http://www.informatik.uni-trier.de/~ley/db/conf/sigir/sigir2004.html
SIGIR 2004
http://doi.acm.org/10.1145/1008992.1008993	0	Keynote Address	adar andrea atlantic bell bulletin bush cadiz canada challenges cikm city communications curtis cutrell daniel data david december digital drucker dumais edward environments eric events eytan france freeman fulfilling gavin gcconf gelernter gemmell gordon grand haystack information jancke january juan july kansas karger lifestreams lueder lynn march memex model monthly multimedia mylifebits nesc november personal pins press proceedings raman references retrieval robbins roger sarin seen sigir sigmod stein steven storage store stuff susan system think toronto user vannevar vision wong
http://doi.acm.org/10.1145/1008992.1009004	7	A Formal Study of Information Retrieval Heuristics	academic addison advances amati american analysis applied approach approaches approximations automatic based basis brief buckley classical clustering committee computer conference croft data divergence document documentation ective editors engineering erty evaluation exploring forum from fuhr generation harman hill html http ieee importance impossibility inference information introduction jones journal kleinberg kluwer language letin little logic management mcgill mcgraw measuring methods model modeling models modern network nips nist occurrence overview pages poisson ponte probabilistic proceedings processing publications publishers pubs query randomness references relevance retrieval rijbergen rijsbergen robertson salton science search sept sigir similarity simple singhal smoothing society some space sparck special study systems technical term terms text theorem theoretical theory theuse transactions transformation trec turtle uncertain voorhees walker weighted weighting weights wesley with wong workshop yang zhai zobel
http://doi.acm.org/10.1145/1008992.1009097	85	Aggregated Feature Retrieval for MPEG-7 via Clustering	acknowledged acknowledgments adding aggregated approach around attempts based boldareva chiariglione collection combination combining conclusions conference contains descriptions different directorate document each effective enterprise european feature gaithersburg gratefully have hiemstra hierarchical high http ianeva index informatics information integrate into ireland language last mart meaning method might mpeg news over overview pattaya pisa preparation proceedings programme programmes project proportion query ranked references report required research results retrieval schema semantic shot shots similar similarity smeaton sources stages standards support supported text that this track trec upon useful video visit visual vries westerveld when work yager
http://doi.acm.org/10.1145/1008992.1009034	30	Feature Selection using Linear Classifier Weights: Interaction with Classification Models	aaai accurate advances algorithms anderson andrew april august australia bases bayes bhattacharya bologna brain brank bundled cambridge categorization chakrabarti chang chicago china class classification cold comparison conf conference constructing cortes crediting data databases david decision discriminant distribution dunja engineering event fast feature fields finance foundations francisco frank frayling grobelnik hong icml information international italy janez jason joachims jour july just kamal karger kaufmann kernel kong krauth large lawrence learning linear lkopf mach machine machines mahesh making marc marko mccallum methods mili mining mladeni model models morgan multiclass multiple nata networks neural neurocomputing nigam optimal organization other perceptron physics practical press probabilistic proc proceedings programs projections psych pushpak quinlan rakshit references rennie reprinted research review right rosenblatt rosenfeld ross scale selection september shih shourya siam sindhwani soumen soundalgekar stability storage subrata support sydney text textml theoretic thorsten trees unbalanced using vapnik vector very vikas vldb werner with workshop zard
http://doi.acm.org/10.1145/1008992.1009092	80	Searching Databases for Sematically-related Schemas	abbreviated across algorithm allows application applicationsp approach approaches attribute attributes automatic barbara beneventano bergamaschi bernstein between business castano chang china clifton close coma combination conference consisting correspondences crossworlds cupid data database databases describing diego disparate doan documents domingos dominogos drawn edding eleventh engineering erformance flexible flooding frequent from further garcia generic graph halevy have hawaii heterogeneous hong identifying information integration interfaces international italy ject jects jose journal june know kong lack large larger learning ledge library machine madhavan match matches matching meanings melnik memb molina networks neural numb observe omit ontologies order over pages proceedings query rahm reconciling references results retrieval rome santa schema schemas semantic semint sept sigmod similarity some sources space statistical survey system tend term tested that there tool using versatile very vincini vldb when wide with words world
http://doi.acm.org/10.1145/1008992.1009100	88	A Study of Topic Similarity Measures	agents annual association communications computational conference distributional document experimental factors harman important information linguistics maes measures meeting overload proceedings ranking reduce references sigir similarity study that work
http://doi.acm.org/10.1145/1008992.1009136	124	Semantic Video Classification by Integrating Unlabeled Samples for Classifier Training	able according accuracy accurate achieve adaptive adding additional advantage advantages algorithm also annotation assigned automatic automatically available average based because best between browsing cannot cant cantly capability careful categorized cation certain chang change cient class classes classi classieled clip come components concept concepts conclusions configure consist content contexts correctly corresponds cost databases death defigure demo dence described detection diagnosis dialog directly discovery discriminative distribution distributions documents does dorior dramatically dynamic ective eled eliminated elmagarmid enable ensamble ercentage erences erent ertected escape etween existing extrema finally following formance found foundation framework from functions further future gastroinstinal given global groups hand have high hours html http icantly ieee image important improve included increases incremental indexing initializa initialized integrating interclassi interpret into ject jects jfan known labeled labeling large learn learning lecture limited local machine mains margin matching maximum mccallum medical merging mitchell mixture model modeling models modfor more multi multiple national necessary negative nigam nite note noting novel number online only operations organizing oriented other over paper parameters performance performing ples posteclassi posterior precision pretation probability processing proposed provide ratio reasonably reasons reduces references relationship releclassi relevant representation require required results retrieval salient sample samples samtraumatic science scores semantic sensitive should sigir sigmm signi signiftest sizes solution some splitting starting structure supported surgery take techniques test tested text that their then they this thrun thus time tion towards training trans uncc uncertain under unis unknown unlabeled updated used using vant video when with without work worth
http://doi.acm.org/10.1145/1008992.1009043	37	Learning Phonetic Similarity for Matching Named Entity ∗ Translations and Mining New Translations	abderrazak abney ahuja algorithms allen amman applications arab baghdad bill blair blunkett boucher bryan burjanadze chavez chechnya cheung china cicek daniel david detecting diana donald flows franke hall huang hussein lavrov madrid magnanti mideast mikhail mining musharraf name network nino orlin paisley pakistani party pervez powell prentice putin rahim references rohani rumsfeld saakashvili saddam sergei shevardnadze sistani syria taipei talabani tevzadze theory thompson tommy tony translations unionist unseen vladimir yilmaz
http://doi.acm.org/10.1145/1008992.1009125	113	A Review of Relevance Feedback Experiments at the 2003 ∗ Reliable Information Access (RIA) Workshop.	blind buckland buckley commerce current department documents editors effect engines england expansion fail failure feedback institute montgomery national nist notes november number pills poison preparation proceedings publication query references relevant shef sigir special standards technology terra varying voorhees warren
http://doi.acm.org/10.1145/1008992.1009060	51	Learning Effective Ranking Functions for Newsgroup Search	aboutness access achieve achieved adaptation adaptive adequate after aggregation allan also although amount analyses analysis analyzed analyzing andrew annotation annual appear applying approach approaches assumption attributes author authors automatically based baseline batch been behavior bethesda boundary bridging buckley budapest callan canada carried cause chakrabarti chen chile china cikm collections combination combine combining communities comparison computing conclusion conference confirmed content context copenhagen cottrell create croft data davis denmark detect detected development different discovery distillation document documentation documents does domain dublin each edges effective efficient embley engine enhanced evaluation even evidence expansion expected experiments extracting fact factors fail features feedback filtering find finding fiore focused found fourth framework from fuhr fujita full fulllength function functionbased functions furthermore fusion future fuzzy gaithersburg generic genetic global gordon harman hearst hidden higher homepage hong however human hungary hyperlinks hypermedia important improve improved improvement improving incorporating indicate information integration international ireland item jiang joshi journal justsystem kaszkiel kazai knowledge known kong laguna lalmas larger learn learning left level lewis like linear local logistic louisiana machine machines management marc markov markup maryland master meggs message messages method might minneapolis minnesota mittendorf model models moffat more much multiple nanyang neural newsgroup newsgroups nist object observed ogilvie okapi orleans other over overview page pagerank pages paper partial passage pathak pattern pennsylvania perceived performance philadelphia pittsburgh plaunt play press previous probabilistic problem proceeding proceedings processing programming projects properties proportional proposed pseudorelevance publication quality query rafael ranking reasons recognition record references reflections regression related relationship relevant representation representations research retrieval revisited robertson roelleke role routing sack salton same scale schauble schemes scott search searches seattle second section segmentation semantic serve service seventeenth shaw sigir sigmod significant singapore sixteenth sixth smith source spain special still string structure structured structuring subtopic such suggests support surprised switzerland symposium system systems tags task tasks tawde technological teirnan tenth terms test text that then there these thesis this thread threads topic toronto towards traditional training trec tree twelfth twentieth understanding unexplored unique university used usenet using value vector very visual vogt washington ways website weighting were when which wide wilkinson with within words work world would zhang zhou zobel zurich
http://doi.acm.org/10.1145/1008992.1009073	61	Image Based Gisting in CLIR	barnard basic basics boyesbraem canada categories chandler cognitive conference conjunction data disambiguation edmonton forsyth from gray held human johnson june language learning linguistic london meaning mervis natural objects pictures psychology references rosch routledge semiotics sense technology with word words workshop
http://doi.acm.org/10.1145/1008992.1009117	105	Multiple Sources of Evidence for XML Retrieval	documents duisburg ective evaluation hiemstra http inex informatik information initiative language models pages path references retrieval sigir structured thesis twente university using wilkinson xpath
http://doi.acm.org/10.1145/1008992.1009075	63	Discovery of Aggregate Usage Profiles based on Clustering Information Needs	adaptive aggregate april arti campbell case cial conceptual data developing discovery etzioni evaluation framework glasgow information intel know ledge ligence mining mobasher model nakagawa needs ostensive perkowitz personalization references september sites study thesis towards university usage
http://doi.acm.org/10.1145/1008992.1009141	129	Armadillo: Harvesting Information for the Semantic Web	alexiei anant andrew annotation automated bootstrapping budap chapman ciravegna conf crete daniel david dill dingli eiron esws fabio gibson gruhl guha harvest hungary information intl jagopalan jason jhingran john kanungo learning nadav proc references seeker semantic semtag sridhar stephen tapas tomkins tomlin wilks yorick zien
