http://www.informatik.uni-trier.de/~ley/db/conf/sigir/sigir2005.html
SIGIR 2005
http://doi.acm.org/10.1145/1076034.1076124	69	Bootstrapping Dictionaries for Cross-Language Information Retrieval	acquisition across algorithm algorithms americas amia annotation annual answering applied approach approaches arist artificial association automated automatic based bethesda bilingual biomedical buckley buitelaar chan chen cheng chien chugur collection combination comparable complex computational concept conceptual conf conference corpora coupling cross crossing daumke dejean development dictionary diekema different document domain donohoe dorow editor eichmann english enrich evaluation extraction fall fernandes findings from fung gaussier german gonzalo group hahn headings hedlund hersh hickam honeck identification indexing informatics information intelligence interactive interest interlingua interlingual international intl jarvelin journal katz keskustalo knight koehn language languages large learning leone lexical lexicon library lingual linguistics machine marko marton medford media medical medicine meeting metathesaurus methods model monolingual morphology multilingual naacl national natural nohama nonparallel oard pages parallel passage pirkola poprat porter problems proceedings processing program quantitative queries question raileanu rapp references research resource resources retrieval review riao ripplinger rogati ruiz sacaleanu sadat schulz science selection semantic sheridan sigir siglex special specific srinivasan statistical stripping subject suffix symposium system technology tellex teng test text texts thesauri today tool translating translation translations trec unified unknown unrelated unsupervised using verdejo view vines vintar volk wang wermter widdows williams with word words workshop yang zhang
http://doi.acm.org/10.1145/1076034.1076101	50	Generic Soft Pattern Models for Definitional Question Answering	aaai abductive about acceptable according accordingly algorithm also alterations amsterdam answer answering answers appendix applications approach automatic based because bensley besides biol biology blair boston bowden brown cambridge canada cannot cases choose chua clark combining comparative complete computational contain containing covered create data definition definitional definitions dempster different each editors edmonton embedded essential estimation evaluate evaluation extraction factoid five following form foundations fragments from generating gold goldensohn gram groups guideline guidelines harabagiu haussler hazen hidden hildebrandt holland hovy hybrid incomplete information interpolated jelinek journal july katz knowledge krogh laird language learning likelihood list lists machine make manning markov matched maximum mckeown mercer mian mining modeling models moldovan most multiple muslea naacl natural news nugget nuggets occurrence often online only original other parameters part pattern patterns perform philadelphia phrases practice present press proc processing protein purpose question questions ravichandran reasoning recognition references related required retrieval rouge royal rubin same schlaikjer schtze search sentence sentences sheffield sigir sjolander society soft source sources sparse standard statistical statistics study summaries surface survey system systems target tasks techniques text that these this track treat trec ungrammatical unsupervised using voorhees williams with workshop xiao york
http://doi.acm.org/10.1145/1076034.1076144	87	Expectation of F-measures: Tractable Exact Computation and Some Empirical Observations of its Properties	automomous bayesian butterworths categorization chai chieu classification classifiers ergen evaluating filtering information lewis london online optimizing proc references retrieval rijsb sigir strategies study systems text thresholding yang
http://doi.acm.org/10.1145/1076034.1076103	52	Question Answering Passage Retrieval Using Dependency Relations	answering attardi berger brown cisternino computational curin della estimation final formica information jahr knight lafferty linguistics machine mathematics melamed mercer onaizan parameter pietra piqasso pisa proc purdy question references report retrieval sigir simi smith statistical summer system tommasi translation trec workshop yarowsky
http://doi.acm.org/10.1145/1076034.1076178	121	Dirichlet PageRank	analysis anatomy applied authoritative brin computer dels engine environment erlinked ertextual information kleinb lafferty language large level link metho networks othing page pages references retrieval scale search sigir sources study zhai
http://doi.acm.org/10.1145/1076034.1076068	24	Exploiting the Hierarchical Structure for Link Analysis*	aaai addison aggregation algorithms analyzing anatomy annual applying approximation artificial asano associated automatic baeza between bharat boston brin broder castillo chakrabarti chang compilation conference data davison development distillation dynamics efficient engine engines enhanced environment from gibson graph henzinger hyperlink hyperlinked hyperlinks hypertextual icdm ieee impact improved information intelligence international jean joshi july kleinberg large lempel linkage links lisbon list lncs longman markup mining modern nepotistic neto page pagerank pages popularity portugal press proc proceedings publishing quality raghavan rajagopalan recognizing references research resource retrieval ribeiro ruhl saint scale search september sigir site sites spire springer structure tags tawde text topic using wesley whom wide wise world yates
http://doi.acm.org/10.1145/1076034.1076063	20	Accurately Interpreting Clickthrough Data as Implicit Feedback	analysis applications approach based behavior belkin best breadth carpenter chains choice clickthrough colors comparing comprehension computer conference consumer data databases depth design designed determinants development discovery display dumais editor effects engines ester etra european evaluating experiences explicit facilitate feedback feusner filtering first fixation fixations from fuhr functions gazetracker goldberg granka guide halverson hembrooke holland hornof human implications implicit improve inferred information interaction interactive interests international jaana jameson joachims jose journal just karnawat kaski kelly kemp kieling kloeckner know lankford learning leclerc ledge letin lewenstein link lists long maps match measures mining morita movement movements mydland newman nondurables novel optimizing optimum organizing page pages personalized pkdd polynomial practice preference preferences presented principle principles probability proceedings processes processing psychological query radlinski ramamohanarao rank ranking rayner reading relevance report research result retrieval review russo ruthven salogarvi scott search self shinoda sigir sigkdd software spencer stimson study symposium systems talk task tasks techniques term text theory time track tracking transactions trec understanding user using viewing white wichansky wirschum workshop york
http://doi.acm.org/10.1145/1076034.1076060	18	An Application of Text Categorization Methods to Gene Ontology Annotation	abbreviation abstracts accomplishments acids again alexandre algorithm american amos annotation annual anton applications applied approach apweiler ariel association automated bairoch based beth bhuptiraju biocomputing bioinform bioinformatics biology biomedical biomedicine brief burr categorization cathy centered challenges claire cohen comparative computational computing conditional conference craven daniel daraselia data dayanik debole definitions development dictionary digital dimacs document documents donna donovan egorov elisabeth entity examination experiments exploiting fabrizio feature feldman fields first fluck forum fradkin franca fujita future game gasteiger gattiker gene genkin genomic genomics gerard hagit hanisch hearst heinz hersh high hill hirschman hypotheses ichi identification identifying ieee informatics information informative international introduction jesus joint jong journal juliane julie kantor knowledge kraemer language learning length lewis libraries limsoon linguistics literature locuslink lovins lynette machine madigan maglott maria mark marti martin mcgill mcgraw mechanical medical medline menkov methods mevissen michael mining modern name named names natural ncbi nikolai nlpba nucleic ontology overview pacific page pages park patolis pedersen plans playing practical proceedings processing prot protein proteins pruitt quality ralf random recognition references refseq report research resource resources results retrieval revisiting rich rolf ronen ross rules salton schwartz scientific sebastiani selection sergei sets settles shatkay sigir simple stemming study sumio supervised swiss symposium syntactic term terms text theodor track translation trec trembl tsujii using volume weighting william wong workshop yang year yiming yuryev zimmer zone
http://doi.acm.org/10.1145/1076034.1076169	112	Translating Pieces of Words	accurate achieve addressed align aligning alphabet also amounts annual approach association based benefits between bilingual braschler brown calculated char character chinese church clear clef closely common computational concert conference contrasts corpus cross della derive derived development dictionary distinguished each effective efficacy estimation european eurospider evaluation even example experiments extrinsically first fjoch forum french from future fuzzy giza gram grams have html http information international into keskustalo language languages large level lingual linguistics machine many mapping mappings mathematics mayfield mcnamee memory mercer method might model models need needed obviating orthography papers parallel parameter perform performed pietra pirkola porter present proceedings program references related requires research retrieval revised rules rvelin same second segmentation sets several should side sigir simultaneously small snowball spelling statistical statistically study such system tartarus term terms test texts that thereby thing third this though time toivonen tokenization training translate translation translations types uble used uses using validated variants visala vocabulary ways which while with without word words work workshop
http://doi.acm.org/10.1145/1076034.1076142	85	Testing Algorithms is like Testing Students	algina annual brennan buckley chris classical conference crocker development effect effectiveness ellen error experiment generalizability holt information international introduction james judgments linda measurement modern noreen primer proceedings publications references relevance research retrieval richard rinehart robert sage shavelson sigir size springerverlag test theory topic variations voorhees webb winston
http://doi.acm.org/10.1145/1076034.1076115	62	A Markov Random Field Model for Term Dependencies	advances allan also amati amherst applied approach approaches automatic average avgp bahadur based basis biterm boosting buckley capturing cikm clickthrough climbing combining comparison computer conditional conf conference cooccurrence croft data denote dependence dependencies development direct discovery discriminative divergence document documentation each eighth endence engines entropy ercentage european evaluation examination expansion fagan fields from fuhr full general generalized generative given greiff guestrin henderson hill improvement indep indexing indicate indicates indri inference information international intl joachims journal know koller labeling lafferty language lavrenko lazarsfeld learning ledge lewis losee machine management margin markov massachusetts maximization maximum mccallum mean measuring methods metrics metzler mining mishne mitre model modeling models morgan nallapati network networks neural nips operations optimal optimizing over pages paired parameter parenthesis pereira phrase phrases ponte precision principle probabilistic probability proc processing queries query random randomness rank ranking references relevance report research retrieval rijke rijsbergen robert robertson salton search segmenting sentence sequence settings sigir sigkdd significance significant smoothing song srihari srikanth statistical strohman structured study suggested symb syntactic systems table tailed taskar technical tenth terabyte term test text theoretical theory thesis through track transactions trec trees truncating turtle university using values variant variants weighting where with zhai
http://doi.acm.org/10.1145/1076034.1076079	33	Title Extraction from Bodies of HTML Documents and its Application to Web Page Retrieval	aaai agents american amitay analysis applied association automatic breuel callan carmel chapter chidlovskii classification collins columbia computational computing conference construction craswell craven crescenzi cues cutler darlow data databases description detection distillation document documents domains ecml eikvil eleventh european evans extraction filtering findiing freitag from generation grammar hawking hmms html human improve induction informal information informing international internet journal klavans knowledge language large learning lempel linguistics machine matching mccallum mckeown mecca meng merialdo multilingual naacl named news newsblaster nists north novelty ogilvie overview page pages proceedings ragetli references report retrieval rijke roadrunner science second seventh shih shrinkage sites soffer structural structure summarization survey symposium systems tags technical technologies technology text thompson topic towards track trec twelfth twenty usenix using very vldb volume wide with workshop world wrapper wrappingoriented zhang
http://doi.acm.org/10.1145/1076034.1076093	44	Web-based Acquisition of Japanese Katakana Variants	anchor annual approximate asian association automatic based chien coling computational computing conclusions conference construction corpus could cross data detecting detection determining disambiguation document dowling each easily edit eration erson ervised extraction from future generation hall http identifying ieice increasing information international ipsj japan japanese journal katakana kawai keep language large learning linguistics list masuyama matching mechanically meeting method methods metrics microsoft mining nakagawa name normalization notations ohtake orted orthographic osed ours pages presented previous proceedings processing prop queries references retrieval sekiguchi sekine shirai shishib shoda similarity srihari string sugiyama supp surveys system tamagawa text transaction transactions translation transliterated tsuda unlike using variant variants weakly weight with work yahoo yamamoto
http://doi.acm.org/10.1145/1076034.1076170	113	Cross-Language Text Classification	applied automatic catalog charles cmejrek computational consortium czech data dependency disambiguation engineering english extraction from inflection karolinum language letin lexicon linguistic linguistics mathematical minnen morphological morphology natural parallel philadelphia prague press processing references rich texts translation treebank university
http://doi.acm.org/10.1145/1076034.1076135	78	Probabilistic Hyperspace Analogue to Language	achieved annual applied approach approaches available barwise baseline bayes best bruza cambridge cognitive collection combination combining compare comparison computer concept concepts conceptual configurations constructed context could dependencies details different dimensional direct directional discourse discovering discussion distributed document documents effective empirical expansion experiment experimental explorations five flow follow from full gardenfors geometry high however http indexed information initial journal lafferty language limit limitations livesay logic lowe lund mcdonald meeting memory methods model modeling modelling models number obtained original orleans outperformed pages paisley parameter perform performance performed phal porter press priming probabilistic proceedings processes provided query querying rank refer references removed report reports research restricted results retrieval route schemes science selected seligman semantic sentences seventeen shall shown sigir significance significantly similar since size smoothed smoothing society song space spaces standard stemming stop street study submitted suggested systems table technical terms test then theoretical this thought titles topics tracts trec uniform upper used using values various wall weighting were where whether wilcoxon window with within words zhai
http://doi.acm.org/10.1145/1076034.1076125	70	A Maximum Coherence Model for Dictionary-based Cross-language Information Retrieval	academic across adriani ambiguities ambiguity amsterdam annual approach approaches asian association ballesteros based bertoldi best bilingual burges chen choquette classification clef clir comput computational computers computing conference croft cross daelemans data davis decaying dependence development dictionary diego different disambiguation discovery editor editors eigth embedding engine entropy evaluation expansion experiments federico fifth filtering forum from gill grefenstette hall harman hiemstra http huang hull ijcai improving information institute international iral jang know kraaij lafferty language languages lavrenko learning ledge lingual linguist linguistics lisbon machine machines maeda mathworks maximum mccallum meeting mining model models multilingual mutual myung national nigam nist nmsu notes number occurrence optimization pages papers park pattern phrasal pohlmann practical prentice press proceedings publication query querying recognition references relations relevance research resolve resolving retr retrieval revised rodopi sadat salton search second sense september sigir sima simard similarity smart special springer standards statistical studies support syntactic system systems techniques technology term text track translation translations trec tutorial twenty uemura using vector veenstra verlag volume voorhees weighting weischedel with working workshop wright yoshikawa zavrel zhang zhou