http://doi.acm.org/10.1145/1008992.1009032	29	Restrictive Clustering and Metaclustering for Self-Organizing Document Collections	addison advanced after agency agrawal algebra algorithm algorithms applied association baeza bagging base basis berry bingo biwer bookmark breiman brien burges business categorization cation cidr classi cluster clustering comb combination combining comparative comparison computer concepts conference construction cvpr data database databases dataset decomp defense discovery document documentation documents dumais efore eiro engineering ensembles ergen ervised ester evaluating fast feature figure foundations framework fred gathering generation gosh graupmann hartigan hasan http icml imdb induced information innovative intelligent international internet jain japan jects journal jrennie kamb kaufmann know knowledge kriegel language large learning ledge lewis linear loss machine machines madison manning mapa mapb mapc masumoto means meta metaclustering metho methods mining modern morgan movie multiple natural neto newsg newsgroups occurence ortal osition pages partitions pattern pedersen predictors press proc proceedings processing recognition references research restrictive results retrieval reuse reuters rijsb robust rules sander schuetze search selection siersdorfer singular sizov society spaces speech springer srikant statistical statistics strehl study supervised supp system systems technical techniques technology text theobald theoretical theory tsvm tutorial underlying using value vapnik vector vision vldb weikum wesley wiley wise with wong workshop yang yates york zimmer
http://doi.acm.org/10.1145/1008992.1009122	110	Information Retrieval Using Hierarchical Dirichlet Processes	approach beal beaulieu berkeley blei cesses conference croft cross deling department development dirichlet ertson gull hanco hiemstra hierarchical information jordan kraaij language okapi pages ponte references research retrieval statistics technical text track trec twenty walker
http://doi.acm.org/10.1145/1008992.1009110	98	Formal Multiple-Bernoulli Models for Language Modeling	annual applied approach bayesian conference constraints croft deling dels details development eighth ertson erty extension form general hiemstra informaion information international know language ledge management metho model note omitted othing pages ponte press proceedings references research results retrieval sigir similar smoothing song space study that tipping zaragoza zhai
http://doi.acm.org/10.1145/1008992.1009077	65	The Effect of Back-Formulating Questions in Question Answering Evaluation	aaai analysis answering appear application based challenge clef directions ects evaluation evalution fukumoto ipsj irex isahara japanese kato language lrec magnini masui metrics multigrade multiple notes ntcir overview paraphrasing performance proceedings project question references relevance sakai sekine spring symposium takahashi tasks their track trec voorhees workshop
http://doi.acm.org/10.1145/1008992.1009131	119	Constructing a Text Corpus for Inexact Duplicate Detection	above addresses after against agreed agreement algorithm also analysis applied april arbitration assessing assessor assessors baseline before bene between calculating candidate carletta cation chance chowdhury cial cikm classi clustering collection combined commissioned compare comparison complete computational conclusions concordances conrad corpus cument cuments customary data decisions degree deployments details detection determine determining distribution document domain done duplicate duplicates duplication dynamic electronic elsevier environment environments erformance evaluation exceed expanded expected exploration fast feature finding follow forms found frieder furthermore garc glassman greater grossman growth have here hypothesis identi identical identify indicators inexact initial inter invited july kappa language learning least level library linguistics machine manasse march mccab meta methodology methods molina natural near need news nition nominally observed occurs olean online optimal order over pages pair pairs parameters presented press principled problem proceedings proved queries query real references reliability replicas replication result resultant resulted resulting results retrieval reviewed round rounds scaled schrib science scientists scope searchers sense serve sets shrivakumar sigir signature signi significance springer statistic statistics strong subsequent summary syntactic table task tasks team teams test than that these this tois total training trials tried turtle uncovered underscoring used users using validate value values various verlag webdb where whether which while working workshop world worth would yields zweig
http://doi.acm.org/10.1145/1008992.1009109	97	Context Sensitive Vocabulary And its Application in Protein Secondary Structure Prediction	accuracy accurate active alignments american analysis application approach association associations bank barton between biochemistry biol biological carb cases categorization cation certain chemical choose chou comparative compute concept concepts conference conformation corresp cpxxai cross data datasets deerwester describ detect discovering discovery document dumais ears ective enzymes erent eriment erties etter evaluation examination example expression expressions fasman ferredoxin formations formerly frequently furnas ganapathira genome gram harshman have help high highest human identi improvement indexing information journal king landauer language largest latent loop machine matrix meaningful measure method methods might multiple ndings novel occur occurrences onding onell order ortant overlap pairs pattern patterns physico pick prediction prop protein proteins reddy references region regular relationships reliable results retrieval rosenfeld rost sander science scores search secondary seetharaman segment semantic semantics sentence sequence sequences sheet shows sigir similar similarly singular sites society some sternb structure structures study supp table technique technologies tend terminal text than that these this those understand validation values vector weisser where whether which whole with word words yang
http://doi.acm.org/10.1145/1008992.1009013	14	Polynomial Filtering in Latent Semantic Indexing for Information Retrieval	absence algebra algorithm algorithms also amer analysis appl applications approximation average averaged axis baltimore based beginning begun behav berry blaisdell bottom brien browne buckley causing challenging chisholm collections comp comprehensive comput computations computing conclusion conclusions conditioned constructor cranfield cullum curves data dataset datasets davis december decomposition decompositions deerwester dept discrete document documents dumais edition effectiveness eigenvalues engines englewood erhel evaluating expect explore external fact feedback field fierro figure formulas frakes from furnas gallopoulos golub guyomarc hall harshman have high hopkins horizontal html http illustrated implementation implementing improving indexing info information initial institute instr intelligent interesting interpolation issue iterative john kolda laboratory lanczos landauer large larger latent leary least levels linear loan lters matlab matrices matrix means medline method methodology methods middle minneapolis minnesota modest much national navg nist note number numer observe october orthogonal over pages patras performance phys plot polynomial precision prentice presented press queries query rank recall references related relevance relevant reliable reorthogonalization report retrieval review ridge saad salton search semantic semi sets siam similar sources space sparse squares stemming street structures such supercomputing symmetric systems table technical term text that this though time transactions trec tsvd umsi understanding union university upper using vector very waltham weighting will willoughby with zeimpekis
http://doi.acm.org/10.1145/1008992.1009119	107	Query-Related Data Extraction of Hidden Web Documents	accuracy accurately also analysed approach architecture archives articles automatic based been buttler callan carried categorization caverlee chen classification coding compared computing concept conference connell contained data database databases defined demonstrate detected detects discovering discovery document documents dynamically each engines etzioni examined experiment experiments extracted extraction extracts from generated georgia given gravano have hidden hierarchy high hill information institute international introduction invisible ipeirotis journal knowledge manually manuals mccraw mcgill measure meng modern modified news number objects obtain order originate over overall pages precision preliminary proposed provide qprober query rahardjo randomly ratio recall references related relevant report respectively results retrieval retrieve retrieved routing sahami salton sample sampled sampling search shown sigir subset sugiura system systems table technical technique techniques technology templates terms text that tois total transactions used user using wang which with york
http://doi.acm.org/10.1145/1008992.1009080	68	Subwebs for Specialized Search	alamitos algorithms applications authoritative authority bharat birmingham category cations chang cohn communities computer conf conference create customized development diego distillation environment erlinked flake francisco from gibson giles glover henzinger hypertext ieee improved improving inferring information international internet january journal kaufmann kleinb kruger lawrence learning link lists machine mccallum melb modi morgan ology ourne pages pennock proc proceedings query raghavan references research retrieval saint search sigir society sources symposium topic
http://doi.acm.org/10.1145/1008992.1009026	24	Cluster-Based Retrieval Using Language Models	academic adaptive agglomerative allan american anton anything approach automatic based bennett broadcast butterworths callan carbonell carnegie carp cikm classification cluster clustering collection comparative comparison computer computing conference cornell cranfield critical croft darpa decades department detection development distributed document documentation doddington effectiveness effectivenss efficiency evaluating evaluation evans examining experiments filtering final forum from gather gillick griffiths hamdouchi hearst here hidden hierarchic hierarchical huettner hypothesis ieee information inormation interactive interdocument international jansen jardine jones journal kluwer kraaij lafferty language lavrenko leek leuski london lowe luckhurst management markov mellon methods miller minka model modeling models more mulbregt news novelty part passage pedersen pilot pittsburgh ponte probabilistic proceedings processing publishers query recent redundancy references relevance report results retrieval review revisited rijsbergen robertson rosenfeld scatter schwartz science searching series sigir similarity smoothing society some sparck specific spitters stage statistical storage stream study system systems technical than thesis tombros tong topic tracking transcription trec trends understanding university using villa volume voorhees walker where willet willett with workshop yamron yang zhai zhang
http://doi.acm.org/10.1145/1008992.1009095	83	Design of an E-Book User Interface and Visualizations to Support Reading for Comprehension	above actual adapt aided american amia analysis annual association book browser browsers carthy centred chains comprehension computing conceptual conclusion conducted conference contents contextual cybernetics derived design develop discovering document each ebook effectiveness enhance evaluate evaluation existing experiment experiments factors fall feature goal hammamet harper have heller help hence http human hyperbolic hypothesis ieee important improve informatics information interface international koychev lamping large lexical lifelines measured medical mushlin narrative navigation next patient perceiving performance philadelphia pirie plaisant possible presenting press proceedings profiling propose publishers purposes reader reading real records references relevance research retrieval satisfaction separately sherwood shneiderman simulated sketched smith snyder software step structures studylight symposium systems tasks techniques tested that their thematic this topic tracking trees tunisia understanding user users using vancouver visualization visualizations visualizing will with within work world would
http://doi.acm.org/10.1145/1008992.1009126	114	Supporting Federated Information Sharing Communities	awareness cation cial communities computer computing conference content designing etzee factors farnham flake giles human identi ieee improved kelly lawrence letin liechti line onsibility organization overview pages participation press proceedings references resp self sigchi siggroup sung systems user
http://doi.acm.org/10.1145/1008992.1009120	108	The Patent Retrieval Task in the Fourth NTCIR Workshop	answering automatic fujii information iwayama kando ntcir overview patent proceedings question references research retrieval summarization takano task text third workshop
http://doi.acm.org/10.1145/1008992.1009114	102	Automatic Recognition of Reading Levels from User Queries	about accuracy adult approach available based branch callan categories chang chissom cikm cjlin clear combines csie derivation different distinguished download engine enlisted evaluates examine excite experiment fishburn formula formulas from further general good grad grading gunning hill http iowa journal kincaid klare level library libsvm machines madsci mcgraw mclaughlin measurement memphis millington model naval navy next observe obtained over pages performance personnel press proceedings queries readability reading references report reported research results rogers science scientific search second separation smog software state station statistical support table technique tennessee that university vector whether writing
http://doi.acm.org/10.1145/1008992.1009029	26	Document Clustering by Concept Factorization	advances aided algorithm algorithms analysis applications artibary atlanta baker based belonging bipartite canada capabilities categorization cation chan cikm classi cluster clustering comparisons computer concept corpus cuts data dates decomp depicted design dhillon dimensionality ding distributional distributions document documents each ectral ergraph factorization fast figure finland following francisco function georgia gong graph icdm ieee image indexing information intel introduction jects jordan karypis kernel kernelized large learning ligence machine machines malik matrix mccallum means mika model modha muller multilevel multiplicative nature negative nement networks neural nonnegative normalized ositions pages partitioning parts pattern performance points proceedings processing programming quadratic ratio ratsch reduction references relaxation respectively retrieval reuters same saul schlag scholkopf segmentation selection seung sigir sigkdd simon sizes spaces sparse subspace supp symbol systems table tamp text toronto trans transactions tsuda using vector version vertex volume weiss with words zien