http://doi.acm.org/10.1145/1076034.1076074	29	Optimization Strategies for Complex Queries	activity addison advanced annual applications approaches author bell broder brown buckley callan carmel center combining comments compressing computer conclusions conference croft database development documents edition efficient ening evaluation expert expressed fast files findings flood gigabytes grant harding herscovici images indexing inference information inquery inteldexa international invaluable inverted jamie john know knuth language ledge level lewit ligent management managing massachusetts material mendations metzler model moffat necessarily network opinions optimization optimizations pages paper part press proceedings process processing programming queries query reading reand recomart references reflect research retrieval search searches searching self september sigir soffer sons sorting sponsor strategies strength structured supported syst system systems text this those trans turtle twelfth using vector volume were wesley whose wiley witten work zien zobel
http://doi.acm.org/10.1145/1076034.1076175	118	Scalable Hierarchical Topic Detection	allan annual approach architecture based bolivar browsing burnett cluster clustering collection collections committees computing conference crimmins cument cutting data definition detection development document dynamic efficient evaluation event feng flexible flynn gather hierarchical http index information international intrinsic irsg jain karger know kraaij language large ledge linking loquium lrec management modeling multilingual murty news newspaper nist pages pantel pedersen plan press proceedings quinn references report research retrieval review scatter sigir smeaton speech spitters streams surveys task tests texts thesis topic tracking trieschnigg tukey twelfth twente university unsupervised variations with workshop
http://doi.acm.org/10.1145/1076034.1076138	81	Automatic Web Query Classification Using Labeled and Unlabeled Training Data	according approach based categorizing cikm class classification document foundations geographical gravano hatzivassiloglou information kang language lexical lichtenstein locality manning natural pennsylvania press processing queries query references relationships resnik retrieval schutze selection sigir statitical thesis type university
http://doi.acm.org/10.1145/1076034.1076127	71	Hidden Markov Models for Automatic Annotation and Content-Based Retrieval of Images and Video	amir annotated annual barnard blei conference data duygulu forsyth freitas international jordan journal learning machine matching modeling november pages pictures proc references research retrieval sigir system trecvid video words
http://doi.acm.org/10.1145/1076034.1076158	101	3D Viewpoint-based Photo Search and Information Browsing	accurate archaeological architectural archives byzantine calibration camera computer conference database digital efficient furukawa ieee image integrated isprs july kadobayashi kanjo kawai machine models pages pattern presentation proceedings recognition references ruins system technique techniques tsai vision workshop yoshimoto
http://doi.acm.org/10.1145/1076034.1076147	90	Top Subset Retrieval on Large Collections Using Sorted Indices	achieved algorithm american average best clear conference consistency consistent cost davis degradation document documents early effective engine european evaluated expense ferguson filtered formats frequency from gain gurrin however improvement increase index indexes indices information interesting kretser level maintain managed march mean moffat more nearest neighbour note other over performed performs perhaps persin poorly precision problem proc proceeding ranking references regular retrieval rijsbergen sacks science search seen sigir smeaton society sorted space storage sweightntfsorted terabyte termination terms tfsorted that there these this upperbounds using vector weightedntf weightntfsorted weighttfsorted which while wilkins with zobel
http://doi.acm.org/10.1145/1076034.1076091	43	Learning to Extract Information from Semi-structured Text using a Discriminative Context Free Grammar	aaai addresses algorithm algorithms andrew approach artificial august australia automatically based bayesian boosting borkar bouckaert cardie caruana claire classification collins committe computer conditional conf conference constrained corinna cortes culotta data david deshmukh discriminative emnlp empirical engineering environment experiments extracting extraction fernando fields francisco free freund from hidden high hodor ieee information intel interactive international john kaufmann kaustubh kristjansson labeling lafferty language large learning letin level ligence machine margin markov mccallum methods mining models morgan natural network networks pages paul perceptron pereira pierce precision probabilistic proc proceedings processing proposal random references remco report rich rosenberg sarawagi schapire segmenting sequence society structure sunita support sydney technical text textml theory training using vapnik vector vina viola vladimir with workshop
http://doi.acm.org/10.1145/1076034.1076161	104	Major Topic Detection and Its Application to Opinion Summarization	based classification conference dave dependency domain event extraction fukumoto fukushima gallery harman institute lawrence mining morinaga national novelty opinion overview peanut pennock product references reputations retrieval reviews semantic sigir sigkdd soboroff standards suzuki tateishi technology text track tracking trec twelfth yamanishi
http://doi.acm.org/10.1145/1076034.1076044	5	A Study of Factors Affecting the Utility of Implicit Relevance Feedback	academic algorithms american appendix assessments bell borlund brajnik busha campbell checkboxes complexity components conceptions conference context data developing document documentation european evaluation experimental explicit feedback hall harman harter help implicit information interactive interface interfaces international interpretation journal librarianship library mark methods mizzaro model modification needs ostensive other prentice press proceedings query ranking references relevance relevant research retrieval rijsbergen ruthven science searchers searching sentence series society strategic structures summary system systems task tasso techniques technology title titles user venuti york
http://doi.acm.org/10.1145/1076034.1076156	99	Finding Semantically Similar Questions Based on Their Answers	annual answer applications approach approaches artificial based berger bridging burke caruana case chasm cohn conference croft finder finding freitag hammond http information intel international knowledge language lemur lexical ligence lytinen martin mittal modeling naver navigation pages ponte proceedings references retrieval sigir statistical
http://doi.acm.org/10.1145/1076034.1076066	22	Detecting Phrase-Level Duplication on the World Wide Web	aaai american amitay analysis applications artificial between bharat broder capocelli carmel center challenges chang clustering clusters communications computer computing conference congress connectivity damn darlow data databases davison detecting duplicate editors engines evolution fetterly fingerprinting forum functionality glassman graph harvard henzinger hypermedia hypertext ieee intelligence international july june kumar large latin lempel linkage links locate maghoul manasse method methods mining motwani najork near nepotistic pages patterns polynomials rabin raghavan rajagopalan random recognizing references report research ruhl santis scale science search security sequences sigir silverstein site sites soffer some sonar spam springer stata statistical statistics structural structure study syntactic technology tomkins university using vaccaro verlag whom wide wiener workshop world zweig
http://doi.acm.org/10.1145/1076034.1076110	58	Modeling Task-Genre Relationships for IR in the Workplace	academy accepted allan american analysis annual approach athens australasian background based behaviour belkin broder brooks bruce bystrom challenges ciccolo communication conference database documentation dunedin emerging engineers enterprise environment extensions fidel forum framework frameworks freund from generation genres greece greenwood hansen hawking human information informing ingwersen jarvelin journal language libraries life management media meeting methods modeling moving needs next oddy office organizational orlikowski part patent presented proceedings process real references registration research retrieval review science search seeking sigir society software structurational studies studying swedish systems task tasks technology theory toms towards units unlimited using vakkari village waterhouse wilson work workshop yates zealand
http://doi.acm.org/10.1145/1076034.1076182	125	Mining Translations of OOV Terms from the Web through Cross-lingual Query Expansion	accuracies anchor automatic based chen cheng chien compared comput corp corpus cost cross detection different each entity equivalence extraction feature features figures huang information japan july language linguist listed livetrans minimization mining mixed multi multilingual named overall pages parallel press prior proceeding provided quality queries query recognition reference references resnik retrieval sapp sigir smith system systran systransoft table talip teng terms text time translating translation translations translingual unknown using vines vogel waib wang with workshop zhang
http://doi.acm.org/10.1145/1076034.1076070	25	Web-Page Summarization Using Clickthrough Data	abstracts advances algebra algorithm amitay analysis annual approach around asian asker association automatic automatically based bergen berry bostrom bouchon brien browsing buyukkokten cambridge carbonell categories categorization chapter chen chien chuang classification clustering computation computational conference creation cubesvd decision delort development devices document documents domain dumais eleventh engine enhanced enriching european evaluation extraction firmin form from garcia generate generic goldstein gong gram handheld hierarchical hirschman house hovy huang hulth hyperlinks hypermedia hypertext information intel intelligent international ject jonsson journal kantrowitz karlgren keyword klein know knowledge krishnapuram kummamuru language latent ledge ligent linear linguistics literature logs lotlikar luhn management mani mapping maybury measure meng metadata metrics meunier mittal molina monothetic norway novel occurrence paepcke page pages paris parts personalized press proc proceedings processing queries query references relevance research results retrieval rifqi search seeing selection semantic sentence shen siam sigir singal sites statistics summac summaries summarising summarization summarizing sundheim support systems taxonomies technology tenth terms text texts thematic there through tipster transactions user using whole wide world yang york zeng zhang
http://doi.acm.org/10.1145/1076034.1076089	41	Controlling Overlap in Content-Oriented XML Retrieval	aalto advances algebra annual april arvola assessments automatic avignon based bayesian beaulieu bruno callan canada carmel clarke comp comprehensive computer conference configurable consens consistent content cumulated dagstuhl data decemb dehaan development diego documents dynamic editors effort encoding eriments ertson evaluation exhaustive extension fields filtering fragments france fuhr gain gaithersburg gallinari germany hierarchical holistic hybrid indexing inen inex information initiative interactive international interval irrelevance johann joins july june junkkari kamps kazai know koudas lalmas language lecture ledge length lncs maarek madison malik management mandelbrod mass matching metrics models most multiple multitext networks nexi next normalization notes novemb ogilvie okapi onent onents optimal oriented orleans overlap ozsu pages pattern pehcevski piwowarski predefined problem proceedings providing published queries query ranking references refinement relevance relevant reliability research retrieval retrieving riao rijke rnsson science searching second septemb seventh sheffield sigir sigmod sigurb simple soffer springer srivastava structured struggling systems taylor techniques tests text third thom tilker tolerance toman toronto track transactions translation trec trix trotman twig unit user using velin vercoustre visited vittaut volume vries walker washington weighted wisconsin with without workshop xirql xquery zaragoza zolt
http://doi.acm.org/10.1145/1076034.1076189	131	A Wireless Natural Language Search Engine	architecture atio august author brazil copyright earch engine figure held ireless owner references salvador search sigir tern wireless
http://doi.acm.org/10.1145/1076034.1076155	98	Predicting Query Difficulty on the Web by Learning Visual Clues	cronen determining effectiveness evaluate framework inferring jensen necessary ounis performance predicting predictors preretrieval proceedings query references search sigir sizes spire townsend using
http://doi.acm.org/10.1145/1076034.1076149	92	The Impact of Evaluation on Multilingual Text Retrieval	access achievements analysis braschler clef cross evaluation experiments ferro forthcoming forum html http information isti language lncs mart mayfield mcnamee merging multilingual nunzio objectives overview peters publications references results retrieval scalable strategies track website with
http://doi.acm.org/10.1145/1076034.1076076	31	Efficiently Decodable and Searchable Natural Language Adaptive Compression	academic adaptive adding addison addressing algorithm algorithms alistair allowing arith arithmetic aspects australian baeza based behavior bell block boyer brisab burrows cambridge carpinelli character cleary code coder codes coding comm comp compressed compressing compression computational computer conf construction coruna data databases dense dept documents ecir efficient effort eiro errors esteller farina fast file flexible generation gigabytes hall heaps horsp huffman human ieee iglesias images indexes indexing individual information inst integer inverted kluwer language least lemp lncs lossless managing matching method minimum modern moffat moore morgan moura natural navarro neal neto neub next optimized over param parama pattern pract practical prentice press principle proc radio raffinot rate redundancy references retrieval salamonsen search searching sequences sequential sigir simple soft sorting spain spire string strings stuiver systems tarhio text theoretical thesis tois turpin univ universal university using variable wesley wheeler witten word yates zipf ziviani
http://doi.acm.org/10.1145/1076034.1076109	57	When Will Information Retrieval Be "Good Enough"?	aaai abdul abstract accompanying accuracy addison aggregation ahrenberg allan alon amherst analysis annual answering apache automatic available based bear behavior belkin blair boyce buckley callan case center chen cheyer classification college communications computer conceptual conference croft current dahlback data development dialog dialogue diaz displaying document documents drori editor editors effectiveness engines evaluation evidence exploration facilitating fail found foundation from full fuller generalization gorin graphical graphics haddadin hard harman hawaii hebrew hierarchies high hobbs http human improving incomplete information intel intelligent interaction interactive interface interfaces international israel issues jakarta jaleel jerusalem jonsson judgment julia kaszkiel kehler koenemann larkey leibniz level ligent list lotem lucene maron martin maryland massachusetts metzler model multi multimodal natural network nist notebook novelty online oriented over overview page pages paper park passage performance plaisant preference previews proceedings processing publication query question reference references relevance report representations representing research results retrieval revisited salience salton science sciences search searcher shneiderman sigir sixth slaughter smucker soboroff software special spoken strohman studies study swan symposium system systems tanin technical text thirteenth track transformation trec turtle twelfth umass university user using voorhees wade weiland wesley wilkinson with wizard zobel
http://doi.acm.org/10.1145/1076034.1076164	107	Analysis of Recursive Feature Elimination Methods	analysis applications approximation cancer categorization change classification classifiers commonly deriva dified during feature features function gene guyon icml jian large learning logistic loss machine machines metho minimize obviously partial penalized plicity process ranks recursive redundant references regression ridge rocchio scale selection simy style supp text their used using vector weight will yang