http://doi.acm.org/10.1145/1008992.1008997	2	Scaling IR-System Evaluation using Term Relevance Sets	abdur algorithms amitay amsterdam andrei annual answer answers application applied april august austrailia automatic based beitzel breadth broder buckley budapest cahan cambridge cance carmel category challenges charles chowdhury chris cient clef clevedron clues clustering collections comparing computer computing conference connor cran craswell crawling crestani cross cyril david dennis development discrete donna driven editor editors eighth einat ellen empirical environments eric evaluating evaluation evolution experiments expressions fabio fagin fazli fetterly first forum fourteenth frank from full garc giles glassman grossman harman hawking hector herscovici high hong hungary imperfect index indexing information institute international isdn janet ject jensen john jork journal judgments june junghoo juru justin know kong kumar language languages large lawrence ledge lexical libraries lists maarek management manasse marc mark measure melbourne methodologies methods miki molina names national netherlands networks nicholas nick ning nist nuray october ordering overview page pages papers patrick patterns paul petruschka philosophy potential press proceeding proceedings processing pruning quality rabia ranking ravi recognition references relations relevance reliable research results retrieval right scale science scienti search searching second shengli siam sigir signi sivakumar sixth smadja soboro soda software soubbotin stability standards steve steven study summary symposium syntactic systems taxonomies technology tenth terabyte test tests text thistlewaite through titles track trec twelfth using volume voorhees wide wiener with without workshop world yael yields yoelle zobel
http://doi.acm.org/10.1145/1008992.1009055	46	Automatic Image Annotation by Using Concept-Sensitive Salient Objects for Image Content Representation	access achieved algorithm analysis annals annotated annotation appearances approach aslandogan automatic barnard based bayes belongie benitez bird blei branard campbell carson cation cbsa chang ciency classi clustering comanicu comp comparison composite computer concept conclusion consist content cross csvt data database discriminant dividing duygulu early eccv ective enable erent erformance etween extensions extraction feature features feedback figueiredo figure forsyth framework france freitas from future garden gomes good greenspan grimson gupta hsieh iccv icml ieee image images independent indexing inferring information instance integrated intl isee jain ject jects jeon john jordan journal jsilovic king knowledge krishnan kullback lavrenko learning leibler level lexicon libraries library linguistic lyon machine machines malik manmatha maron matching mathematical mclachlan mean media medianet meer misrm mixture model modeling models more mori most multi multilevel multimedia multimodal multiple natural navigation network networks neural nips nite noting novel numb oliva onents optimal organization outdoor pami paper perceptual performance picture pictures point precision proc processing proposed quantizing querying ratan recall recognition references region relevance relevant representation research results retrieval rishe robust rogowitz sailing salient santini scene scenes segmentation semantic semantics sensitive shift sigir sigmm simplicity smeulders smith sons space spatial spie statistical statistics structural subspace sychay system systems takahashi technique template templates that their this thomas torralba toward training trans transformation translation troscianko understanding user using vailaya vector versus very video view vision visual vocabulary wang where wiederhold wiley with word wordnet words works workshop worring worth yang years york zhang
http://doi.acm.org/10.1145/1008992.1009104	92	The NRRC Reliable Information Access (RIA) Workshop	annual april buckley conference croft cronen development eighth from harman have information international learned loquium pages performance predicting proceedings query references research retrieval sigir text townsend track trec walz what zhou
http://doi.acm.org/10.1145/1008992.1009018	18	Tuning Before Feedback: Combining Ranking Discovery and Blind Feedback for Robust Retrieval	abstraction adaptation affecting algorithms american analyses annelise annual applications approach approaches arbor artificial assumptions automatic banzhaf bartell based beaulieu belew belkin bigi buckley cambridge carpineto chung cikm cliffs combination communications comparison computer computers conference consumer context cottrell data decision development discovery document documentation editor effective effects empirical engineering englewood enhanced evidence evolution expansion experiments exploring factors feedback fifteenth fireder fitness forum foundations fourth framework francisco francone fuhr function functions fusion gatford generic genetic gerard gordon grossman hall hancock harman hawaii hicss holland ieee improving indexing inductive inferring information ingwersen international introduction jersey jones journal june kaufmann keller khan khor knowledge koza langdon lasvegas learning linear logistic lundquist management mark matching means method michigan model moffat morgan mori multiple natural needs nevada nicholas nist nordin okapi optimization pathak pejtersen performance peter pfeifer poli prentice press probabilistic probability proc proceedings processing profiling programming programs publication publishers query ranked ranking references regression relevance research retrieval revisited robertson rocchio romano salton science sciences scores search selection seventeenth sigir similarity smart society some space special specific springer study support system systems technology term text theoretic trans transactions trec twentieth unified university using vector velag vogt walker weighting with york zobel
http://doi.acm.org/10.1145/1008992.1009065	55	Query Based Event Extraction along a Timeline	accurate allan annotation annual artificial association barzilay beth coincidence columbia computational conference context development document dunning elhadad ferro fusion generation george guidelines gupta inderjeet inferring information intelligence internati james journal kathleen khandelwal linguistics lisa mani mckeown meeting methods michael mitre multi multidocument mutlidocument news noemie onal ordering pages paraphrasing proceedings rahul references regina report research retrieval sentence sigir statistics strategies summaries summarization sundheim surprise technical temporal thesis tides topics university version vikas wilson
http://doi.acm.org/10.1145/1008992.1009133	121	Automatic Sense Disambiguation for Acronyms	abbreviation acronym acronyms advances annual approach association based burges christiane christopher computational computing database david disambiguation editor editors electronic entropy fellbaum fraser joachims july kernel large learning lexical linguistics making manuel maximum medical meeting methods normalization pakhomov philadelphia practical press proceedings references rivaling scale scholkopf school science semi sense serguei simon smola supervised support texts thesis thorsten university unsupervised vector word wordnet yarowsky zahariev
http://doi.acm.org/10.1145/1008992.1009107	95	Information Extraction Using Two-Phase Pattern Discovery	arasu around baumgartner built cation classi communications complete computer concepts containing correct cowie credit data dbworld discovery documents ected ective ectiveness engineering enhancing experiment extracted extracting extraction flesca from garcia given gottlob half identical information kamb kaufmann lehnert lixto mails measure mining molina morgan order pages partial pattern pslnl randomly references regions sample school science selected semistructured shepherd should sigmod south still structured synopses system techniques tested text that thesis university unstructured using visual vldb waim wales weight wise with would zhang
http://doi.acm.org/10.1145/1008992.1009112	100	Information Retrieval for Language Tutoring: An Overview of the REAP Project	acquisition adaptively adult adults although analysis annual applied approach aspects associated based berg bikel boston both broder callan captures capturing chakrabarti child children classification coherence collins complex comprehension conference constraints constructs content crawling croft crude current curriculum delos difficulty digital directory discourse discovery dublin each evaluation evolve example experiments fact feedback fiction finder finn focus focused foltz forum general goals google grade grammar high histogram http hypothesis including incrementally information interests international joint kids kintsch kushmerick landauer language latent lavrenko learning lemur level lexical libraries long make matching methods miller model modeling models more naacl name natural nist novelty nymbol ogilvie orleans performance personalisation plan predicting proc proceedings processes processing profiles progress prototype provide publication read reader reading reap recommender references relevance resource retrieval schwartz search semantic series sigir simple smyth special specific student studies such system systems taxonomy teens test text texts that they this thompson three through toolkit topic topics trec updated user using vary vocabulary weischedel well wide will with word wordfrequency workshop world year
http://doi.acm.org/10.1145/1008992.1009083	71	A Two-stage Mixture Model for Pseudo Feedback	associated average based cadez cation cikm classi clustering conference data discovery divergence document editors evaluate experiments federation feedback framework gaffney general harman html http individuals information international knowledge lafferty language management meeting minimization mining mixture model models nist objects pages performance precision press probabilistic proceedings pseudo publications pubs queries query references retrieval risk sept sigir smyth societies special tenth test text trec voorhees with zhai
http://doi.acm.org/10.1145/1008992.1009001	5	Forming Test Collections with No System Pooling	accessed accuracy agreement allan also altman analysis annual appear assessing assessment automatic average based baseline bases bates been being between bland boolean bpref british buckley building cahan cambridge categorization cieri clarke clinical clustered collection collections combination combining communication communications comparing complete computer conference consideration construction cormack corpora correlation correlations crosslanguage degraded depending depends described design designed detection determining development different distinguish distributions document documentation documents dunlop each editors effective effectiveness efficient eguchi employ encyclopedia engine engines establishing evaluation event examine experimenters experimenting experiments extended facto feng filtering first formation formed from further future gaithersburg garofolo gilbert graff greater harman harmandas have here high highly however html http hypertext ideal image incomplete indexing indicate ineffective information intention international introduced john johnson jones journal judgments kando kendall kluwer kotz kuriyama laboratory lancet large lewis liberman library likely links manmatha martey mean means measure measurement measuring methods modeling more most multiple narrow need next nicholas nist nozue ntcir number order organization other outputs overview page palmer panel perfect performance personal phrasal pointed pooling pools pptest precision presented proceedings produce produced progress provides provision publication qrel qrels quantity range rank ranking rankings rath recently references relevance relevant reliable rennert report representations research results retrieval rijsbergen robertson runs salton same sample sanderson scale sciences score search searchday searchenginewatch searches sets shaw sheridan sigir single situations small smaller soboroff some sons special specifically speech spoken standard stanford statistical step strassel stuart study such sullivan system systems task test testing tests text than that this topic track tracking trec trecs true uble university used users using values variants variations varies very voorhees watch wechsler well when where which wide wiley will with without words work workshop zobel
http://doi.acm.org/10.1145/1008992.1009078	66	Effect of Varying Number of Documents in Blind Feedback	access annual buckley conference development harman information international nrrc press proceedings references reliable research retrieval sheffield sigir workshop
http://doi.acm.org/10.1145/1008992.1009061	52	Language-specific Models in Multilingual Topic Tracking	academic adaptive allan applications based college cross detection dissertation dlrg event filter filtering glue http information introduction kluwer language maryland monolingual oard organization papers park publishers references space text thesis topic tracking university vector