http://doi.acm.org/10.1145/1076034.1076187	129	A CLIR Interface to a Web Search Engine	aspects august author based brazil copyright cross document from hahn hedlund held indexing informatics information international jarvelin journal language lingual management medical mono morpheme morphology owner perspective pirkola processing references retrieval salvador schulz semantics sigir swedish
http://doi.acm.org/10.1145/1076034.1076128	72	Exploiting Ontologies for Automatic Image Annotation	active algorithm analysis annotated annotation answering approach associated automatic barnard based bernoulli better between blei brown caption chai chains cikm ciocca classes classification coherent coling computational computer conference cross cusano cuts cvpr data della dempster distances dividing document duygulu effective electronic estimation european experiments features feng first fixed forsyth freitas from hierarchy hoffmann http icml ieee image images imaging improving incomplete indexing intel international internet ject jeon jordan journal kang laird language lavrenko learning lexical lexicon ligence ligent likelihood linguistic linguistics machine madison malik management manmatha mathematics maximum mccallum media memo mercer mitchell model modeling models moldovan mori multimedia multiple nips nist nlpir normalized novischi occurrence pages parameter pattern pictures pietra proceedings projects puzicha quantizing question quigley recognition references regularizing relevance retrieval rosenfeld royal rubin schettini segmentation semantic semantics series shrinkage sigir smeaton society statistical statistics storage takahashi text trans transformation translation trec trecvid using vector video vision vocabulary wang with word words workshop
http://doi.acm.org/10.1145/1076034.1076163	106	Profile-based Event Tracking	academic allan available based boston cikm clustering definition detection evalplan evaluation event flexible hierarchical http information intrinsic james kluwer nist organization orleans plan proceedings publishers references speech task tests topic tracking
http://doi.acm.org/10.1145/1076034.1076132	75	A Testbed for People Searching Strategies in the WWW	allan center computer conference coreference corpus cross department disambiguating document garg gooi guha information intelligent large massachusetts people press proceedings references report retrieval scale science search technical univ wide world
http://doi.acm.org/10.1145/1076034.1076072	27	Do Summaries Help? A Task-Based Evaluation of Multi-Document Summarization	aaai abstracts alberta amigo analysis analyst argument artificial assistant automated automatic barcelona barzilay based basis berlin between blair bloedorn bodnar brandow case center clustering colbath college columbia condensation conference consensus construction content crescent daily daume demo developed diego distribution document domain eacl echihabi edmunton electronic elhadad empirical evaluating evaluation evans evidence examining exclusive experiments extracts factoid fifteenth forth from galliers generator gleans goldensohn gonzalo gram graph halteren hand hatzivassiloglou hovy hughes human independent information initial intel intelligence island january jing joint jones klavans kubala language lecture ligence ligent logical madrid management mani manuscript marcu marshaling matching mckeown method methods military mitze multi munteanu naacl national natural nenkova news newsblaster newsinessence nice notes occurance pages passonneau peinado penas philadelphia proceeding proceedings process processing proposal prototype providence publications pyramid radev raghavan real references research rethinking review rhode sable schiffman schum search second selection sentence session sigelman sign single soricut spain sparck springer statistics strategic study sumamries summaries summarization summarizing sundara symposium synthesis system systems task tasks technology teufel text third time tracking understanding using verdejo warning washington with workshop zhang
http://doi.acm.org/10.1145/1076034.1076040	2	Why Spectral Retrieval Works	able according acid actually affect again algorithm algorithms also american analysis ando angle appear approach assumed asymmetric asymmetry automatic azar balance based basis been boost both buckley call carolina categorization chapel characteristic clustering collection column communications computational computations computed concepts conclusions consider corresponding could curve curves data decomposition deerwester definition detecting different dimensionality dimensionless dimensions ding direction discovery document documents dumais dupret dwarkadas each edition effectiveness effects efron eigenvalue entries estimators evaluation evidence example expand expansion expect experimental external factors fashion fiat figure figures findings first form foundation foundations frequent from functionality furnas further generalization give given gives golub good hard harshman have heart hence hersh hickam hill hofmann hopkins husbands icdm identical identify identifying ijcai illustrated immediately improves including incorporate indexing indices individual information inter interactive into introduced intuitive iterative johns jordan journal just kamvar karlin klein knowledge kontostathis landauer large latent lead learning left lemma leone lewis like list loan looking machine main makes manning matrix mcsherry measurement method methods mining model more much needed nips normalized north note nothing nucleic number observation obtain ohsumed only optimal orthogonal orthogonality other outlook outperform pages pairs pairwise papadimitriou patterns peer perfectly phenomenon pods possible possibly pottenger practice precision present previous principled probabilistic probability problems proceedings process prompt proportions proposed query raghavan ranked rationale reduction references related relatedness relations rescaling research residual retrieval reuters riao rise saia salton same scaling scenario schemes science scores semantic sequence shape sigir similarity simon singular smoothness society space special spectral steps stoc straightforward strong such symmetric synonyms systems taken tamaki tang term terms test text that then theoretical there therefore thesaurus these thesis they third this underlying unsupervised used user value variable vector vectors vempala versa version vice view viewed weiss what when whenever where which widely will with wong word work workshop would yang
http://doi.acm.org/10.1145/1076034.1076046	7	User Term Feedback in Interactive Text-based Image Retrieval	american analysis andrews anick annotated april assistant based belgium belkin berkeley blei brussels categories choi clough collection computing conference content context cool curcuits data determining development eurovision expansion factors features feedback from full functionality guide harman hearst henniger history http human ieee image imageclef images information intelligent interaction interactive interface international issues iterative jordan journal king koenemann learning marchetti modeling multimedia nicholas norman novel paraphrase park photographic press proc proceeding proceedings provide queries query ranking rasmussen references reid relevance research results retrieval riao sanderson science scott search searching seeking semantic shef sigir society space strategies system systems technique technology terminological text tipirneni towards transaction trec user using video york zhang
http://doi.acm.org/10.1145/1076034.1076112	60	The Loquacious User: A Document-Independent Source of Terms for Query Expansion	accuracy acquisition actual adding allan analysis analyzed anick annual based beaulieu behavior belkin brussels buckland buckley callan canada carballo case cognitive communications computing conference cool croft cronen current deng design development dinstl document documentation documents dublin edition effect effectiveness efthimiadis elements engines environment evaluation examining expansion experiments exploration factors fail feedback find finland franzen from gatford government grenoble grunfeld hancock hard harman helping high hill html http human impact information ingwersen interaction interactive interface interfaces international introduction irinterface iterative jansen jarvelin jones journal kalgren katz kekalainen kelly know koenemann kwok length library life ling magennis management mcgraw melbourne multiple muresan needs office okapi overview park people perez performance perspectives philadelphia pircs pittsburgh point potential predicting printing procedures proceedings processes processing queries query real reference references refinement reformulation relevance representations research retrieval retrieved review rijsbergen robertson robust ruthven salton saracevic science search services sheffield sigchi sigir sikora spink staff structure study support system systems tampere tang technology terminological text theory they third thirteenth toronto towards townsend track trec twelfth user users using verbosity view volume voorhees walker washington what with work yuan zhou
http://doi.acm.org/10.1145/1076034.1076049	9	Improving Collection Selection with Overlap Awareness in P2P Search Engines	aberer academic access across acuna adaptive addressable advances allowable applications approach architecture architectures available balakrishnan based bibfinder bloom buchmann build building byers callan chakrabarti chord cikm cluster coding collection collections combining commun communications communities compressed comput computer computing conference considine content continuous coverage croft cuenca data database decentralized decision delis delivery detection development different discovering discovery distributed druschel effectively efficient engine errors estimating evaluating experiments expressions fagin filtering filters florescu framework francis francisco from fuhr full fuzzy galanis galanx ganguly garcia garofalakis gloss gossiping grabs gravano handley hash hauswirth hernandez http hypertext ieee improving index inference information informed integration international internet ject journal kaashoek kambhampati karger karp kaufmann kharrazi kluwer know koller language large ledge levy location long lookup martin mathur melnik merging methods middleware mining minka mitzenmacher modeling models molina morgan morris multiple netw network networked networks nguyen nist nottelmann novelty odissea offs ogilvie over overlap overlay pages paral pastry peer peery planetp polytechnic poster powerdb press probabilistic proceedings processing protocols publishers punceva quality raghavan rakaposhi rastogi ratnasamy recommended redundancy references report research resource results retrieval rost routing rowstron rutgers santa scalable scale schek schenker schmidt search searching selection sept service shanmugasunderam sharing sigcomm sigir sigmod source space statistics statminer stoica streams structures suel symposium syst system systems technical technologies text theoretic thomas time tomasic trade trans transactions trec univ university update using version vldb wang wisc with witt yang yuanwang zhang
http://doi.acm.org/10.1145/1076034.1076134	77	A Geometric Interpretation of R-precision and Its Correlation with Average Precision	analysis annual blustein buckley conference data development evaluating evaluation harman information international measure overview pages press proceedings references research retrieval seventh sigir stability statistical sutcliffe tague text third trec voorhees
http://doi.acm.org/10.1145/1076034.1076117	64	Gravitation-Based Model for Information Retrieval	amati based buckley communications computer divergence document editors fang formal from fuhr heuristics information jones journal kaufmann length measuring mitra model models morgan normalization pivoted probabilistic proceedings randomness readings references retrieval rijsbergen salton sigir singhal space sparck study systems transactions vector willett wong yang zhai
http://doi.acm.org/10.1145/1076034.1076166	109	Revisiting the Effect of Topic Set Size on Retrieval Error	annals annual approximate approximated approximation between buckley clearly compare conference connection development effect empirical error evaluating evaluation exactly experiment exponential follows form found function goal here html http indeed information integral international mathematical measure model nist note pages past press probability proceedings proposed reer references replacing research results retrieval same september show sigir size stability statistics that theoretic theoretical therefore they topic trec voorhees williams with
http://doi.acm.org/10.1145/1076034.1076086	39	Integrating Word Relationships into Language Models	algorithm berger chen conference data dempster dependence development empirical from goodman harvard hauptmann incomplete information journal lafferty laird language likelihood maximum model modeling pages proceedings references research retrieval royal rubin sigir smoothing society statistical study tech techniques title translation university zhai
http://doi.acm.org/10.1145/1076034.1076039	1	Orthogonal Locality Preserving Indexing	above accu accura accuracy advances after akademiai always american analysis ando average bartell based baseline belew belkin best both bousquet budapest case chicago chung classes classification clustering computations computer conference confidence cottrell database decreases deerwester department different dimension dimensionality ding document does drastically duda dumais dwarkadas edition eigenmaps eline embedding estimation evaluated factorization figure filtering finding fluctuates framework from furnas generalization geometric global golub gong graph harshman hart high hoboken hofmann holland homology hopkins improves indexing information inter interscience intrinsic iterative johns journal kegl kiado kokiopoulou landauer langford laplacian large latent limits linear loan locality locally lovasz luxburg maniold matching mathematics matrix measurement model multidimensional mutu mutual negative neural niyogi nonlinear north number numbers olpi optimal outperform packing papadimitriou pattern peer performance plummer polynomial precision preserving press principles probabilistic probability proc proceedings processing racy raghavan random reach reduction references regional regularization report representation rescaling residual retrieval roweis saad samples saul scaling science semantic series sigir silva similarity sindhwani slightly smale society space special spectral stork submanifolds symp systems tamaki tang technical techniques tenenbaum their theory university using vempala very volume weinberger wiley with
http://doi.acm.org/10.1145/1076034.1076146	89	On Evaluation of Adaptive Topic Tracking Systems	active adaptive allan based bayesian callan connel evaluation exploitation exploration feng filtering icml kumaran learning notes raghavan references shah umass working zhang
http://doi.acm.org/10.1145/1076034.1076151	94	Customizing Information Access According to Domain and Task Knowledge: The OntoExplo System	aimsa approach aussenac author background berlin bhogal broekstra buel bullet cecchini challenges collections commerce corpus crowlesmith dealing document dope electronic evaluate existing explore exploring fensel figure fluit gilles harmelen hernandez indexing intelligent isbn kampman knowledge large management mothe mulligen ontologies project receipt references repositories result riao scerri selection semantic silver solar springer stephano stuckenschmidt system technology upon verlag waard with
http://doi.acm.org/10.1145/1076034.1076172	115	Automated Evaluation of Search Engine Performance via Implicit User Feedback	absence action actions american annual application background been behavior bharat bibliography calculated capture chalmers city clicks conducted conducting content data delos digital distinct dublin dumais each efficiency engine engineers environment errors evaluation excellence favorites feedback forum framework from google have implicit implicitly indicated inferring information initial interesting interests ireland joachims june kelly libraries link major manufacturing measures meeting metrics modeling naturalistic network noted oard object observable observed organization pairs performance personalisation positive preference preferences presence print proceedings quality quite recommender relevance remaining report respectively results running save science search searches second setting sigir society such system systems technology teevan that tracking university unobtrusively user using villa washington weigend were with workshop
http://doi.acm.org/10.1145/1076034.1076191	133	Hierarchical Text Summarization for WAP-Enabled Mobile Devices	amsterdam andreas august author automatic benjamins brazil browsing buyukkokten computational copyright devices dragomir eduard efficient form garcia handheld hector held hovy inderjeet introduction issue john kaljuvee kathleen linguistics mani mckeown molina oliver orkut owner paepcke page radev references salvador sigir special summarization syst terry trans using winograd
http://doi.acm.org/10.1145/1076034.1076084	37	Relevance Information: A Loss of Entropy but a Gain for IDF?	access according activation addition after amati american annual anthology applied approximations arguments automatic available based basis beaulieu because between binary buckley butterworths canada chapter church classical collection component conclusions conference corpora could cument dbms decreases deling delling dels development deviations discriminativeness divergence document documentation documents edition effect effective eighth eing elleke engineering entropy erational ergen eriments ertson estimates estimating estimations event experimentally explicit feedback first form framework france frequencies frequency from function gaithersburg gale general generation glasgow grenoble hanco hiemstra high impact improving include incorporated incorporation independent inference information informative informativeness initially interactive international into inverse investigated issue jones journal july june justifying kazai kluwer know lafferty lalmas language large ledge london long loss makes management maryland matrix maximal measure measures measuring metho mirror model models motivated next novemb occurrence oisson okapi original overall pages paper parsimonious particular poisson positive probabilistic probabilities probability proceedings processing prove queries query randomness reasoning reasons references relating relationship relevance relevant research respectively result retrieval review revised revising rijsb ruthven salton science search second seventeenth sigir simple smaller society some space sparck spreading springer stephen steps subtracted success summary superior survey system systems term terms test text that theoretical thesis third this toronto transactions transformation trec tsikrika understanding unfortunately university values verlag very viewed vries walker weighted weighting which with withindocument wong workshop zaragoza zhai
http://doi.acm.org/10.1145/1076034.1076180	123	Indexing Emails and Email Threads for Retrieval	apache chris cscw derek docs exploiting http improve interactive java lucene mail orleans posters references rohall schmandt stern steven structure summarization
http://doi.acm.org/10.1145/1076034.1076052	12	A Utility Theoretic Approach to Determining Optimal Wait Times in Distributed Information Retrieval	accuracy adaptive advances agarwal agent agents alan among annual antecedents applications approach april aslam autonomous baeza better brand broker bruce brynjolfsson business callan choice clay collection collections colleen commonsense comparison computer conference consumer context cost croft customer data database databases decision designing development distributed document doorenbos econometric economics editor efficient effort ellen engine eric erik etzioni examination finland first focs foundations fred fuhr fusion gaithersburg garcia gathering generalizing gloss glover goaloriented goose graphics gravano gupta hanks harman henry hierarchies hosanagar hypermedia industrial inference information interface international internet issue james jamie jiang johnson journal just karp kehoe kelly krishnan laboratory laguna laird larcker large learning lessig lieberman madani maes making malaga management marina marketing matters mcfadden media merge metasearch michael models modern moenaert moffat molina montague montgomery narendra neto networked networks optimization oren pages paper pattie payne perceived pitkow press probabilistic proc proceedings products psychometric real references regression research results retreival retrieval rhodes ribeiro rogers rudy sampled scalable science sciences search searching seattle selection selker september shopbot shopping shugan sigir simulation smith souder space spain special steven still strategies survey sutton systems tenth text theoretic thinking third time transaction trec usability usefulness user using utility vector very visualization vldb voorhees waarts wedding weld wide william winter with working world yates york zhihong zobel
http://doi.acm.org/10.1145/1076034.1076082	36	Multi-labelled Classification Using Maximum Entropy Method	aaai advances alberta algorithms alternating among analysis annual applications approach argonne banff based becker benson berkley between biometrics biostatistics boostexter boosting cambridge canada carnegie categories categorization category chen chua cikm clare classification comite comparison comparisons computer conference conll constraint correlation correlations corresponding crammer data decision della development discovery discriminative division document dotted elisseeff entropy estimation european examination family features fields figure filtering finland first gaussian gene gilleron godbole hastie hierarchical hofmann http icml ieee ijcai indicates individual inducing information intelligence international jaynes kernel king knowledge label labeled labelled laboratory lafferty lagrangian learning line linear list logistic machine machines mailing malouf management manual mathematics matrix maximum mccallum mcinnes mechanics mellon method methods mfom microarrays mining mixture model models more multi multiclass multipliers national neural nigam obermayer oles online pakdd parameter parameters parametric pattern peled penalized phenotype physical pietra pkdd points press principles prior proc proceedings processing random ranking references regression regularized relation report research retr retrieval reuters review revision robust rosenfeld roth saito sarawagi sarich schapire school science sigir singer sixth smoothing springer statistical strict support system systems tampere technical text their theory thirteenth tommasi trained transactions trees trend triangle twenty ueda university upper user using vector verlag washington weston where which wilcoxon with workshop yang zhang zimak
http://doi.acm.org/10.1145/1076034.1076179	122	A Retrospective Study of Probabilistic Context-Based Retrieval	above american application approach approaches approximations average avoid based baseline best blind bruza chen chung clarke collections compares comparison comparisons conference context contexts croft databases davis dependencies depends determined different document documentation documents each effective effectiveness efficient empirically every expected experiment exploring factoid factor feedback formal forum fuzzy here heuristic increased information initially interpretation into jones journal judgments kaszkiel kretser kupiec kwong language locality location matching mathematical mean mentioned methods model modeling modified moffat next operators order ours parameterized passage pedersen performance poisson ponte precision predictive present presentation probabilistic proc proposed queries question ranking references relevance report results retrieval retrospective robertson rprecision sacks scheme science score search should shows sigir similarity simple simplicity size society some song space sparck specific specificity spurious statistical substituting summarizer summary table term terms terra that through title tois trainable trec used using various walker weight weighted weighting weights when with workshop would zobel
http://doi.acm.org/10.1145/1076034.1076119	65	Impedance Coupling in Content-targeted Advertising	addison additional advanced advert advertisement advertisers advertising adwords african allo americas among annual april associating attempt attitude august average baeza based baseline because bhargava brazilian browsed browser called cancun case categorization cation chan choi cobweb collection commerce communication comparison comput computer conclusions conference consideration consumers containing content contents contenttargeted coupling craswell crawler croft customization daeredita database decisions delivery development digital directly distinct distribution edition editors effective efficient eighth eiro electronic eleventh elief enablement endo engine engines ergen ethical ethicomp evaluated evaluating evaluation evidence expanding feasible feng fifth figures first five form from gains gaithersburg garcia generated golgher google haig hawking high hoffman html http human hyperlink impacts impedance indicate indicated inference information institute intel international internet into investigated investigation isdn issues july kamba kaufmann keywords kohda koseki laender langheinrich large learning ligent longman maryland match matching measurement media merging metrics mexico million modern more morgan muntz nakamura national nearest neighb neto netw network networks nist novak novemb obtained ogle osed other over overview page pages paid pearl penno placement planned plausible point possible practical precision press probabilistic problem proceedings processing provided publishers publishing quality real reasoning recognize references relative research results retrieval rijsb roukos rules sample scientists scoring search select septemb seventh sigir silva smith social solve south spire springer standards strategies strategy string such switzerland symposium syst systems taken targeted techniques technologies technologists technology terms test text that there these they this thistlewaite thousand through time toward track transactions trec troubador turtle ubiquitous understanding unintrusive using veloso verlag very visibility vocabulary wang website weideman were wesley which wide with work world yang yates yielded zhang ziviani zurich
http://doi.acm.org/10.1145/1076034.1076054	13	Robustness of Adaptive Filtering Methods In a Cross-benchmark Evaluation	about accuracy adaptation adaptive after allan amount applied based bayesian been benchmark boosting both callan cambridge categorization challenge cikm classification classifiers combine concluding conclusions consistency corpus cross data decided detection document documents duddington each effective elements estimation evaluation event evidently examples feedback filtering final first fiscus focusing following friedman from goal hastie highly hull improve incremental incrementally information kisiel labeled learning likelihood linear local main make margin maximum methods microsoft model negative obtain oles open optimization organization overview parameter past perform performance positive predicted prediction presented priors process provided randomly references regression regularized relevance remarks report respect results retr robertson robustness rocchio sampled schapire settings sigir significantly singer singhal small soboroff springer statistical statistics steps studied study summary table terms test text their then this thoroughly those thresholds tibshirani topic track tracking train training trec true updated using validation walker wheatley which while will with yang zhang
http://doi.acm.org/10.1145/1076034.1076042	4	The Maximum Entropy Method for Analyzing Retrieval Measures	aaai about actual agree ambiguity annual approach autonomous berger bollmann buckley classification communication comput conditional conference cooper correlation cover critical current dervin development distribution document dudik editors effectiveness elements entropy evaluating evaluation fawcett figure filtering first greiff icml ieee ijcai improvements inferred information international investigation jaynes john journal jung kagolovsky kantor kaufmann kendall lafferty language learning lewis linguist losee machine marcus mathematical maximum mccallum measure measures mechanics method methods mishra mixtures modeling models moehr morgan natural needs nigam nilan optimizing page pages part pavlov pennock performance phillips physical pietra ponte popescul precision press principle probabilistic proc proceedings processing publishers quality raghavan rankings rationale ratnaparkhi readings recall references relative research resolution retrieval review saracevic schapire science selecting shannon sigir sons species springer stability statistical status syst system systems technical technology text theory thomas trans trec twenty ungar using volume voorhees when wiley workshop york
http://doi.acm.org/10.1145/1076034.1076194	136	A Web Mining Research Platform	almaden august author brazil cluster composed computational conf conj content copyright data diglib enhancing exposes held http ieee integrating jose know knowledge large ledge long management methods mining monitoring online owner platform portal proc process publication references repository reservation resource salvador service sigir stanford storage store system term testbed user virtual webbase webfountain with workshop
http://doi.acm.org/10.1145/1076034.1076120	66	Improving Web Search Results Using Affinity Graph	addison annual australia automatic baeza calvo carbonell classification conference content development digital diversitybased document documents goldstein information international journal longman managing melbourne modern neto proceedings producing references reordering reranking research retrieval ribeiro sigir summaries wesley with yates
http://doi.acm.org/10.1145/1076034.1076153	96	Live Visual Relevance Feedback for Query Formulation	algebra american applications berry decompositions document fierro hoenkamp information journal linear numerical operators orthogonal rank references retrieval science society space technology unitary with
http://doi.acm.org/10.1145/1076034.1076129	73	A Database Centric View of Semantic Image Annotation and Retrieval	analysis annotated annotation approach asian asilomar august australia automated automatic ballard barber barnard based bayesian benchmarking bernoulli blei browsing california carbonetto chang classification coding color compression computer computers conf conference content contextual croft cvpr dashed data databases december digital discriminant distance distributions dividing duda duygulu early eccv efficient equitz faloutsos feature features feedback feng figure finite first fixed flickner form forsyth framework freitas fully functions fundamentals general glasman gupta hafner hall hart hier hierarchies high histogram iccv ieee ijcv image images indexing information instance intelligence intelligent international jain jeon jmlr john jordan jose journal juang july june language lavrenko learning level lexicon libraries library line lippman lozano machine makov management manipulation manjunath manmatha maron matching meeting melbourne miller mixture model modeling models mori multimedia multiple nature niblack nips object omohundro pages pattern pektovic pentland perez photobook picard pichunter pictures place ponte prentice proc project qbic quadratic quantizing query querying rabiner recognition references relevance representation result results retrieval santini sawhney scalable sclaroff selection semantics shape shows sigir signals singapore smeulders smith snowbird sons speech spie springer statistical storage stork straight swain system systems takahashi taubin texture theory titterington trans transformation translation using utah vailaya vapnik vasconcelos vector verlag video vision visualseek vocabulary wiley with word words workshop worring yanker years yianilos zhang
http://doi.acm.org/10.1145/1076034.1076168	111	Mining Multimedia Salient Concepts for Incremental Information Extraction	algorithm analysis annotation automatic average bari barnard based bayesian best browsing canada color computer concepts conference considered content corel cross data dimensions duygulu early elements entropy european evaluated examples experimental experiments feature features figueiredo figure finite fixed forsyth freitas friedman gabor gaithersburg generative gupta hall hastie heesch howarth icml ieee image imahes inference intelligence international italy jain jeon koller lavrenko learning level lexicon machine magalh manmatha marginal mean measures media mining mixture model models more neapolitan networks object only optimal pattern pickering precision precisions prediction prentice presented presents probabilistic proceedings recognition reduce refer references relevance reported result results retrieval retrievals sahami santini search selection sigir smeulders space springer statistical stock subset table tamura test tested than that tibshirani toronto total toward train training transactions translation trecvid unsupervised used using video vision vocabulary vries were westerveld with workshop worring worst yavlinsky years
http://doi.acm.org/10.1145/1076034.1076136	79	Basic Issues on the Processing of Web Queries	aasheim american architecture authoritative barb barroso chile cluster conference congress coupled dean digital distributed dobr eiro engines environment erformance erlinked first holzle http ieee journal kleinb latin libraries lidal main micro multi neto ogle page pages planet proceedings query references risvik santiago search sources third tier tightly volume
http://doi.acm.org/10.1145/1076034.1076123	68	Iterative Translation Disambiguation for Cross-Language Information Retrieval	accurate adriani algorithm alignment ambiguities ambiguity analysis annals annual another application approach approaches asian association based bigrams bootstrap braschler brin bringing brooks buckley butterworths cambridge chen church citation clef coincidence cole coling combining company comparisons computational conference crof cross data database davison decaying decision dempster dependency development dictionary digital disambiguation document dunning dutch edition editor editors effective effectiveness effects efron electronical empirical engine european evaluation exploring extraction fellbaum forum foundations fourth frequencies from german gonzalo goodman hanks harman hinkley hollink incomplete inference information international introduction iral italian jackknife jang journal kamps keller kikui kitchens kluck laird language languages lapata lexical lexicography library likelihood lingual linguistics linguitics list lncs look machine maeda management manning maximum meeting methods mitra model modeling models modern moffat mono monolingual monz morphological motwani mutual myaeng natural nist norms obtain occurrence order page pagerank pages parametric park part performance peters phrase pirkola porter press probabilistic proceedings processing program publication publishing query ranking references relations report research resolve resolving retrieval rijke rijsbergen royal rubin sadat savoy schmid schutze science search sense setups shallow sidl sigir significance similarity singhal smart smoothing society special speech springer stanford statistical statistics stripping structure study suffix surprise syntactic tagging technical techniques template term tests text their translation trec trees uemura university unseen using vectors venugopal verlag vogel waibel weighting wilbur wilkinson winograd with word wordnet workshop yoshikawa zhou zobel