http://doi.acm.org/10.1145/1008992.1009058	49	Human versus Machine in the Topic Distillation Task	achieved acsys algorithms analyzing annual answer answering approaches april assessor associated august authoritative automatic average background based batch belkin between bharat borlund brajnik braque bring brooks case category chakrabarti chan chen classification cluster clustering collection compilation completion components computing cone conference content context cool coverage craswell csiro cutrell depth descriptions design development developments difference direction discrete distillation document documentation documents domain dumais each engine enhanced enough environment environments evaluating evaluation evaluations evidence experiment experimental experiments exploring factors finding first focused fuller future gaithersburg gather gibson give hard harman hawking hearst henzinger hersh hierarchical hierarchy higher however human hyperlink hyperlinked hyperlinks hypothesis improved indicate information interaction interactive interface interfaces international joshi journal judgment karadi kelly kleinberg korfhage kraemer large level linear list lists louisiana management marchetti markup mclean measured mizzaro muresan nearly needs neutral nist november oddy olsen olson opposite optimizing order orleans overlap overview part pedersen perception performance performed post price proc proceeding proceedings processing quality question questionnaire raghavan rajagopalan range reexamining references regarding relevance relevant research resource results retrieval retrieved rutgers sacherek same scatter score scores search searcher searches sept seventh show showing shows siam sigir significant sochats sources specifying spring structure study subjects such summary support supposed switzerland symposium systems table tags task tasso tawde tendency tends text than that theory there this thistlewaite thoms tool topic track tracks trec trystan turpin upstill usefulness user users using veerasamy vibe viewing visualization voorhees washington well while wide wilkinson williams work world zurich
http://doi.acm.org/10.1145/1008992.1009064	54	Evaluation of an Extraction-Based Approach to Answering Definitional Questions	aaai accurately achieved advancing algorithm also analysis another answering answers appear appositives approach area automatic available based because become between biographical carry chapter close components conclusively copulas correlation cost created currently definitional demonstrate described design directions editor effective empirical employs evaluation evaluations extracted extraction feature features finally found future grams human hybrid important improve improved information interests larger less licuanan linguistic maybury method model more most observed outputs overview patterns performance performed potential press principled procedure proceedings profiles propositions question questions ranking redundancy references relations removal roadblock rouge scale scores scoring shared significantly small state strategy structured study subjective sufficient suggesting system technology that this tools track training trec types used useful user using valuable value very voorhees weischedel when work
http://doi.acm.org/10.1145/1008992.1009039	34	An Effective Approach to Document Retrieval via Utilizing WordNet and Recognizing Phrases	aboutness academic adapted addison advanced algorithm algorithms amati ambiguity analysis applications approach artificial automatic baeza banerjee bape based brill broadcoverage buckley cardie carpineto chugur cigarran clarke coling colloquium comparison computational concept cone conference copyright cormack cornell craswell cream croft database deng department dictionaries dinstl disambiguate disambiguation divergence document efficient electronic engineering eric evaluation expanding expansion experiments extended fagan fang feature feedback fellbaum framework francisco frieder from fujita genomics global gloss gonzalo grossman grunfeld hard hawk hawkcras hawking heuristics honma improve improving indexing information intelligence intelligent international irsg issue jasis joint jones journal justsystem kaufmann keenbow kluwer knowledge krov krovetz kwok language learning lesk lewis lexical lexicography line linguistics local lynam machine mano mbsc measure measuring meng methods miha mihalcea mihamold mill miller mitra model models modern moldovan morgan multitext narita natural neto oakes ogawa okapi omnh optimization overlaps overview parser pattern pedersen penn pennsylvania performance phrase phrases pine pircs press principar principle principles probabilistic processing programs publishers queries query quin quinlan randomness readable references reflections relatedness relations relevance retrieval revisited riao ribeiro rich richardson rijsbergen robertson robust robustness romano ross salton sander sanderson search selection semantic sense senses shang sigdoc sigir singhal smeaton sparck special specific statistical stokoe structured structuring sumio synsets syntactic tagger tait task tell term terms terra text thesis tois topic track trec treebank turtle uiuc university usage using verdejo voor voorhees walker weighting weights wesley with word wordnet workshop xucro yates yeung zhai zhaitao
http://doi.acm.org/10.1145/1008992.1009127	115	The Effect of Document Retrieval Quality on Factoid Question Answering Performance	above accuracy actual adhoc algorithms amsterdam analysis answer answerextraction answering answers applied approaches aslam behavior better between buckley callan cant carbonell cikm clarit clarke clear collins combination combinations component components conclusions conclusive condorcet cormack corresponding coverage currently czuba december demonstrated discussion dissertation document documents does duggan effectiveness erts especially evaluation evans exact exception experiment experiments extraction fernandes found from fusion given gures have hiyakumoto huang illustrate improved improvement improves included increasing information input javelin katz kemkes laszlo lita lynam management marton mean methods mitamura mitra montague monz multiple multitext murtagh nevertheless nist note nyberg obtained optimal oracle overall passage passages passed pedro perform performance phase preference presented press proc processing provided quantitative question rank reciprocal references results retrieval runs sabir salton searches selection seven shaw show sigir signi singhal smart specialized state statistical svoboda system systems table tellex tequesta terra than that there these this thompson tilker track trec typically umass universiteit used using voorhees were when which with within
http://doi.acm.org/10.1145/1008992.1009105	93	On Evaluating Web Search With Very Few Relevant Documents	august buckley chris craswell david ellen eriment error finland gaithersburg hawking mingfang nick overview proceedings references retrieval ross sigir size tamp topic track trec voorhees wilkinson
http://doi.acm.org/10.1145/1008992.1009035	31	Web-page Classification through Summarization	aaai abstracts adaptation algebra algorithm american analysis annual approach april association athens attardi automated automatic automatically based bayes berger berry bled bouchon brien bringing browsing building butterworth buyukkokten cambridge canada categorization categorizing chakrabarti chen china chris chuang cikm classification classifying coling comparative comparison computational computing conference context cortes creation data deerwester delort describing development devices discourse document dumais eacl ecml eliminating england enhanced european event extracting extraction feature features flake foltz from function furnas gaetano garcia gavin generalized generic glover gong greece gulli handheld harshman hill hong html http hutchison hyperlinks hypertext icml ieee importance indexing indyk inference information instance intelligence intelligent international introduction joachims journal jplatt june kalita kolcz kong kupiec laham landauer lanzarone latent learning linear linguistics link literature london louisiana luhn machine machines many martin mccallum mcgraw measure meeting metamodel meunier micro minimal mining mitchell mittal model models moens molina naive nature networks nigam noisy numerical object ocelot october optimization order orleans paepcke page pages park parts pattern pedersen porter porterstemmer poster prabakarmurthi press proc proceedings processes recipes references relevance relevant research results retrieval review rifqi rijsbergen scalable science scientific search sebastiani seeing segments selection semantic sentence sentences sequential sets siam sigir sigmod slovenia society soft springer states statistical stemming structure study summarization summarizer summarizing support surveys system tartarus task technique teufel text textual thai theory thesaurus towards trainable transactions transductive tsioutsiouliklis united university using vapnik vector verlag website whole with workshop yang zhang zhou
http://doi.acm.org/10.1145/1008992.1009096	84	Toward Better Weighting of Anchors	actual anatomy anchor answer average beaulieu best brin budap columns correct craswell despite document ective engine ertson fagin gaithersburg gatford genvl hanco hawking high hungary hypertextual indicate information kumar large length library link maryland mcbryan mccurley nding normalisation normalised novak novemb okapi only orleans overview page pages penalised places proceedings query rank ranks references relative response results robertson runs scale scores search searching severely show sigir site sivakumar subscripts surrogate table taming term text times tomlin tools track trec type used using very walker when wilkinson williamson words workplace wwww
http://doi.acm.org/10.1145/1008992.1009040	35	Web-a-Where: Geotagging Web Content	advanced analysis annual answering applications approach asker august baker based biomedical boosting bukatin cairo canada carpuat chunk classi coling columbia computational computer computing conference conll crane darmstadt dence detecting digital ding disambiguating division ecdl edmonton egypt endent entities entity eriksson erta european evaluation evidence exploiting extracting extraction framework franz from geographic geographical geospatial gravano grounding heights historical hybrid indep information infoxtract july language larsen lecture leidner libraries library linguistics location malouf mapping markov mccurley mcnamee meeting message models munro naacl named names natural navigation ngai normalization notes octob olsson overview pages patrick philadelphia press proc proceedings processing protein question rauch ravin recogniser recognition references research resources results science scop septemb shivakumar sinclair slinerc smith spatial springer srihari sundheim sydney syntax tagger tagging taip taiwan technical technology terms text understanding using vldb wacholder watson webb when whitelaw wide without workshop world yang yorktown zhou
http://doi.acm.org/10.1145/1008992.1009090	78	Learning Patterns to Answer Open Domain Questions on the Web	access agichtein always analysis another answer answering answers approach banko based been better both bottleneck brill cache candidate case central computational concept concerned conf conference containing content demonstrated detection direct distributed driven dumais each engine error example explore feasibility feasible fetching finds from gpse gravano grewal harabagiu harman have identified ieee index information internal june language lawrence learning likely linguistics minutes moldovan more much multiple natural objective only pages pair parallel parallelized part pattern patterns performance preferably probabilistic proceeding proceedings processing proof prototype proximity query question radev real references requests responsiveness retrieval scalability search send sentence server since solution soubbotin specific speech strings study successfully surdeanu system systematic systems tagging tenth text that time training transactions transformation transformations trec voorhees were when which with within workstations would
http://doi.acm.org/10.1145/1008992.1009135	123	Evaluating Content-Based Filters for Image and Video Retrieval*	ahlberg based content coupling current displays dynamic evaluation filters http information interaction lncs nist nlpir press proc projects query references retrieval santini seeking shneiderman smeulders springer starfield state theart tight trec trecvid verlag video visual with worring
http://doi.acm.org/10.1145/1008992.1009118	106	Verifying a Chinese Collection for Text Categorization	amit analysis approximations based buckley categorization chris collections degraded dmitry document effective gerard http info khmelev length measure model normalization poisson probabilistic proc references repetition retr retrieval robertson salton sigir simple singhal some symp teahan text verification walker webgenie weighted william
http://doi.acm.org/10.1145/1008992.1009052	44	Using Bayesian Priors to Combine Classifiers for Adaptive Filtering	aaai algorithm allan american analysis applied approach ault automatic bartell based bayes belew bennett bias bienenstock boosting buckley callan categorization cation chapman classi collection combination combining comparison computation cottrell croft cross development dilemma discriminative document doursat dumais ecml ective edition empirical ertson erty estimation evaluation event experiments feedback filtering geman generative hall hastie hersh hickam horvitz hull hybrid icml incremental indicators inference information informative interactive joachims jones jordan journal langauge language large larkey lavrenko learning leone likelihood logistic ltering lwin machine maritz maximum mccallum meta method methods metrics models multiple naive networks neural nigam nips nist ohsumed pages pedersen pierce prentice probabilistic processing raina ranked references regression relevance reliability research results retrieval rocchio rubinstein schapire schutze science search shen sigir singer singhal smart smoothing society sparck strategies study supp system systems terms test text thresholding thresholds track trec using validation variance vector volume weighting with workshop yang zhai zhang