http://doi.acm.org/10.1145/1076034.1076051	11	Modeling Search Engine Effectiveness for Federated Search	academic addison advances american approaches australasian australian avrahami baeza broglio buckley callan collections comparison conference craswell croft database distributed editor experiments federated fedlemur information inquery institute journal kluwer management methods mitra modern nation national neto press proceedings processing project publication publishers real references retrieval ribeiro salton science search selecting singhal smart society souza special standards techniques technology text thesis thom tipster trec university using wesley with world yates zobel
http://doi.acm.org/10.1145/1076034.1076116	63	An Exploration of Axiomatic Approaches to Information Retrieval	academic addison advanced advances american analysis application applied approach approaches approximations automatic axiomatic axioms based basis beaulieu brief bruza buckley ccurrence cheng clustering committee computer conference croft cument data dearing deling dels dern development documentation ecifity editor editors effective enchmarking engineering ergen ertson evaluation exploratory exploring fang fields formal forum fuhr functional gatford generation grieff hanco harman hartiwig heuristics hill huib ieee inference information interpretation introduction investigating jones journal kleinb kluwer lafferty language length letin logics management mcgill mcgraw metho mitra modeling models modern moffat network nips normalization oisson okapi ortance ossibility othing outness overview pages pivoted ponte probabilistic proceedings processing publications publishers query references relevance representation research retrieval rijb sage salton science search sept sigir similarity simple singhal society some song space sparck statistical study systems technical term terms text theorem theoretical theory theuse third towards transactions transformation trec turtle uncertainty using walker weighted weighting wesley wong yang zhai
http://doi.acm.org/10.1145/1076034.1076067	23	Using ODP Metadata to Personalize Search	accurate aggregation algorithms amsterdam analysis annual approach authoritative bandar between bordermanager brin bringing capture capturing citation combating communications computer conference contextualized crawling data database design directory dmoz dwork effect efficient electronic engine engineering engines environment ester experimental filtering first garcia google haveliwala hawaii hill http hyperlinked ieee information international intl jordan journal kleinberg knowledge kolesnikov kriegel kumar lempel lexical link lipton lists mcgraw mclean measuring methods middleton miller molina moran motwani multiple naor netherlands networks novell ongyi ontologies open order page pagerank pendersen persona personalized preferences press principles proc proceedings project rank ranking recommender relevant report roure salsa scaling schubert sciences search semantic sensitive shadbolt sigir similarity sivakumar sources spam stable stanford statistical stochastic structure system systems tanudjaja technical topic transactions trustrank university user using vldb webbase websites widom williamson winer winograd with wordnet words zheng
http://doi.acm.org/10.1145/1076034.1076047	8	Active Feedback in Ad Hoc Information Retrieval	aaai accident active adaptive algorithm american analysis applications approach based bayesian beyond bibliography buckley callan cambridge carnegie catlett chen cikm classification classifiers clickthrough cluster cohen cohn content data divergence document documentation employing engines entropy error estimation evaluation experiments exploitation exploration fall feedback filtering finding forum framework gale given groups hard harman heterogeneous http icml ieee implicit improving independent inferring information introduction jaakkola joachims john jones journal kaufman kelly koller lafferty language learning lemur less lewis little machine mccallum measures mellon methods metrics microsoft minimization model modeling models more multimedia nigam nips notes optimal optimizing pages performance pool preference proceedings query reduction relevance retrieval revisited risk robertson rocchio rousseeuw salton sampling schohn science search sequential shannon shen siegelmann sigir sigkdd smart society sons sparck stanford subtopic supervised support symposium system taylor teevan term text theory thesis through tong toolkit toward track training transactions trec uiuc uncertainty university unpublished user using vector weighting wiley with working zaragoza zhai zhang
http://doi.acm.org/10.1145/1076034.1076050	10	Server Selection Methods in Hybrid Portal Search	abbaci ackerman adamic advanced aiqun algorithms alison alistair allison alto alvarado amit analysis anchor anchors andrei anthony approach approaches april august australasian automated automatic bailey based behaviour bernado better broder bruce callan case center charles christine cikm cluster collection collections combining comparing computing conference connell cope cori cost craig craswell crimmins croft ctor database databases david decision digital directed discovery distributed distribution document dynamics ecic effect effective emmitt engine engines enough estimation evaluating evolutionary experiment factors february finding forum framework francis french fuhr garc gloss gravano hannes hawking henrik henzinger http huberman human improving indices inference information interfaces internet ipeirotis jacques jaime james jamie january jared jinxi july june karger kaszkiel kevin lada language large lempel libraries link luis marais marcin margaret mark merging method michael mingfang modeling models moffat molena monika moran moricz networks nick norbert nottelmann november october ogilvie optimizing orienteering over overview pages palo panaglotis papers paris paul perfect performance peter powell prefetching press prey proc query rasolofo references relevant report research resource result results retrieval robertson rong ronny ross sample savoy search searching segmented selection server shlomo shrinkage sigir sigmod silverstein singhal site size source springer stephen study systems taxonomy technical techniques teevan testbed text theoretic tomasic toward track tradeoffs transactions travis trec trystan upstill using very viles vldb webgrowth weighting when wide wilkinson with workshop world xerox yves
http://doi.acm.org/10.1145/1076034.1076190	132	The Recap System for Identifying Information Flow	able above affected algorithms allowing also annotated answering answers applications arda area article articles aspects august australian author background basis because been bernstein block both brazil browsing bysentence categories ciir circled cloud collections combine come concise consider considerably copyright could council croft data date dead deaths demonstration derived description detection determine developed diminished discover diversity document documentlevel documents elements embedding emphasis engines enters entire eruption even example experimentation fact factual favor finding flow form found from funding general given globe grant have held helens highlighted highly however identification indicators individual influenced information informative instances intended intention interest introduces intuitive involve keywords kinds latter left leveled likely list many match matches matching material measures metzler miles missing model moffat more most newspaper newswire newswires novel number occurred often original other over overview owner part particular passage people persuasive pertinent piece plagiarism possible potential previously properties prototype provided provides publication published queries question range ranked rapid recapitulation reference references related repeated repetition research retrieval reuse reveals rich salvador same search seek seen selected sense sent sentence sentences several showing shown shows sigir similar similarity simple since snippet software some source sources specialist specifically square standard statement statements statistical strength submitted such summaries supported system task terms text than that then there third this those timeline tracking transformations translation trust typical used useful user usual valid value verbatim when where which while whole wide with without words work years zobel
http://doi.acm.org/10.1145/1076034.1076055	14	A Probabilistic Model for Retrospective News Event Detection	aaai accordingly actually adaptive algorithm algorithms allan also although analysis approach approaches articles asterisk auto best better bigger bikel both brants callan caputo carbonell cause center characteristics chen choose classification clustering combined complex components conclustions conditioned conference consistently contents contextual data dataset detection development discovery documents dynamic each easily easy effectively effectiveness elements entities especially event events examine experimental experiments explicitly failure farahat feature figure filtering final find fitful fitness fits from furthermore future ghosh ground hastie hidden high hoberman homepage however http illustrates impact implement independent indicate inference inflexible information intel international ject jman journal keywords know kumaran labeled language lavrenko learning learns ledge ligent like line machine main many markov mccallum measure measures meng mining minka mitchell modal model modeled models mooney multi mutual name named news next nigam nist noises none novelty numb number obtain osition other overcomes page paper papka partition parts performance pierce practice prediction principle proc processing propose quality reasons redundancy references refine report representation representations research respectively results retrieval retrospective robert satisfied schwartz search seemed selected selection series sigir sigkdd similarity simplified since sliding some space sparseness speech springer statistical statistics step strehl study such summer system systems tests text than that them they think this three thrun thus tibshirani time timestamps topic touch tracking traditional trevor tried true truth typical under understand unlabeled usages using vectors version wayne weischedel were what which will windows with wong work works workshop yang zhang
http://doi.acm.org/10.1145/1076034.1076099	49	Boosted Decision Trees for Word Recognition in Handwritten Document Retrieval	aaai addison additive advances algorithm alon analysis annals annual approach approximate architecture arcing artifical athitsos australasian available based bayesian behaviour bengio boosted boosting boostmap breiman buja bunke cambridge cascade categories center centric civr class classification classifiers closer collection comprehensive computer computing conference croft cursive cvpr detection dial document documents dynamic editor efficient effort engine estimation experiments fast features fergus freund friedman govindara handwriting handwritten hastie historical hmms holistic howe http human ieee image images impacts improve independent indexing inference information integral intel intelligent international ject jones journal july kaufmann kollios lafferty language lavrenko learning least lemur libraries ligence ligent line logistic look machine manmatha manuscript manuscripts marti matching maybury mease media method metzler minimization model modeling models moffat morgan multi network november offline pages pami pattern performance perona plamondon ponte press principle probability proc proceedings programs quantile query quinlan rankings rapid rath recognition references regression report research retrieva retrieval retrieving risk robust scale schapire scientific sclaroff search second segmentation shot sigir similarity simple society space spotting srihari srimal statistical statistics submitted survey symposium system systems technical technique text texts theories thirteenth tibshirani tieu time toolkit trans transactions trees unconstrained unsupervised using video view vinciarelli viola vision volume warping wesley with without word workshop world wyner zhai zipf
http://doi.acm.org/10.1145/1076034.1076114	61	A Study of the Dirichlet Priors for Term Frequency Normalisation	achieve acknowledgments addison also amati amsterdam annual application applications applied apply applying association bailey based beaulieu better buckley chen classical compostela computational computer computing conducted conference cost could craswell data default degroot department development device dirichlet divergence document documentation ecir edition editors eighth empirical enable estimation european experiments found fourth francisco frequency from functionalities funded funds gaithersburg gatford gelsema glasgow goodman grant have hawking high http indexing information informative interesting international interpolated interpretation investigates ject jelinek jones journal kanal lafferty language large length leverhulme linguistics lowering march markov measuring mechanism meeting mercer method methods mitra modeling models modular more most netherlands nineth nist normalisation normalization number october okapi only orleans ounis overhead overview pages parameter parameters particular pattern payne performance pivoted platform possible practice previously priors probabilistic probability proceedings proposed providing publication query randomness rapid recognition references research rest retrieval rijsbergen robertson santiago scale science setting sigir singhal smooth smoothing solution source spain sparck sparse special specificity statistical statistics study suggested switzerland systems techniques term terms terrier text these thesis thirty this though tois track transactions trec trust tuning university using version volume voorhees walker were wesley which while will work would zhai zurich
http://doi.acm.org/10.1145/1076034.1076143	86	Evaluating the Impact of Selection Noise in Community-Based Web Search	adapted adaptive analysis balfe based blott boydell briggs camous collab community conference connor coyle dels detection engine engines eriments etition european existing exploiting ferguson freyne gaughan genomic gurrin information interaction manipulating modeling murphy novelty orative press proceedings query references regularity relevance retrieval search searching similarity smeaton smyth springer terabyte text thirteenth trec user verlag wilkins
http://doi.acm.org/10.1145/1076034.1076177	120	Self-Organizing Distributed Collaborative Filtering	advanced algorithms analysis annual based being belonging between both breese buddy class collaborative communication computing conference consequently cpeer distributed document download downloaded elements eleventh empirical evaluation file filtering follows generated generation heckerman hofmann ijcai imaging indicates information international item items kadie karypis kluwer know lafferty lagendijk language latent ledge locally management methods minimize modeling models moors networks number only other peer performs pisson pouwelse predictive probabilistic proc puzicha query recommendation references reinders relevance relevances report research retrieval robust school search series sharing south store survey systems table tables technical tells tenth that time towards transaction univeristy update updated using wales wang when where within words zhai
http://doi.acm.org/10.1145/1076034.1076139	82	Surrogate Scoring for Improved Metasearch Precision	advances aslam beitzel buckley chowdhury cikm combination condorcet determining effective effectiveness evaluate evaluating evaluation fast framework fusion improved information jasist jensen kernel learning machines measure methods minimal montague multiple necessary nist operational optimization pass platt press query references requirements retrieval same scalable search searches sequential shaw sigir sizes stability strategies support system systems training trec using vector voorhees
http://doi.acm.org/10.1145/1076034.1076098	48	A Phonotactic-Semantic Paradigm for Automatic Spoken Document Classification	alshawi bytype catalog classification edmonton effective extracted from hltnaacl http models phonotactic proceedings reference references results speech unsupervised upenn utterance with