http://doi.acm.org/10.1145/1008992.1009094	82	Context-Based Question-Answering Evaluation	aaai answering answers chen complex diekema directions evaluation finding gaithersburg harwell implementing liddy lrec march maybury mean nist overview press proceedings programs question questions references report results spring symposium tice track trec trends using voorhees what within workshop yilmazel
http://doi.acm.org/10.1145/1008992.1009016	16	GaP: A Factor Model for Discrete Data	acquisition advances algorithms allocation analysis applications applied august based becker blei cambridge clustering comp conference decomp development dietterich dirichlet discourse document dumais editors endent erty factorization fast foltz ghahramani gong hofmann human ieee indep indexing induction information introduction jordan journal kearns knowledge laham landauer language latent learning like machine matrix methods models negative networks neural nips oint onent osition pages plato press probabilistic problem proc processes processing progress psychological references representation research retrieval review rinen robust semantic seung sigir singular smoothing solla solution study systems theory trans value volume zhai
http://doi.acm.org/10.1145/1008992.1009050	42	A Collaborative Filtering Algorithm and Evaluation Metric that Accurately Model the User Experience	algorithms analysis artificial benefits borchers breese choices collaborative conference dahlen data dead design empirical evaluating filtering fourteenth francisco good heckerman herlocker information intelligence issue jump kadie kaufmann konstan minnesota morgan movielens neighborhoodbased predictive proceedings recommendation references retrieval riedl starting system systems terveen tois transactions uncertainty university user with
http://doi.acm.org/10.1145/1008992.1009142	130	UKSearch - Search with Automatically Acquired Domain Knowledge	acquired adaptable august automatically beijing collections conference documents domain engineering evaluation ieee intel international july know knowledge kruschwitz language ledge ligent natural pages partially proceedings processing references results search structured system systems
http://doi.acm.org/10.1145/1008992.1009138	126	An Implicit System for Predicting Interests	approach author cikm conference copyright detecting european feedback held implicit implicitly information jose july models needs owner proceedings references retrieval rijsbergen ruthven sheffield sigir simulated south study white yorkshire
http://doi.acm.org/10.1145/1008992.1009044	38	Text Classification and Named Entities for New Event Detection	academic accommodate accounting accuracy acknowledgments across alarms algorithm allan also annotation applications attempts author based baseline basic bikel boostexter both brants callan came cantly caputo carb carthy categories categorization center chen classi clustering comp computational conclusions conditioned conference consistent consortium contributed corp corpus cost could creation croft curve curves data database depth detection develop development dexa direction document documents eech elieve entities erent erformance erforming erman etter evaluations event expert expressed extensions factored false farahat figure first form fraction future gain grant hard harding harness have high http human illustrated improvement improvements increased index inference information inquery intelligent international interpretation into jects jman kluwer know krovetz language latter lavrenko lead learning learns ledge like line linguistic lists machine made management material metrics miss model more morphology multi multiple name named ndings necessarily need ninth nist norm novelty numb onell onsor oosting opinions oral organization orted osite pages papka part pass performance presented probability proceed proceedings process processing promising publishers random recommendations references regions representation result retrieval rules runs schapire schwartz should shown sigir sigkdd signi simple singer single situations space spawarsyscen speech stage stokes stop stories story study such summer supp system systems technical technology temp tests text that their these this those time title topic tracking using utility utilizing vector viewing wayne weighted weischedel were what while with work workshop yang zhang
http://doi.acm.org/10.1145/1008992.1009103	91	Evaluation of the Real and Perceived Value of Automatic and Interactive Query Expansion	actual analysis annual anova apache auto beaulieu behavior belkin case context croft dependent documentation each easier effectiveness efthimiadis examining expansion experiments fisher hersh improving independent information informational interaction interactive interfaces jakarta journal koenemann local lucene magennis measures mode none perform performed post potential presented query question references repeated report results retrieval review rijsbergen ruthven scaling search semi showed sigir significantly study subject support systems table technology test that track transactions trec users variable which with within
http://doi.acm.org/10.1145/1008992.1009079	67	Eye-Tracking Analysis of User Behavior in WWW Search	about above abstract abstracts actions again altavista analysis around average bars before below break broder bulletin clicked clicks clickthrough conference data depicts discovery document each effect engines error evaluate explore feedback figure first forum frequency from henzinger hewlett hibikino implicit important individual inferred information instance interesting jaana japan joachims kaski kitakyushu know knowledge kojo laboratories large links list look making many maps marais mining moricz movements number observed only optimizing organizing packard page particularly performance presented proceedings processing psychological query rank ranked rayner reading references relevance report research result results retrieval salogarvi scan search selected selection self september serve seven sigir silverstein spent system taxonomy technical that there third thoroughly time user users using very viewing when with workshop wsom years
http://doi.acm.org/10.1145/1008992.1009048	41	Hourly Analysis of a Very Large Topically Categorized Web Query Log	actual actualpos addition algorithms also although american analysis anchor applications article audio automatic available average beitzel berry best bibliography both broder budapest caching canada categories categorization category certain change changes chowdhury cikm clustering cmis commerce computer conclusions conference constant content continuing correlation coverage craswell critical csiro data degree delivery despite developing development distribution drive driven during eastman editor effective efficient eiron electronic engine engineering engines european evaluation evaluations examining example excite exploring fall february feedback figure finance find finding fluctuation focuses fonseca forum framework frequency from further future game general giles given goodrum great griffiths grossman hallaron hawking henzinger higher highest hong hour hours http ieee image impact implications infocom information interests international internet investigates investigating jansen january jensen journal july june known kong large lawrence lempel level life locality logs longitudinal magnitude management marais markatos mccurley meira might mining moran moricz most moura multimedia names nature needs neto networking next nickc november october online only operational operators optimal orleans over overlap ozmultu ozmutlu pages pair pass past patterns peak pigfish policy pooch popularity populi poster posters potentially predictive prefetching preserving press priori priority proc proceedings processing public pubs queries query raghavan rank ranked ranking rankings real received reformulation relevance remains repeated reports representative requirements research results reuse review riberio ross saracevic saraiva scalable scheme science search searchenginewatch searches searching seattle september service services sets sever sigir silverstein similarly society spink stable stream studies study subject submitted sullivan systems taxonomies taxonomy technology term terms test text that their they this throughout time titles topics toronto total traffic transactions trends twice type understanding user users using varies vastly versus very video volume wang watch which whose wide wolfram work workshop world worth yang zhang ziviani
http://doi.acm.org/10.1145/1008992.1009124	112	A Study of Methods for Normalizing User Ratings in Collaborative Filtering	algorithms analysis annual architecture artificial bergstrom breese cikm collaborative computer conference cooperative decoupled empirical filtering flexible gaussian grouplens heckerman hofmann iacovou icml information intelligence international kadie knowledge latent learning machine management mixture model models netnews open pages predictive preferences proc ratings resnick riedl semantic sigir suchak supported uncertainty with work zhai
http://doi.acm.org/10.1145/1008992.1009068	57	Block-level Link Analysis	aaai algorithm algorithms amento analysis analyzing anatomy annual approach artificial asia associated authoritative authority automatic based bharat block brian brin bringing budapest chakrabarti china citation compilation conference content davison development distillation document documentation documents does effect efthimiadis engine enhanced environment expansion expert exponentiation extracting extraction feedback fred gibson gregory henzinger hill hits hungary hyperlink hyperlinked hyperlinks hypertextual importance improved improving information integrating intelligence international joel joshi journal july kleinberg large learning lempel links linkstructure list markup matrix mean microsoft miller model models modifications moran motwani nature nepotistic object okapi order overview pacific page pagerank pages predicting press proc proceedings projects pseudorelevance quality query raghavan rajagopalan ranking ratings recognizing records references report representation research resource retrieval review robertson salsa scale schaefer search segmentation seventh sigir song sources springer stanford statistical stochastic structure systems tags tawde technical technology terveen text theory topic track trec twelfth university using vapnik vips visionbased visual wide winograd with world york
http://doi.acm.org/10.1145/1008992.1009111	99	User Biased Document Language Modelling	applied approach australia berkley comparative conclusion conference consistently croft development discussion erty experiments finland hofmann improved indexing information infromation jones language latent managament melbourne methods model modeling models orleans pages performance ponte prior probabilistc probabilistic processing references retrieval robertson semantic sigir smoothing sparck stage study tampere using walker zhai
http://doi.acm.org/10.1145/1008992.1008998	3	Using Temporal Profiles of Queries for Precision Prediction	academic allan annual august based bursty callan cikm collins conference constructing croft cronen data development diaz discovery erformance erty feng fisher frank hierarchical http implementations inference information international java jensen jones july kaufmann kleinb kluwer know krovetz labs language larkey lavrenko learning ledge lemur machine management mining modeling models morgan morphology novemb ogilvie oral pages practical predicting press proceedings process publishers queries query references relevance research retrieval sigir sigkdd sixteenth statistical streams strohman structure swan technical techniques temp thompson time timelines timemines toolkit tools townsend truong turtle usage viewing waikato weka with witten word yahoo zhai zhou
http://doi.acm.org/10.1145/1008992.1009082	70	Measuring Pseudo Relevance Feedback & CLIR	affected after ambiguity american assessing automatic ballesteros based being buckley change changes cikm clef clough comparing contradictory contrast croft cross crosslanguage cycle degrading different documents dunlop ecir effectiveness effects evaluation expansion feedback found from higher hundred image improving information interactive iteration journal kldivergence lafferty language mayfield mcnamee measures mira mitra model more only particularly pnorm position proceedings provides purely quality query rank ranking references reflecting reflection reflections relevance relevant resolving resources results retrieval sanderson science score second selecting show shows sigir singhal society somewhat table techniques terms thirty this track translation when zhai
http://doi.acm.org/10.1145/1008992.1009000	4	Retrieval Evaluation with Incomplete Information	algorithms alistair american annual august australia automatic available based bene british bruce buckley butterworths cahan cambridge cance chapter charles chris christopher cient clarke clef cleverdon collection collections computer conference construction cormack cornell cran croft cross cyril determining development documents donna ecial ectiveness edition editors eisenb ellen environments erfect ergen eriment eriments error eval evaluating evaluation fazli fourteenth frei from google gordon harman highly html http ideal index information international january jones journal judgments justin language languages large lecture library management mark measure measurement measuring melb mizzaro need nicholas nist notes numb nuray oratory ourne overview package pages palmer patrick philosophy precision preference press proceedings processing provision publication rabia ranking recall references relevance relevant reliable research reserach results retrieval rijsb rorvig ross scalability scale science search seventh sigir signi simple sixth size smart society sparck stability stefano systems technology test tests text topic trec twenty uble university user variations voorhees what whyuse wilkinson with without workshop wrong york
http://doi.acm.org/10.1145/1008992.1009132	120	Why Current IR Engines Fail	able above access addition address addressed almost america analysis analyzed annual another antarctic antarctica application applied aspect aspects assigned assignment assignments associated attack attempt automatically available been begun being belongs between both buckley cantly categories category cause chris close collection come concept conclusion conference construct constructed contributing could countries crucial cuba cult current declines deliberately despite development different disasters discover discuss discussing distinguish documents does donna during each efforts emphasize emphasized enough episode europe example except expand expansion experiencing explicit explorations export extremely fact factor fail failure failures fashion fertilization finally focus forged from full general give given half harman have help human identify import important improve improvement incidents increase increasing information instances international investigations irrelevant just knowledge language least like likely list long loss main major many methods missed missing more much natural necessary need needed notions nrrc occurred ocean once order other others outside over overall performance placed planned point poor possible possibly present prevent primary problem proceedings producing proximity quantity query quite reached realize reasons references relable related relationship relationships remote represented require required research restrictive results retrieval retrieve retrieving root roughly same sample scienti secondary seem seen semantic semantics sensing ship should show sigir signi some sorted sorts spaceborne spotted steel stem stemming stolen straightforward studied substantial success such sugar suggests summer surprising system systematic systems technical techniques technology tended term terms than that then there these this those though through thus together tools topic topics tourism towards transportation tunnels type understanding understood used using various very vitro want weather well were what which while will with without worked workshop world would
http://doi.acm.org/10.1145/1008992.1009106	94	A Music Recommender Based on Audio Features	algorithms analysis approach audio based breese canada carried classification clustering collaborative conf cook corpus empirical evaluation experimental experiments files filtering genre george heckerman hybrid ieee item kardie karypis konstan mono music perry pieces predictive proc ratings real recommendation recommender references sarwar signals stored system tenth tsap tzanetakis users which with word
http://doi.acm.org/10.1145/1008992.1009025	23	Parsimonious Language Models for Information Retrieval	abstracts academic adaptive advanced advances algorithm alternative american approach approaches automatic average back based bayesian berger bertoldi broadcast butterworths callan called cantly center ciency cikm civr combination communication conference creation croft cross darpa data dempster describ detection development digital disambiguation discussions document dordrecht down ecdl ecial edition editor editors eighth endence entropy entry equation erent erforms ergen ertson erty estimation european evaluating exact examples federico feedback figure foundations framework from gadde general graphs hidden hiemstra hofmann however html http icml image improved incomplete indep indexing information institut intel international jones jong journal keith kluwer know kraaij laird language latent lavrenko learning ledge leek libraries ligent likelihood linear lingual literature ltering luhn machine managemen management manning markov massachusetts maximum method methods miller minimization minka mixtures mizzaro model modeling modelling models multimedia multiple natural news newswire nguyen nist notion novelty oils ortance other overview page pages parsimonious plus ponte precision preface press prior probabilistic probabilities proceedings processing pruning publication publishers query radio ratio recall recent redundancy references relevance research retrieval review rijsb risk routing royal rubin same sankar saracevic schutze schwartz science search second section semantic seventh shows sigir signi sista size smaller society song sparck speech statistical step stolcke story strategies system task technology terms text than thesis thinking third three topic topics track tracking transcription translation translations trec twenty understanding uses using vasconcelos video visual voorhees vries walls weighting weischedel weng westerveld which whole with word workshop worse zaragoza zhai zhang
http://doi.acm.org/10.1145/1008992.1009084	72	Natural Language Processing for Browse Help	actes angleterre annual australia based borlund cambridge college colloquium conference croft cumulated development evaluation first from gain half harman have indicators inen information ingwersen institute interactive learned life measures melbourne moffat national overview performance press proceedings publication ranked references relative relevance research retrieval rijsbergen rvelin sidney sigir special standards sussex systems techniques technology text tois transactions trec what wilkinson york zobel
http://doi.acm.org/10.1145/1008992.1009123	111	Broken Plural Detection for Arabic Information Retrieval	abduelbaset access algorithm analysis anne arabic assessed assessment august ballesteros basic clearly collection comp compared comparing computing confirming connell corpus created data department designing developed digital documents effect effectiveness evaluate evaluated evaluation evens experiment extracted figure finland four from full gaithersburg garside goweder greenstone hani highly http improving incorporated index indexing indicate individual inflected information jasist judgments kharashi khoja lancaster lancs language languages large larkey lerc library light mahmoud martha method methodologies methods natural nist obtained occurrence omari outperforms over parametric particular performance popovic processing proposed queries query rank references referred relevance results retrieval roeck root roots salem search shown sigir sign signed significance significant significantly slovene standard statistical stem stemme stemmer stemming stems substantial such suggest suggests sughaiyer system tailed tampere task terms test testing tests text textual that their they this three trec umass university used users using waikato wilcoxon willett with word words workshops zealand
http://doi.acm.org/10.1145/1008992.1009031	28	Document Clustering via Adaptive Subspace Iteration	academic acknowledgments adaptive advances agent agents aggarwal agglomerative aggregates agrawal aided algebra algorithm algorithms allerton allows almost also alternating analysis anderberg annual applications applied approach aspects associated austin authors autoclass automatic autonomous based bayesian berger berlin beyer bialek biclustering biology bipartite birch bock boley bottleneck buzo categorical categorization cation chain cheeseman cheng church cient cikm classi cluster clustering clusters cofd columns comments communication communications comparative competitive computations computer computing concepts conceptual conclusions conference control coupled cover criterion cure cuts cybernetics data database databases datasets dawak department design dhillon dimension dimensional dimensionality ding discovery discriminant distance document documents domeniconi dual dubes each eigenvectors eighth element elements evaluation experimental experiments explicit exploration expression factorization fast feature fifteenth fischer foundations freeman friedman functions gehrke generalized generation generative ghosh gini globerson goldstein golub gong govaert grant grants graph graphs grateful gray grid gross grouping guha gunopulos hagen hall hard hartigan hastie hastings helpful hierarchical high hopkins http huisinga iccv icdm icdt icml idea identi ieee image inference information insights intelligence intelligent international introduced invariant ismb iterative iteratively jain jieping john johns johnson jordan kahng kamber karypis kato kaufmann kelly knowledge kumar language large learn learning lecture leung linde linear livny loan locally ltering mach machine malik mallela management markov matrix mccallum meaningful meshes method methods metric mining minnesota mobasher modeling models modha molecular moore morgan multi multivariate nature nearest nearly negative neighbor neural nips nishisato nite normalized notes numerical objects ogihara operators opitz optimization paper park part partitioning parts pattern peng pereira perform perona perturbation planar point prediction prentice press probabilistic proc procedures proceedings processing procopiuc projected providing publishers quantization query raghavan ramakrishnan rastogi ratio reduction references reminiscent report results retrieval reversible review reviewers rigoutsos rows scale scaling schutte science section segmentation self seung shaft shim siam sigir sigkdd sigmod simon simultaneous slonim somewhat sons space spaces spectral spielman springer statistical structure study stutz subspace suggested suggestions supported symposium system systems taylor technical techniques teng texas text thank that their theoretic theory third this thomas tibshirani tishby toolkit toronto trans transactions twelfth unifying university updating useful using vector verlag very viable view vision volume want warehousing webace weiss when wichern wide wiley with wolf word words works workshop world york zhang zhao zhong
http://doi.acm.org/10.1145/1008992.1009099	87	Collaborative Filing in a Document Repository	analysis applied assistant bharat butterworths card clustering cooperative data desktop document foraging hierarchical information jobson london manage marais multivariate personal pirolli process psychological recent references retrieval review rijsbergen springer supporting surfing trends uist verlag willett with york
http://doi.acm.org/10.1145/1008992.1009008	10	The Overlap Problem in Content-Oriented XML Retrieval Evaluation	action advantages american analysis applications approach approaches appropriate assessments assistee avignon base based benchmarks bericht blanken bollmann clark components computer conclusions consortium content contribution cooper could critical cumulated current dagstuhl data december derose develop different distinguishes documentation dortmund editors effect effectiveness effort elements employed european evaluating evaluation existing expected explain facilitate framework france from fuhr functions gain general germany govert grabs graded here high however http ideal implementations inex inference informatik information informations initiative instantiation integrating intelligent intend investigation irrelevance issues january jarvelin jasist journal jung kazai kekalainen lalmas language languages length letter lncs main malik measure measures metrics misses model modeling models near nition ordering ordinateur oriented overcomes overlap overlapping paper partial path pointed possible precision prede probabilistic problem problems proceedings proposals provided raghavan rather recall recherche recommendation reference references relevance report research result results retrieval retrieved reviewed rewarded riao satisfaction schek schenkel science search second should side single skewed society some springer successes such suitable systems technical techniques technischer technology that these this through tois tolerance transactions trivial uniduisburg unit university user users using value version volume vries weak weikum with within without wong workshop xpath year
http://doi.acm.org/10.1145/1008992.1009101	89	Effectiveness of Web Page Classification on Finding List Answers	answering banko brill conference data dumais francisco intensive kaufmann learning machine morgan notebook overview proceedings programs question quinlan references retrieval tenth text track trec twelfth voorhees
http://doi.acm.org/10.1145/1008992.1009072	60	A Hybrid Statistical/Linguistic model for Generating News Story Gists	analysis applications assessing automatic carthy chain cicling cohesion college computer conjunction corpus dept detection domain doran dublin dunnion evaluation extraction factorial gram hovy html http impact informal jones lexical methods naacl nist nlpir occurrence pilot proceedings projects quinlan references rulequest schemes science scoring sentence sigir spark speech statistics stokes study summaries summarisation summarization summary tests text thesis topic tracking tutorial university unix using with workshop
http://doi.acm.org/10.1145/1008992.1009010	12	Configurable Indexing and Ranking for XML Information Retrieval	adding allows also amer american appropriate aspx assessment avoid baeza bidirectional canada carmel compact conclusion configurable configure content ctree data databases dataguides development document documents edition enables enabling evaluation fall false fargo fernandez flexible formulation forum fragments fuhr generating goldman grabs grossjohann guide guidelines hill http index indexing inex inexdemo inexsearch information initiative introduction issue journal kazai lalmas language level maarek malik mandelbrod mass matching mcgill mcgraw meuss modern negatives number optimization orleans outlook paper phrase piwowarski proceedings processing proper proposed qmir qmul query querying ranking references relevance report retrieval salton schek schlieder science searching second select semistructured sigir society soffer spaces srivastava system tags technical technology thefly theobald this topic toronto tree types ucla users vector vldb volume webdb weikum which widom workshop xirql xpath yahia yates york
http://doi.acm.org/10.1145/1008992.1009121	109	Measuring Ineffectiveness	appear buckley chris ellen error evaluating evaluation experiment measure overview pages proceedings references retrieval robust sigir size stability topic track trec voorhees