http://doi.acm.org/10.1145/1076034.1076059	17	SimFusion: Measuring Similarity using Unified Relationship Matrix	advances agglomerative alberta algebra algorithm algorithmic american analysis anatomy annual approach approximate asia atlantic attributes authoritative automatic baltimore based beeferman belgium belief berger berkeley between bibliographic bibliometrics borchers boston brauen brin browse brussels bush calado california canada categorical catid chakrabarti chapter cikm citation cliffs clustering clusters collaborative communications computer computing concepts conference coupling croft cyberspace data database databases davison dean deerwester department design development digital discovery document documentation documents dumais dynamic editor edmonton eighth engine engineering englewood environment experiments exploratory external factors feedback filtering finding flake foundations fourth framework francisco fuhr furnas fusion ghemawat gibson giles groups hall harshman henzinger herlocker heterogeneous hill http human hyperlinked hypertextual identifying ieee implementation improve inference information integration intellectual inter international interrelated introduction isdn iterative jacm jersey july kallenberg kessler kleinberg knowledge konstan kumar landauer large larson latent lawrence libraries link literature logs management mannila mapreduce march maryland mcgraw measure measurement meeting mining model modeling modern modification monthly mrssa multi muntz neto network networks objects october operating optimal organization osdi over pacific page pages papers past performing popescul prentice probabilistic probability probes proc proceedings processing queries query ragavan raghavan rajagopalan recommender references reinforcement related relational relationship relevance report research resnick retrieval reuse ribeiro riedl rocchio rolleke ronkainen salton scale science scientific search seattle section semantic seventh sever shanghai sigir sigkdd simfusion similarity simplified simrank sixth small smart society sources space special spreading springer structuralcontext structure switzerland symposium system systems tech technical temporal text think tods tois tomkins toronto toward transactions trends turtle type ungar unification unified user using varian vector verlag very virginia washington wide widom wong world york zhang ziarko zurich
http://doi.acm.org/10.1145/1076034.1076106	54	Relevance Weighting for Query Independent Evidence	adding additional adjustment after although amento analysis anatomy annual appropriate australian authority baseline begun bishop brin buckley cambridge characterize characters cikm classification clickdistance cocoon combinatorics computer computing conference could craswell cument cuments development direction directly document each effective effectiveness engine entry ertextual ertson estimate estimates evidence example experiments expert extension fagin feature features fields figure finding first flat floe form functional gaithersburg gives good hard have hawking hiemstra hill home however improve improvement indegree indep independent indicate inferior informaion information international investigation isdn itself kamps kang know kraaij kumar large ledge length level limitation line lines link management maryland mccurley mean meaning microsoft mitra more much multiple national needed networks neural normalization novak novemb only ortance outcome overview oxford page pagerank pages pandurangan pattern pivoted plus possible precise predicting press prior probabilities proceedings quality query raghavan ranking rather ratings recognition references related relevance research result results retrieval rijke rnsson saria scale score search searching second short sigir sigmoid sigurb simple singhal sivakumar skip slight slop specific springer static step structure syst system systems taking taylor terveen than that there thesis thirteenth this tomlin track tracks training trans trec tuning twelfth university upfal upstill upward urls used using verlag very weighted weights were westerveld wide williamson workplace world would york zaragoza
http://doi.acm.org/10.1145/1076034.1076185	127	An Industrial-Strength Content-based Music Recommendation System	audio cano contest description gouyon herrera html http ismir koppenberger press publications references serra streich wack
http://doi.acm.org/10.1145/1076034.1076131	74	Analysis of Factoid Questions for Effective Relation Extraction	accounting analysis animate answering asist asked askmsr banko before brill broder carroll clark cogex color complex context cover coverage date dumais echihabi emnlp etzioni extraction fleischman forum founding frequent furthermore gerber harabagiu hermjakob hovy information instance jeeves junk knowitall learning length list literature location logic maiorano moldovan more most neighbors observed offline online ozmutlu patterns people piquant preliminary prover quantity query question questions ravichandran references relation relations reports results sample scale search server shows sigir speed spink strategies surdeanu surface system table taxonomy text than that they total trec type types webclopedia what with
http://doi.acm.org/10.1145/1076034.1076058	16	OCFS: Optimal Orthogonal Centroid Feature Selection for Text Categorization	alberta algebra aliferis analysis annual applications applied automated banff based belkin berlin canada categorization chambers characterization classification component computing conference croft decomposition discriminant douglas duda eddy fabrizio feature filter first florida franca generalized generalizing gentle greengrass haerdle hart howland icml ieee information intelligence international ioannis james jolliffe kaufmann lang learning linear machine morgan netnews newsweeder november numerical park pattern principal proceedings references retrieval review science selection sheather singular spriger springer statistics stork supervised survey symposium techniques technology term text theoretical tierney transactions twenty using value verlag weighting wiley york
http://doi.acm.org/10.1145/1076034.1076056	15	Scalable Collaborative Filtering Using Cluster-based Smoothing*	aaai algorithm algorithms alleviate analysis annual application applications applying approach architecture artificial associative august balabanovic based bergstrom berkeley better billsus breese brochers brown case chen cimca class claypool clus cluster clustering collaborative combining communication communications computational computer conf conference conner constant content contentbased cooperative data desouza development diagnosis different dimensionality eigentaste empirical environments ester evaluating evaluation feature figure filtering filters fisher foster framework francisco generalized giles gokhale goldberg gram grouplens gupta heckerman herlocker hildrum hofmann hong horvitz huang hybrid iacovou information instance intelligence international itembased items joint kadie karypis kaufmann kohrs kong konstan language latent lawrence learning linguistics machine maltz management means measure memory menlo mercer merialdo methods miller mining miranda model models morgan murnikov national natural netes netnews newman news newspaper nicholas number online open pages park pazzani pennock performance performing perkins personality pietra popescul predictive press probabilistic problem proc proceedings puzicha recommendation recommender reduction references research researech resnick retrieval riedl roeder sartin sarwar selection shoham sigir similarity sixteenth small soboroff space sparse sparsity study suchak supported swami systems techniques terveen text thomas time transactions uncertainty ungar unified usenet value vector vuduc webkdd weighting wide work workshop world xing zeng zhou
http://doi.acm.org/10.1145/1076034.1076094	45	On The Collective Classification of Email "Speech Acts"	aaai ablex action acts agreement analysis annual appear approach archives assessing association barcelona bayesian behavior berger carletta carvalho categorization chakrabarti chickering claims classification classify cognition cohen collaborative collective communication computational computers conditional conference construction coordination corp data deliberation della dependency discovering distributions eighth electronic email emnlp enhanced entropy espinosa evidence fields filtering flores four from fussell gallagher game geman gibbs heckerman heuristics html http hypelinks hypertext identifying ieee ijcai images improves inducing indyk inference intel into iterative jensen journal july kadie kappa kleinberg kraut language leadership learning lerch leusky ligence ling linguistics machine management maximum meek methods minorthird mitchell model models murakoshi naacl names natural negotiations networks neville norwood ochimizu ontological organizational pacific parsing pattern people pereira perspective pietra predicting press proc proceedings processing progress publications publishing random rattigan references regularities relational relations relaxation research restoration roles rounthwaite schoop seattle shallow shimazu sigir sigkdd sigmod simulated sourceforge spain speech stage statistic statistical stochastic structure survey tasks teams text transactions understanding using visualization vitor washington winograd with work workgroups working workshop
http://doi.acm.org/10.1145/1076034.1076102	51	Evaluation of Resources for Question Answering Evaluation	academic allan american amsterdam annotation annual answer answering answers aslib association bakshi banko based brill buckley building case chapter cieri clarke cleverdon collection collections computation computational computer conference construction context cormack corpora counting cranfield criteria data detection determining development document documents dumais editor effect effectiveness efficient eleventh england error evaluating evaluation event experiment experiments extracting factors fernandes from good graff human huynh ifip incomplete indexing information institute intensive interact interaction international ject journal judgments karger katz keen kluwer knowledge language large liberal liberman linguistics logic makes martey marton massachusetts measure measurement meeting mills mining monz naacl negligible ninth north norwell organization overview pages palmer performance press proceedings publishers quan question references relevance reliable rennert research results retrieval reusable role scale science sigir sinha size society sormunen stability strassel study systems techniques technology tellex tenth test text thesis tice topic track tracking trec university using variations volumes voorhees what with zobel
http://doi.acm.org/10.1145/1076034.1076148	91	Relation between PLSA and NMF and Implications	advances algorithms aligning analysis based buntine categorising chen clustering computer document documents ecml extensions factorisation factorization finite gaussier gong goutte hierarchical hofmann information jects kaufmann latent learning lecture matrix mclachlan mixture model models morgan multinomial nature negative nips notes pages parts peel popat press probabilistic references retrieval science semantic seung sigir springer using variational wiley words yamada
http://doi.acm.org/10.1145/1076034.1076183	126	On Redundancy of Training Corpus for Text Categorization: A Perspective of Geometry	analysis automatic based categorization classifiers collections etition examination generalized instance khmelev kisiel measure proc references scalability sigir teahan text using verification yang zhang
http://doi.acm.org/10.1145/1076034.1076090	42	Publish/Subscribe Functionality in IR Environments using Structured Overlay Networks	achieve additional after algorithm algorithms also among anked balancing cause causes closely conclusion considered cost developed dhtrie differences different distribute distributed distribution document does easier efficiently entries evaluation expected experimented fcache figure filtering first follows from graph group here however important imposed imposing increase increases leads load manages message messages misses more much needed network nodes notice number observation observe observed occurring outlook overloading peers perform point price protocols purposes queries received reduced references relatively requests responsibilities revealed routing scenario section show shown shows significant significantly small splitting strengths summary task terms that this time uniform uniformly updated values weaknesses when which while window with
http://doi.acm.org/10.1145/1076034.1076141	84	Characterization of a Simple Case of the Reassignment of Document Identifiers as a Pattern Sequencing Problem	account address also although annual applications assigning barreiro bell better bits blanco blandford blello case chen chung clustering coded codes coding compressing compression computers conference consequences considering cument cuments data dern development dimensionality dkftk document documents ecir edition encoding enhance erty european file final finds fink formalized francisco fulltext gigabytes given goal good graph here heuristic however identifier identifiers ieee images important index indexes indexing information integer international into inverted january kaufmann lists lncs logs management managing metho minimal minimization minimize moffat morgan most must notice obtain only operations order orlando pattern perego posting practical problem problems proceedings processing producing products prop publishing ratios real reassignment reduction references reordering represent research results retrieval scheme search sequencing shann shieh sigir silvestri similarity solution static strategy take that this through unary using weighted with witten would
http://doi.acm.org/10.1145/1076034.1076159	102	Examination and Enhancement of a Ring-Structured Graphical Search Interface Based on Usability Testing	addison advanced advances again annotate baeza browsing compare computing concentric conference could environment find following found groups huang image imagegrouper images information integrating intelligent interactive international joint kajiyama kando letters ljubomir make manola march modern more munehiro nakamaru nakazato namely neto obtained ohno organize pages participants performed previous proceedings reading recent references relevant results retrieval ribeiro ring same search searching second september seven simple soft specializing students symposium system systems task test testing they this thomas three university usability using view visual were wesley with yates
http://doi.acm.org/10.1145/1076034.1076174	117	Using DragPushing to Refine Centroid Text Classifiers	advances algorithms analysis approach based bengio calahan categorization centroid classification classifiers data document experimental handling html http idiap information karypis lewis linear methods misfit model neural papka phang pkdd processing project projects references refinement regularized research result reuters schapire sigkdd svmtorch systems text theo training winnow zhang
http://doi.acm.org/10.1145/1076034.1076087	40	PageRank without Hyperlinks: Structural Re-Ranking using Links Induced by Language Models	access accomplishing accuracy acknowledgments adjectives advances aggregates algorithms allan alternative amit analysis anastasios anatomy andrew annual anton application applied approach artificial association assume attention authoritative automatic available based baselines better bipartite book brin bringing bruce buckley callan carmel case centrality chains charles chengxiang chirag chris christopher cikm citation classification cluster clustering combined combining companion comprise computation computational computations conclusion conference connection corp corpus croft cross cuts daniel data david deguzman detection dhillon different dinit direct distributions djoerd document documentation documents domshlak dragomir eacl ecific ectral edition editors education edward effective effectiveness eighth elieve emnlp employing endency engine engineering entire entry environment ergen eriments erkan erlinked ertextual etter etween eugene evaluated evaluating evaluation even explicitly exploring extraction feedback fernando filtering finding forum fraction framework francis fruitful further future gabriel garfield gather gene generation geoffrey goal golub good graph graphs grassmann grimmett gune happ hatzivassiloglou have hearst heuristics heyman hidden hiemstra hierarchic high hopkins human icml implemented improving inderjit inducing influence influx information initial instantiations intel inter interactive international into introduction intuition isolateddocument iterative james jamie jectivity joachims john johns journal kathleen kenney kleinb kluwer know kraaij kristina kurland lafferty laflamme lalmas language large lavrenko lawrence learning ledge leek lemur length leuski lexical lexrank ligence likelihood lillian line linguistics literature loan machine main management managment mandar manning markov markovian marti matrix maximum mckeown meeting method methods michael mihalcea miller minimization minimum mitra mixture model modeling models more mounia naftali narin necessary neural ninth nips noam normalization novel numb numerical ogilvie operations optimized order oren orientation ortance osed other othesis ottleneck overall oxford page pages pang partitioning paul pedersen pereira peter physics pinski pivoted pollard ponte poster predicting preliminary press princeton prior probabilities probability proceedings processes processing promising prop proximity pseudo publications query rada radev random ranking ratio real reasonable recursive reexamining references regenerative relationships relaxation relevance relevant research restrict results retrieval returned revealed review richard rijsb risk rivaling ruthven salience scale scatter schwartz science scientific search seem seems semantic sentence sentiment sentimental sergey series seventh several shah showed sigir sigkdd similarly since singhal slonim smoothing solution sources space stage standard state statistical steady steven stewart sticking stirzaker structural structure study successful summarization surprising survey system systems taksar tarau techniques technology tenth text textrank texts than thank that theory there they thijs thing third this thomas thorsten those time tishby tombros tool toolkit topic toutanova track tracking transductive trec twenty uniform university using vasileios vector veera victor villa volume walk weighted were wessel westerveld whether which wide willett william winfried with within word words world would xiaoyan xiaoyong yield zhai
http://doi.acm.org/10.1145/1076034.1076045	6	Context-Sensitive Information Retrieval Using Implicit Feedback	acknowledgments adaptive adar allan american amherst anonymous applied automatic award based belkin bharat capture career challenges chien cikm city clickthrough comments concept constructed context contextualized croft cronen cutrell data delos demo description digital display divergence document dumais dynamic ecir effects effort engine engines environments expansion experiments explicit exploiting feedback finkelstein forum foundation from gabrilovich hall hatano haystack history horvitz huang identification implicit inferring information interactive joachims jose journal kansas karger kelly lafferty language larvrenko libraries logs material matias methods model modeling models national numbers optimizing oyang page pages part peng personalisation personalization personalized perspective placing poster preference prentice probabilistic proceeding proceedings processing proeedings profile queries query ranking recommender references relevance retrieval reviewers revisited rijsbergen rivlin rocchio ruppin ruthven sarin scaling schuurmans science search searchpad second session shen sigir sigkdd simulated smart smoothing society solan sriram statistical study suggestion sugiyama support supported system systems task technology teevan term thank their this time townsend under understanding university upon useful user users using white widom with without wolfman work workshop yoshikawa zhai
http://doi.acm.org/10.1145/1076034.1076160	103	Shortcomings of Latent Models in Supervised Settings	allo analysis andrew blei canny cation data david dirichlet discrete factor hofmann indexing john jordan journal latent learning machine michael nips pages probabilistic references research semantic sigir thomas
http://doi.acm.org/10.1145/1076034.1076080	34	Multi-Label Informed Latent Semantic Indexing	aaai achieves actually advances agree also american analysis ando annual appendix application applying approach arbitrary arsenin back bartlett benchmark between biometrika cambridge canonical case categorization classification collection columns combination compare component computation conclusion conference considered correlation corresponding cost cristianini cvrt deerwester define denote derivative diagonal document dumais easy efore eigenvalue eigenvalues eigenvector eigenvectors elements equation first fixed flating following formalism formed friedman function furnas generalized given hardoon harshman hastie have hence hilbert holds holloway honour hotelling however improves indexing information inserting inter international into iterative ject jection jective jopt journal kernel kernels known label lagrange landauer largest latent learning least lewis lies linear locality london machine matrix maximal mccallum means measurement methods minimum mixture model modeling moment muller multi must neural nonlinear notation obtain obtained obviously only optimal optimization orthogonal overview pages papers part partial pattern perspectives posed precision preserving press principal probability problem problems procedure proceedings proof proofs prove proved proves references regression relations replace report representation reproducing research resp respect results rewritten rose rosipal rotation royal rule same satatistical satisfying scaling scholkopf science second semantic sets setting shawe sigir similarity simplicity since smallest smola society soft solution solutions solving space span springer squares statistics subspace suggests support symmetric szedmak taylor technical term text that then theorem therefore this thus tibshirani tikhonov trained trejo univeristy university vaat value variables vaxt vbbt vector verlag where which wiley with wold workshop write yang york zero
http://doi.acm.org/10.1145/1076034.1076162	105	Using Query Term Order for Result Summarisation	amsterdam automatic benjamins boston devlin experience feature john liang mani proceedings references selection summarisation summarising sunderland tait
http://doi.acm.org/10.1145/1076034.1076192	134	Manjal - A Text Mining System for MEDLINE	august author beneficial biolink biology brazil copyright curcumin diseases fish graphviz held http knowledge libbus longa medicine medline mining owner pages perspectives postulating public raynaud references retinal role salvador sehgal sigir srinivasan swanson syndrome undiscovered
http://doi.acm.org/10.1145/1076034.1076171	114	A Temporally Adaptive Content-Based Relevance Ranking Algorithm	allan analysis annual applying banff based beitzel brin bringing buntine categorized chowdhury ciety citation computer conference data development digital discrete endent engine exploring frieder grossman gupta hourly http ieee indep information intel international jakulin ject jensen kauppalehti khandelwal large library ligence motwani oral order page pagerank pages perki perttu press proceedings query ranking references research retrieval search sigir stanford summaries technical technologies temp topic topically topics trends very winograd
http://doi.acm.org/10.1145/1076034.1076077	32	Efficient and Self-tuning Incremental Query Expansion for Top-k Query Processing	accessible achieve adding aggregation algorithms allan analysis approach approximations aref asso australian automatic balke based behavior behaviour benchmarks billerb bruno buckley candidate cess cessing chang chaudhuri chien ciated ciation cikm clearly combination combining comput concept conclusion conference contextual cost could croft cronen cument data database databases davis decent decomp disambiguation distributed early effect effective efficiency efficient electronic elmagarmid engines ensive environment environments eriments ertson especially evaluating evaluation examination execution exhibiting expansion expansions extension fagin fast feature feedback fellbaum fields figure files filtered framework frei frequency fuhr function fuzzy global good gravano guarantees hard heterogeneous huang hwang icde ilyas image impact improving incr incremental indexes indexing information inputs interactive inverted issues itcc jasis jasist jcdl join kretser kwok large level lewit lexical ling logs long marian memory meng merge method middleware minimal mitra moffat multi multimedia multiple natsev nearest neighb nepal novel ntzer oakes optimal optimization optimized optimizing ordering orhees orting osed ositories outperformed over overview oyang page paper parameters persin pfeifer phrases pircs poisson precision predicates prediction presented press probabilistic probing proved pruning quality queries query queryresult questioning ramakrishna rand ranked ranking recognizing record references relational relations relevance relevant retrieval revisited robust rprec sacks salton scalable schenkel scholer score search searches selection selective self semantic sense session sharma sigir sigmod simple singhal some sort sorted space spire static stoko stream suel suggestion suite supp surrogates syst table tait taylor technique techniques terabyte term termination tested text that them theobald this time tkde tods tois towards townsend track traditional transformation trec turpin using utilizing vague various vector vertically very vldb vries walker weighted weikum while williams with word wordnet zaragoza zhou
http://doi.acm.org/10.1145/1076034.1076154	97	A Dual Index Model for Contextual Information Retrieval	abdul after allan another based been best bpref bruce challenges constant corpus data document documents equal evaluate evaluation experimental experiments forum from functions genomics gordon hard harper have here hmms huang index indexed iscas jaleel jiang known language level measures merge modeling nist notebook number other overview paragraph paragraphs passage performance precision proc proceedings proposed publication references relevant result results retrieval retrieved robert sets sigir special terms they this three topic total track tracks trec tuning uiuc umass univ university used using wdnew weight where york zhai zhang
http://doi.acm.org/10.1145/1076034.1076111	59	Personalizing Search via Automated Analysis of Interests and Activities	adler agent american anick anticipating arbitrary asisit based bharat both browsing budzik cadiz capture case chafee context contextualizing craswell cutrell documents dumais evaluation explicit feedback gauch google hammond hawking highly http inen information intelligence jancke journal kendall labs methods modification needs ontologybased overview personal personalized pretschner proceedings rankings references refinement relevant retrieval retrieving robbins rvelin sarin search searchpad seen sigir society statistical study stuff support system systems terminological ties track trec using watson
http://doi.acm.org/10.1145/1076034.1076165	108	Assessing the Term Independence Assumption in Blind Relevance Feedback	allan analysis approach automatic because bigi carpineto conference context corpus croft data databases diaz effectiveness expansion extension gauch give improving information intervals jaleel larkey meaningful metzler mori multiple noteb omitted points proceedings query rachakonda references retrieval romano smucker strohman systems terms text theoretic thirteenth transactions trec turtle umass wade wang with
http://doi.acm.org/10.1145/1076034.1076137	80	An Interface to Search Human Movements Based on Geographic and Chronological Metadata	access annual archives browsing byrne conference detecting digital events greenberg gustman history international joint large libraries oard oral picheny proceedings ramabhadran references sigir smith soergel structured supporting text
http://doi.acm.org/10.1145/1076034.1076105	53	A Study of Relevance Propagation for Web Search	addison agents algorithms amento amitay authority baeza bharat carmel darlow distillation does environment expert henzinger hill hyperlinked improved information knowledge lempel mean modern neto pages predicting proc proceedings quality ratings references retrieval ribeiro sigir soffer terveen topic trec wesley with yates
http://doi.acm.org/10.1145/1076034.1076157	100	Study of Cross Lingual Information Retrieval Using On-line Translation Systems	about across ambiguity applies approach average ballesteros based baseline bertoldi best bilingual brown chinese choquette clir collection comes computational consists content corpus croft cross crosslanguage curves description document documents each english evaluation examine examples expansion experiment federico figure from generated historical http imageclef improving information into language lavrenko learns length lingual linguistics machine mathematics model models online phrasal picture points precision precisions probabilities proc provided provides queries query ranked recall refer references relevance resolving retrieval shef shows sigir simply statistical summarized system table techniques that these this translate translation translations used using which words
http://doi.acm.org/10.1145/1076034.1076107	55	Detecting Dominant Locations from Search Queries	aaai according accuracy achieved addressing advantage agreed algorithms also alto always american amitay analysis annual another answer answering answers applied approach around askmsr austin banko bases because both bourigault brill butterworths cairo canada capturing categorizing check chunk church cikm classify cluster clusters coling combination computing conclusion conclusions conference consistently consult content contextual continue correct costs cucerzan current data date defined detect detecting detection develop development ding does dominant dumais edition edmonton effectively egypt entity every evidence exclusion existing exists experimental extraction facilitate false fast france from future geographic geographical geography geonames geotagging given gnis google grammatical gravano hatzivassiloglou have hierarchy high hmmbased http hybrid implicit important improve improvement improving including inclusion incurred independent information infoxtract intended intention internal international into issues july know knowing knowledge language large lichenstein lists live local locality location locations logs london look lookup maintenance majority management measure measures meeting method micro minimizes mining model modified most named names nantes natural negatives normalization normalizations north noun novel npanxx numbering open orleans other otherwise outcome outperformed palo paper parser parts performance phil philadelphia phrase phrases plan popular positives postal power presented proc processing program proposed qdls queries query questing real references relevance rely research resource results retrieval rijsbergen scale scopes search second services sheffield shivakumar show sigir sivan soffer solution sources speed spread spring srihari states stochastic suppressed surface symposium system tagger tagging taipei taiwan terminological test texas texts that their them this thus time tokenizing true unified united unrestricted usage user users usgs using usps venue very vldb wareonearth well when where will with work workshop worldwide would yahoo yarowsky zhou
http://doi.acm.org/10.1145/1076034.1076193	135	UCAIR: A Personalized Search Toolbar	according accuracy adaptive additional advantage agent algorithms appealing assisting assume attempt august author back based been boundaries brazil browser button captures clicked clicks clickthrough client coherence collect components constructed context contextual copyright corresponding current data delivered down ecir effective effectively efficiently effort engine engines example existing exploit factors feedback figure first formulation forum from functionality functions general google hatano held high history human identify immediately implicit improve improves improving including inferring infers information interaction joachims jose kelly keywords kunj link major many match model modeling models modification modify module more need next once online optimizing originally over owner page pages past perform personalization personalized phoi plug preference proceedings profile pull pushing queries query rank ranked ranking ranks recognized references reflect relevance report requiring result results retrieval rijsbergen ruthven salvador same search selectively sensitive session shen shown side sigir simulated specifically study submitted such sugiyama summaries summary support technical technique terms that their this those thus toolbar toolbars ucair uiucdcs unseen updated user users using view viewed well when whenever while white with without would yoshikawa zhai
http://doi.acm.org/10.1145/1076034.1076062	19	Combining Eye Movements and Collaborative Filtering for Proactive Information Retrieval	acoustic acoustics adaptation advances affective agents algorithm algorithms allocation analysis anderson appendix applications applying artificial asymptotically attentive automated automating barrett based basilico berlin bibliography blei bolt bounds buntine cambridge campbell challenges codes cognitive collaborative communications computer computing conference content convolutional data decoding direction dirichlet disclosing discriminative display donnelly ecml editors elomaa emotional environments error estimation european experiences experimental expressions extensions facial factors fast feedback filtering finland first forum free from gales gaze general genetics genotype grouplens hands heikkil herlocker hidden hofmann http human hyrskykari icassp icml ieee implicit independent inen inference inferred inferring information institute intel intelligence interact interaction interface interfaces international irem japan jaranta jordan journal journals kaski kaufmann kelly kitakyushu konstan kyushu latent lawrence learning letin letters ligent lisetti list machine mackay maes maglio maltz mannila maps markov marlin maui menozzi methodology miller model modeling models morgan mouth movement movements multilocus multimedia multimodal multinomial nasoz nature network networks neural news optimum organizing oulu pages partala pascal pattern pennock perception performance person pietik popescul population povey preference press pritchard proactive probabilistic proceedings processing profiles protocol psips psychological psychology puolam rabiner rating rauterberg rayner reading recognition recommendation relevance research response responsive retrieval royal salo salvucci science sciences scientific selected self selker semantic sensory series shardanand sigchi sigir signal silv simola social society sparse speech springer starker statistical stephens structure suitor surakka system systems technology teevan tenth theory toivonen transactions trends tutorial twenty ungar unified unifying usenet user using vanhala variational vision viterbi volume ward wesson woodland word workshop writing wsom years york