http://doi.acm.org/10.1145/1008992.1009027	25	Corpus Structure, Language Models, and Ad Hoc Information Retrieval	allan allocation analysis andrew applied approach audio based bayesian blei brown bruce butterworths cache callan cation chengxiang class classi cluster computational computer conference connel croft data david della dependence desouza dirichlet distance distributed djoerd document dyadic dynamic edition editors erty european experiments extension fang feng fisher from gather gram hearst hiemstra hofmann hugo hypothesis icsi ieee importance information inquery institute international iyer james jamie jennifer jinxi john jordan journal kluwer language latent lavrenko learning lemur linguistics long machine mari marti mercer methods michael minimization mixture mixtures model modeling models natural ninth ogilvie optimal ostendorf pages paul pedersen peter pietra ponte probabilistic proceedings processing puzicha query reexamining references relevance report research results retrieval rijsbergen risk robert rukmini scatter science searching semantic sigir smoothing speci speech study systems technical tenth term text thomas tipping toolkit topic transactions trec unsupervised using victor vincent zaragoza zhai
http://doi.acm.org/10.1145/1008992.1009087	75	Evaluation of Filtering Current News Search Results	allan brin chadrasekar chang cikm cost critical data document documents examination extracting feng filtering from function generated glean henzinger information manage manmatha milch news process queryfree references search sigir srinivas syntactic template unstructured using
http://doi.acm.org/10.1145/1008992.1009116	104	Refining Term Weights of Documents Using Term Dependencies	addison agrawal anne applying approach arun association associations baeza berthier between bruce building chang chen classification concept conference croft databases development documents extracting hierarchy huang imielinski information international internet items janming knowledge language large meng ming minig mining modeling modern nanas neto nikolaos ponte proc profile rakesh references representation research retrieval ribeiro ricardo roeck rules semantic sets sheng shian shih sigir sigmod swami term tomasz uren user victoria washington wesley with yates yueh
http://doi.acm.org/10.1145/1008992.1009022	21	Using the Web for Automated Translation Extraction in Cross-Language Information Retrieval	able academic airlines akira annan anti ariel arnold blaster canal cappuccino card cards carlsb carter castro cation cell cells china chinese chiutou clijsters clone cloning commerce commercial company contactless corp credit cross crown daya dingkang dioxin djia donald dongsha dullah electronic embryonic english enron environmental espresso excellence extracted fighter finding forest found france frontieres gage gene genetic ghter given governor gump haihe hezb hormone howard huaihe hussein international islands john johnnie jordan joyner kazuhiro keeping keizo khatami kitano kurosawa kuwaiti laden lancome land liaohe likud logitech lord lyonnais mariners mars martha masako matrix medecins michael mohammad mortor motors murdo nanotechnology nasd nemo nina nino nintendo nissan ntcir number obuchi ollah osama panama party peace pearl peng pinatub pricewaterhouseco princess program proliferation promoting provincial qinshan query references reloaded renault renualt reuni richter rings river rover rumsfeld saddam sans sars sasaki schwarzenegger seattle sharon sino smart songhua space spratly starbucks start station stealth stem stewart strait subic table tafe takeshi terms tokyo topic torrijos tour transaction translation translations treatment treaty trec universities viagra vietnamese viewsonic walker xisha
http://doi.acm.org/10.1145/1008992.1009085	73	Triangulation without Translation	academic achievements across addressing advances advantageous airio along annual appear appears appendices arabic asterisk august automatic average ballesteros banerjee because best between bilingual blind braschler buckley campaign cardie careful center challenges character cikm clef clir clustering combination common comparison complicated conference conjecture conversely core costello croft cross description development dictionary difference direct discussion dissimilar diversity drop effective european evaluation exhibited expansion experimental experiments extend filtering flukes followed forum from given gives gollins gram harman harper hiemstra improving information inserting intelligent international kluwer knowledge kraft lack language languages largest lehtokangas level likely listed little lower management mayfield mcnamee mean mitra models monolingual more morphologically netherlands nist norway notes objectives observations obtained omitted original other outperformed pair pairs passed pathways performance peters piatko pivot point precision proceedings processing publication publishers queries query range ranging recent references reflect reported research resources results retrieval romagnoli runs rvelin sanderson score scores seen selection serve shown sigir significant similar similarity sixth smart source special stabilize strongest such suggest superconcepts system table target technology telematics tenth terms test tested text that them there therefore these thesis this those thus title tokenization transitive translation trec triangulated trondheim twelfth unaltered union used using vallin video volume voorhees walz western whatsoever which whose wilcoxon with within working zobel
http://doi.acm.org/10.1145/1008992.1009014	15	On Scaling Latent Semantic Indexing for Large Peer-to-Peer Systems ∗	accrue adaptable addressable algebra algorithm american analysis applications applied available baltimore based bawa berkhin berry bingham blott brien browne buckley callan categorization centroids cient cikm cluster clustering code collection combinatorial comp computations computer computing concept conference connell content cornell croft data database decomposition decompositions deerwester department developments dhillon dimensional dimensionality ding discovery disruptive distributed document drmac dumais dwarkadas edition editor engines enhanced enterprise environments expanded experiments fast feasibility february first francis frankl furnas garc gatford gloss golub graphs gravano grid hancockbeaulieu handley harshman hashtable hellerstein high hinterberger hopkins hotnets http husbands image implementation indexing industrial industry info infocom information innovation intelligent internet iptps ithaca jason jeon jessup johnson jones jose journal kaashoek kalogeraki karger karp karypis kolda kubiatowicz landauer language large larkey latent learning leary least lemma length letsche lindenstrauss linear loan location lower ltering lukose machine maehara mahalingam manku mannila maryland mathematical mathematics matrices matrix mcarthur merging methods milojicic mining mitra model modeling models modha molina morris multikey nagaraja nathan netlib network networks nievergelt nist normalization nsdi october okapi older organized organizing over overlay overlays papadimitriou park partially patents peer peersearch performance pivoted podc press prete probabilistic proc projection pruyne psearch quantitative raghavan random ratnasamy redmond reduction references reinsel report representation results retrieval review rhea richard robertson rollins rosen routing salton scalable scale schek science sciences search second segmentation selection self semantic semidiscrete sets sevcik shenker siam sigcomm sigir sigkdd similarity simon singhal singular smart society software some source space spaces sparse sphericity squares storage structure structured study survey svdpack symmetric system systems tamaki tang technical techniques text theory third tomasic tools topic topically topics trans transactions trec understanding university using value vector vempala version villars vldb walker weber with wong workshop yang
http://doi.acm.org/10.1145/1008992.1009140	128	ACES: A Contextual Engine for Search	author based cadiz canada copyright cutrell document dumais engine exploiting held history information interactive jancke july kingdom owner personal query ranking references retrieval robbins sarin search seen session shef shen sigir south sriram stuff system toronto united yorkshire zhai
http://doi.acm.org/10.1145/1008992.1009069	58	Usefulness of Hyperlink Structure for Query-Biased Topic Distillation	academic advances agents algorithms amati amitay analysis anatomy anchor annual appendix application approaches approximations authoritative based below bergen bharat brin cacheda carmel carpineto cation center chapman cikm classi cohesiveness collection combining computer conference craswell croft cronen darlow density development distillation distribution divergence document duda dynamic editor effective eleventh employed engine entropy entry environment estimation european evaluation everson experiments filtering fisher formulae framework freq frequency from gaithersburg glasgow gurrin hall hart hawking henzinger herscovici hiemstra hyperlink hyperlinked hypertextual ieee importance improved improving information intelligent international isdn john journal jown juru kang kleinberg kluwer knowledge kraaij kullback large lempel link links london management mean measures measuring model models nding networks nist normalization ounis overview page pages parameter pattern performance plachouras poisson predicting press prior probabilistic probabilities proceedings publication query randomness references relevance research retrieval rijsant rijsbergen robertson romano scale scene schemes scope search sensitive shannon shown sigir silverman simple site size smeaton soffer some sons sources special springer statistics study systems tenth term text theory topic townsend track transactions trec tuning twelfth type university useful using vari variance verlag walker weighted weighting westerveld when where which wiley wilkinson with within york zhou
http://doi.acm.org/10.1145/1008992.1009091	79	Email is a Stage: Discovering People Roles from Email Archives	acts berkeley blanton bush cambridge cation classi computer conference destroy editor emaildetails england essay from house html http icml inference info information international internet joachims language learning lyman machine machines mail messages much philosophy press proceedings projects reagan references research retrieved searle secret sims speech support text transductive tried university using varian vector white york
http://doi.acm.org/10.1145/1008992.1009046	39	Assigning Identifiers to Documents to Enhance the Clustering Property of Fulltext Indexes	aligned annual appear approach australasian based binary blandford blelloch browsing buckley chakrabarti cluster codes codewords collections compression computer conference cornell cutting data database department development discovering document dunedin editor editors francisco from gather hypertext ieee implementation index information international inverted karger kaufmann know large ledge mining morgan pages pedersen press proc proceedings publishers references reordering report research retrieval scatter schewe science sigir smart system technical through tukey university using williams word zealand
http://doi.acm.org/10.1145/1008992.1009129	117	Context-based Methods for Text Categorisation	academic algorithms based categorization cation collections compression conf dumais harper heckerman inductive inform information khmelev kluwer know lang language learning management measure modeling models pages platt proceedings publishers references repetition representations retrieval sahami sigir teahan text using veri
http://doi.acm.org/10.1145/1008992.1009042	36	Focused Named Entity Recognition Using Machine Learning	aaai abstracting access advances algorithm algorithms analysis anlp applied approach approaches association automated automatic barzilay based bayes categorization cation chains chen chunking cikm classi columbia comparison comprehensive computational computer computing concepts conference conll coreference croft cucs damerau data decision discourse discovery docs document domain dual eacl editors edmundson elhadad enchmark endence endent entities entity erent evaluation event exact extraction features finding focus focused formulation freitas from function generalization goetz grishman hierarchical highly hovy huang icml identi identifying indep induction indurkhya information integrated intel international introduction isahara ists johnson jones journal kaestner kaufmann kupiec language lawrie learning lexical ligent linear linguistics loss lrec machine machinery management mani marcu matching maybury mccallum mckeown methods meulder models moens morgan naive named natural neto news nievola nigam nobata noun oles olic ortant osition pages paice pattern pedersen performance phrases pkdd press proceedings processing programs query quinlan recognition references regularized research resolution resources result retrieval rosenb rule sang santos scalable science sekine selection sentence septemb shared sigir soon stories string structured structures summaries summarist summarization summarizer summarizing symb system systems table tagging task technical test teufel text textual third through time topic topics trainable training tree university using weiss winnow with words workshop yang zhang
http://doi.acm.org/10.1145/1008992.1009074	62	Classifying Racist Texts Using A Support Vector Machine	analysis association between bigrams classification colloquium compared conference consistently content corpus domain dropped europaea european figures filtering finn france genre glasgow greevy http icra improved increased information internet joachims kushmerick language lechleiter linguistic linguistica lyon open precision proceedings racism rating reaching recall references representation research resulted retrieval september size slightly smyth societas svmlight training transfer whereas while