http://doi.acm.org/10.1145/1076034.1076150	93	Using Oracle r for Natural Language Document Retrieval An Automatic Query Reformulation Approach	approach automatic conference developments dixon eighth harman lexical mahesh oracle orhees overview proceedings references retrieval salton science text trec
http://doi.acm.org/10.1145/1076034.1076173	116	Dependency Relation Matching for Answer Selection	answer answering answers approach based baseline brazil chua comparison densitybased dependency evaluate evaluation evaluations exclude factoid final from granada have highest main maslennikov minipar module national note parsing passage proc qualifer question questions ranked references relations retrieval salvador score select selecting selection sigir singapore spain state string system systems task test that this those three trec university using workshop yang
http://doi.acm.org/10.1145/1076034.1076145	88	Search Engines and How Students Think They Work	about american artificial basic belew cambridge clearly concepts conceptual content creation distribution efthimiadis engines esoteric figure finding fleiss from hendry however information interaction internet john journal lenhart life like match media methods metrix models more most move online presence press prevalent project property proportions query rankings rates references release reston results science sciences search shown simon simple sketches society some sons statistical submitted systems technology that these this understand university users when wiley with york
http://doi.acm.org/10.1145/1076034.1076075	30	Simplified Similarity Scoring Using Term Ranks	allan amherst annual aslam beaulieu belkin buckley callan center challenges conference croft development dumais early editors effective efficient forum fuhr harman harper held hiemstra hofmann hovy impact information intelligent international kraaij kraft kretser lafferty language lavrenko lewis liddy louisiana manmatha massachusetts mccallum modeling moffat orleans pages ponte prager press proc radev ranking references report research resnik retrieval robertson rosenfeld roukos sanderson schwartz september sigir singhal smeaton space termination transformation turtle university vector voorhees weischedel with workshop york zhai zobel
http://doi.acm.org/10.1145/1076034.1076097	47	Automatic Music Video Summarization Based on Audio-Visual-Text Analysis and Alignment	abstracting acoustics alignment analysis applications asian audio automated automatic automatically bartsch based berkeley catch cepstral chai chorus chroma communication computer conference cooper creating digital effelsberg explore expo extraction features fischer foote france from girgensohn gong highlights ieee image information international istanbul journal juan lausanne lienhart logan media motion movies multimedia music paltz paris partial pfeiffer phrases pins presentation proc processing references representation representations retrieval signal similarity singapore snippets soccer spectral speech structural summaries summarization switzerland temporal thumbnailing tian turkey using vercoe video videos vision visual wakefield waspaa with workshop yeung york zhang
http://doi.acm.org/10.1145/1076034.1076176	119	Noun Sense Induction Using Web Search Results	aaai accuracy agirre ambiguous annual artificial association automatic automatically available baseline bias built categorized chose classifying clusters commonsense computational conference constructed context corpus correctly defined disambiguation discriminating discrimination each effect evaluate evaluation evaluations fewer finding first found fourth frequent from gate http hugo incorporating informed intelligence international into jose july language linguistics lisbon lrec major manually martinez media meeting methods montylingua most national natural nineteenth order pages pedersen percentage portugal presented proceedings purandare queries query random randomly references related represent resources resulting results returned rivaling search sense senseclusters senses some supervised system tank than that then these tools understanding unsupervised vectors were word yarowsky
http://doi.acm.org/10.1145/1076034.1076121	67	Learning to Estimate Query Difficulty	access accompany affinities agreement amati amitay annual apostolico application association automatic based berkeley beyond breiman britain buckley california callan cambridge carmel carpineto chapman classification clickthrough cluster coefficient cohen collections computer conference cook cost croft cronen current darlow data deng development diaz dictionary difficulty dinstl discovery distributed duda ecir editors educational engines european expansion experiments fail farchi fine forests friedman gain glasgow great grunfeld hall harman hart hummingbird inference inferring information institute international joachims john jones juru kernels know kwok language leaning learning lecture ledge lexical machine machinery machines manual matrix maximal mayfield mcnamee measurement melucci mining model models national networks nist nominal notes nrrc olshen optimization optimizing ounis overview oxford pages part pattern performance petruschka piatko pircs plachouras precision predicting prediction predictors press proceedings processing profiles psychological queries query random references refinement regression regularization reliable research retrieval robust robustness romano scales scholkopf science search searching searchserver seattle selective sense sigir smola soffer sons speech spire standards statistics stone stork string sunderland support swen technology temporal terabyte terrier text tomlinson townsend track tracks trec trees university upton using vassilis vector volume voorhees washington wiley with workshop xiang york zhou
http://doi.acm.org/10.1145/1076034.1076095	46	Using Term Informativeness for Named Entity Detection	again algorithm american approach approaches assumptions automatic averaged base baseline bayes best bikel biometrics bookstein breakeven brookes butterworths categorization church classifiers clifton comparisons conference cooley corpora corpus data databases dempster deviation different discovery distribution document documentation effectivenss engineering european evaluated everything experimental features fifth five four frequency fresh from gale gave gives harter historical hollander identification incomplete index indexing individual information institute international inverse joachims john jones journal karger keyword know laird language large learning learns ledge likelihood literature london look machine machines many massachusetts maximum measure measures methods mining mixture mixtures models naacl naive name natural nonparametric only pages papineni parameter part performance poisson poor practice principles probabilistic proceedings proposed ranking references regularization relevant rennie residual results retireval retrieval ridf rifkin rijsbergen royal rubin schwartz science score select selected series sets settings shih society sons specialty statistical storage support swanson swets table tackling technical technology teevan tenth term text that thesis third topcat topic trained training twentieth vector very weighting weischedel what wilcoxon wiley with wolfe words workshop
http://doi.acm.org/10.1145/1076034.1076064	21	Information Retrieval System Evaluation: Effort, Sensitivity, and Reliability	analysis applied automatic bibliography blustein bounds british buckley butterworths cambridge clear collection collections computer conclusions conf conference considering data development documentation documents dunlop effect effectiveness error established evaluating evaluation experiment experiments figure from future harman highly hull ideal incomplete increases indexing inen inference information interaction issue jones journal laboratory large library london lower management march matthews measure measures methods modeling need numbers overview proc processing provision rates relevance relevant reliability reliable report research results retrieval retrieving rijsbergen rvelin same savoy scale scientist sigir significance size stability statistical substantially sutcliffe tague test testing text that time topic trec trend university upper using voorhees when with work zobel
http://doi.acm.org/10.1145/1076034.1076081	35	Text Classification with Kernels on the Multinomial Manifold	about acapulco accuracy acknowledgments addison advances alberta algorithm algorithms amari american analysis anonymous another applications apply approach artificial assuming asymptotic baeza bahlmann based been berg berkeley bethesda between bollmann burkhardt cambridge canada categorisation categorization chang chapelle chemnitz christensen cikm city cjlin class classification classifiers classify clustering comments comparison composite conclusions conditionally conference consequently constructed contribution cristianini csie dabak dagm data dataset definite denver development diffusion dimensionality directly disadvantage discriminative dissimilarity distance distances divergence duin duke dumais each ecml effective embed embedding employ encodes euclidean european examination example examples exploiting extend features filter fisher foundations framework frontiers functions future gaussian generalized generative geodesic geometric geometrical geometry germany global graepel haasdonk haffner handwriting harmonic haussler heckerman helpful herbrich histogram howard http hyperplane hypertext icml icpr ieee image improvements incorporate inductive inference information intelligence international into intrinsic iwfhr jaakkola jebara joachims johnson journal justified kass kernel kernels keysers kind kluwer knowledge kondor kullback kullbackleibler lafferty lang langford language large leads learning lebanon leibler library libsvm line linear locally losing machine machines main management manifold manifolds manscript many margin mathematical mccallum measure measures methods metric metrics mexico modeling models modern moreno multi multimedia multinomial nagaoka negative netnews neto networks neural newsweeder nips nonlinear obermayer other over paclik pairwise paper particularly pattern pekalska perception platt plugged positive practically press priori probability problem problemspecific proceedings processing product proposed prove proximity quebec recognition reduction references regularized related relations relevant represent representations research ressel retrieval reviewers ribeiro riemannian roweis sahami saul scale schoenberg scholkopf science sdorra semigroups seung shawe shown sigir silva simply smola society space spaces sparsity specific spectral spheres springer standard statistical statistics structure substitution such suitable support svms symposium systems tahoe tangent tasks taylor tenenbaum text thank that their then theoretically theory they this toolkit transactions trick tubingen uncertainty university uses using vancouver vapnik vasconcelos vector verlag ways wesley whistler wiley williamstown with works workshop yang yates
http://doi.acm.org/10.1145/1076034.1076133	76	Measure-based Metasearch	across algorithm algorithms also analyses annual aslam associated average averaged combination combine condorcet conference confidence contains cutoffs data development each eleventh evaluated evidence experimental from fusion gaithersburg government improved information international july know ledge level lists management mean metasearch montague multiple november obtain office pages pennsylvania philadelphia precision precisions press printing proceedings queries query question references reported research results retrieval searches second shaw sigir sign significance standard submitted table test tested text that these trec trecs used using values washington with york
http://doi.acm.org/10.1145/1076034.1076186	128	SPIN: Searching Personal Information Networks	address artifacts august author books brazil copyright data details documents edges email entities events explicitly extracted extraction found from hard held http iitb induced information jects mentions network networks organizations other others owner person personal persons places probabilistic references relations reply represent salvador semistructured sigir soft software some soumen sources subscriptions such textual these through trips wrote
http://doi.acm.org/10.1145/1076034.1076085	38	Linear Discriminant Model for Information Retrieval	advances algorithm allan andb andi appear application approach approximations artificial asian audio based bayes biing boosting boundaries burges cambridge capturing changning chengxiang cherry chin chinese chou cikm classification classifiers clickthrough clustering cohen collins combining comparison computational computer computing conference crammer croft data david dependence dependencies dependency development discriminative document duda editors effective efficient emnlp engines error experiments filtering flannery fletcher fourth franz freund general generative goodman graepel guangyuan guihong harman hart herbrich hidden huang hwang icml ieee information informed intelligence iyer jian jianfeng jiangbo joachims john jones jordan joshua journal juang kernel laboratory lafferty language large learning leek linguistics logistic machine making margin markov merezes methods miao michael microsoft miller minimization minimum model modeling models nallapati nature nips numerical obermayer optimization optimizing order ordinal overview pattern peng perceptron peter phrasal poisson ponte practical pragmatic pranking preferences press probabilistic processing query quirk rank ranking rate recipes recognition regression report research retrieval richard risk robert robertson sample scale schapire scholkopt schwartz scientific search segmentation sentence shapire sigir sigkdd simple singer smola some song sons speech springer stage statistical status stork support syntactically system tech technical techniques term teukolsky text theory things track training tran translation trec tree trees univ university using vapnik vector verlag vetterling walker weighted wiley with word yoav yoram york yuan zhai
http://doi.acm.org/10.1145/1076034.1076167	110	Information Sharing through Rational Links and Viewpoint Retrieval	analysis annual awareness century communities conference conferences development doing dring federated five forum from gurrin harp have information international keogh last liechti mcdonald orting overview pages press proceedings quarter references research retrieval sharing siggroup sigir smeaton supp twenty watt what years
http://doi.acm.org/10.1145/1076034.1076041	3	Better than the Real Thing? Iterative Pseudo-Query Processing using Cluster-Based Language Models	aalbersberg access accuracy alan allan american amit analysis applied approach automatic bandhakavi based book bruce buckley building callan center chengxiang chirag chris cikm claire classification cluster clustering collection conference connell corpus croft cronen current daniel david deguzman description detection difficult distributed document documentation documents donna editor editors edward eighth engineering engines error european expansion experiments fail fang federation feedback feng filtering fisher framework from gerard giridhar griffiths hall hard harman harper hema high human hyperlinks ifcs ijsbrand improving incremental induced information inquery intelligent interdocument international invited james jamie jasis jinxi john jones joseph journal karen kaufmann kenney kluwer know kumaran kurland lafferty laflamme lalmas language lavrenko ledge lemur likelihood lillian links luckhurst mandar margaret massachusetts maximum methods minimization mitra mixture model modeling models morgan mounia ninth nist nrrc number ogilvie optimal oren overview pagerank pages paper paul peter pollard ponte poster prentice probabilistic proceedings processing pseudo publication query raghavan ranking ratio readings references relevance reliable report reprinted retrieval review risk robertson rocchio ruthven salton science selective series shah sigir similarity sindhura singhal smart smoothing soboroff societies society sparck special stage stephen steve steven structural structure study survey system systems technical technology tenth test text thomas toolkit topic topics townsend track tracking trec twelfth umass university using veera victor willett without workshop xiao xiaoyan xiaoyong zhai zhou
http://doi.acm.org/10.1145/1076034.1076152	95	Evaluating Semantic Indexing Techniques through Cross-Language Fingerprinting	about also american analysis approach artificial bilingual book brief brown burgess carbonell cards compared context corpora corpus deerwester different discourse document documents dumais each english example experimental experiments explorations frederking from furnas german gives graphics guideline harshman hawking hemingway history hoenkamp illustrate indexing information instead intel italian journal kinds landauer languages latent learning ligence livesay lund material nine novel nvdia operators other paired paradigm parts play procedure processes references removed retrieval roles science semantic sentences site sketch society space specifications split stopwords take technology that their three time translated translations translingual unitary which will with words yang