http://doi.acm.org/10.1145/1008992.1009070	59	Block-based Web Search	algorithm annual asia automatic bailey based boundary callan chakrabarti china cmis collection conference connectivity content craswell crivellari csiro data development discovery distillation document documents dublin embley engineering enhanced evidence experiments expository extracting hawking hearst homepage http hyperlinks information international jiang joshi level link management markup meeting melucci microsoft multi ninth pacific page pages paragraph passage philadelphia press proc proceedings processing purpose record references report representation research retrieval segmentation seventeenth sigir sigmod structure tags tawde technical test text topic tracks trec trecweb using vips visionbased visual weighting
http://doi.acm.org/10.1145/1008992.1009102	90	Detection and Translation of OOV Terms Prior to Query Time	accuracy additional anchor arguably asian automated bilingual building candidate cedict chen chien chinese computer conclusion conference contained correct cross development dict dictionaries docs documents eaker ectiveness english erson european evaluate extracted extraction found fourth from further given glasgow however html http improve information international investigating jects khudanpur language levow location loquium mandarin mandarintools mcewan meng mining native oard organization ounis pages parallel proc proceedings processing queries references related research results retrieval ruthven schone scotland shown sigir some speech strictly such table tang terms text that this title transactions translation translations translingual tung twenty upenn used using vines wang were when with zhang
http://doi.acm.org/10.1145/1008992.1009115	103	A Joint Framework for Collaborative and Content Filtering	advances algorithms also although analysis architecture arti based basu bergstrom breese cation cial classi cohen collab combinations computer conference content cooperative corp correlation courtesy crammer describ digital discussion empirical equipment evaluation features four grouplens have heckerman hirsh http iacovou identity imdb improvement indicate information intel investigated items jrank kardie kernels ligence ltering minor national netnews neural only oration orative pages pranking predictive proceedings processing quadratic ranking recommendation references resnick results riedl same singer social suchak supported systematically systems testing there training uncertainty used useful users using various very when with work
http://doi.acm.org/10.1145/1008992.1009005	8	Probabilistic Model for Contextual Retrieval	agglomerative allan amherst beeferman berger boston brown center challenges clustering della engine held information intelligent lafferty language machine massachusetts mathematics mercer modeling parameter pietra proceedings query references report retrieval search september sigir sigkdd statistical translation university workshop
http://doi.acm.org/10.1145/1008992.1009130	118	eMailSift: Mining-based Approaches To Email Classification	agarwal asso based ciation concepts conference data databases etween graph hill holder ieee information intel introduction items kamb kaufmann large ligent mcgill mcgraw mining modern morgan proceedings references retrieval rules salton sets sigmod systems techniques
http://doi.acm.org/10.1145/1008992.1009056	47	A Search Engine for Historical Manuscript Images	aaai acknowledgments adapting alto amherst analysis annot annotated annotation annual approach arti attention august australia automatic barnard based better blei both built bunke canada center challenges challenging choquette ciency cluster collection computer conclusions conf congress copenhagen corfu croft cross currently cursive data datasets denmark describes document documents done duygulu dynamic editor either european evaluation even ever experiments feature figure finally finland forsyth freitas george good govindara great greece handled handwriting handwritten historic historical holistic ieee image images improve improving indexing information intel intelligent interp interpolated investigating january ject jeon jordan journal july june kindly lack language large lavrenko learning lexicon libraries library ligence ligent lingual machine madison manmatha manuscripts marti massachusetts matching maybury media melbourne model modeling models much multi needed nonrel olated original orts page pages palo paper particular pattern performance pictures ponte possible precision presentation press prob probabilistic proc processing processors produces promise provided queries rank ranks rath recognition references relevance remain remains removed report requires results retrieval retrieve retrieving returned scale scanned segmentation semantics september shape show sigir solution space spotting srimal stages statistical synthetic system tampere task technical technique that theories this time toronto training translation univ using vancouver vision vocabulary volume warping washington when while with word words workshop would
http://doi.acm.org/10.1145/1008992.1009002	6	Building an Information Retrieval Test Collection for Spontaneous Conversational Speech	academic access acoustics again agreement american annual appear archives aslib assessing audio automated automatic based behavior boston browsing buckley byrne carletta categories categorization cation chen cient cieri classi cleverdon collections computational computer concept conference construction cormack corpora corpus cran dagobert detection development devices digital document documentation ectiveness empirical english enriched evaluating evaluation event examination explication factors form franz garofolo godfrey goodman group gustman history http huang identi ieee index information international ject jects joint jones journal judges judgment judgments kappa katter kluwer language large libraries linguistics malach management martin measure measurement meeting methods modeling multilingual number okapi optimal oral organization pages play proceedings processing ramabhadran recognition references relevance research retrieval review rijsbergen robertson rong samuel scale schamber science search second shef sigir signal smoothing society soergel spandh sparck speech spoken spontaneous stability statistic storage story study success supporting swag switchboard tasks techniques technology telephone test tests text topic topical towards track tracking transactions transcription trec types uence underlying understanding variations volume voorhees whittaker william word working xiaoli yang
http://doi.acm.org/10.1145/1008992.1009051	43	An Automatic Weighting Scheme for Collaborative Filtering	algorithm algorithms analysis annual architecture artificial bergstrom breese brochers collaborative computer conference cooperative development empirical filtering fourteenth framework generalized grouplens heckerman herlocker iacovou information intelligence international kadie konstan model netnews nicholas open pages performing predictive proceeding proceedings references research researech resnick retrieval riedl sigir soboroff space suchak supported uncertainty vector work
http://doi.acm.org/10.1145/1008992.1009089	77	Expertise Community Detection	acceptance accepts ackerman acknowledged across ages aggregate algorithm american amore analysis approach aslam assign assigned associates authoritative authorities authority automated average based beach belief below between beyond bias biased brokers build cambridge candidate case cikm classes combinations combining communities compared computed condorcet conference consensus consisting demoir detection differs discovery domain each ecological ecscw editor either enterprise environment erlbaum evaluation evaluations evidence expert expertise experts farr finder finding first five formed from fusion gaithersburg generated generates glaser gleneden government grace granovetter graph harman here high higher hits house however hubs hyperlinked improved inclusive journal judges judgment kleinberg knowledge known koushik lawrence likelihood list locator management managing mapping maybury mccune mean measures member mitigate modlin montague more most multiple nature nominate nominated office oregon organizational organizations others overall pages panel peer person pipek practitioners precision presented press printing problem proc properly publishers queries query ranked rankings rate rated recent references reflecting relevance replaced responders results retrieval revealing sample sampling scores searches self september sharing shaw since snowball sociology software sources specific strength subsequent survey surveyed system systems table taken test text than that then three ties towards trec trust used user using various vetted washington waves weak weight were which workshop wulf yimam
http://doi.acm.org/10.1145/1008992.1009012	13	Locality Preserving Indexing for Document Representation	advances american analysis ando applications athens bartell belew belkin bingham canada case chung clustering conference conferences copenhagen cottrell data deerwester denmark devroye dimensionality discovery document dumais eigenmaps embedding furnas generalization graph greece gyorfi harshman image improves indexing information inter international iterative journal knowledge landauer laplacian latent lugosi mannila mathematics measurement mining multidimensional neural niyogi number optimal orleans pattern precision probabilistic proc processing projection random recognition reduction references regional rescaling residual scaling science semantic series seventh sigir sigkdd similarity society space special spectral springer systems techniques text theory vancouver verlag york
http://doi.acm.org/10.1145/1008992.1009081	69	Comparison of Using Passages and Documents for ∗ Blind Relevance Feedback in Information Retrieval	added approach automatic average based behavior boughanem buckley certain cikm conference deling denote dependent development difference document documents environment erformance ertson erty etween examine expansion feedback from improvement improving information iris isolated jones language length ltering maglaughlin management mitra newby okapi othesize pages pass passage passages proceedings processing qsdr query references relationship relevant research results retrieval reveal routing same scores setting singhal sixth system terms text there this topic trec using walker with yang zhai
http://doi.acm.org/10.1145/1008992.1009057	48	Display Time as Implicit Feedback: Understanding Task Effects	academic activism activities additional address affects alumni amazon american amherst appendix applying assistance auto awards booking books borlund brain brown browsing bystrom catalog change check checking chen chords claypool companies compile compiler complexity computer concert conference conferences conflict consult cooper correct course database development digital directions directory dissertation documentation download downloads ebay effectiveness email employment english entertainment evaluating evaluation exam exit fantasy fellowships find flights follow friend friends funding general gigs government grants greek groups guitar health hobbies holidays homepage homepages homework household housing images implicit indicators information ingwersen install instruction insurance intellectual intelligent interactive interest interests interfaces international jarvelin jobs journal know knowledge language league learn learning legal library locate look looking maintenance management mandolin maps material materials measure method movie movies music napster network news online operating options oral packet page pages paper papers parking part people person personnel philosophy political possessions predicting preparing printing proceedings processing profs project property publication purchase python quals quotes read reading record recording recreation references registration relevance rent repair research resources retrieval review reviewing reviews running schedule schedules scheduling scholar science search searching seeking selecting selling setting shop shopping soccer society software sources sports staying student studio studying stuff subject subjective submitting systems tablatures task tasks teaching technology theater tickets touch train transcripts translation travel trouble university updates user utilization varying viewing waseda weather website what with work writing
http://doi.acm.org/10.1145/1008992.1009017	17	Belief Revision for Adaptive Information Retrieval	about aboutness academic adaptive agents agrawal alchourr algorithms american analysis annual anytime application arti aspects association august automatic automatically available axioms barreiro base based bases belief benchmarking berkeley bocca boolean brisebois bruce bruza california cambridge canada card carlo categorization change cheng chile chris cial classical classifying cohen commonsense complexity computational computer conference context contraction counterfactual crestani croft data databases defeasible description development developments document documents douglas dublin dubois dynamics editor editors engineering englewood entrenchment epistemic erlbaum expectations experiments extended fabio fast feedback fields fifteenth filtering fisher fourteenth francisco fredric from functional functions gaithersburg gardenfors general grove hall handbook harman hearst hierarchically hill hillsdale http huibers hull human icml imaging inference information informational intel interaction international introduction investigating ireland iterated japan jarke jersey joint jones jorge journal july kaufmann kindo kluwer know knowledge kohlas koller lalmas lang large lawrence learning ledge lepage ligence logan logic logical logics losada ltering machine makinson march martha marti maryland massachusetts matthias mcgill mcgraw meet mellish methods mining model modeling modelling models modern montr moral moran morgan morimoto moshe nagoya nashville newell nist nonmonotonic norwell november october operator organizes outline paci pages partial personal perspective pollack possibilistic prade prentice press probabilistic proceedings processing psychology publishers pubs ranking rdenfors reasoning reece references relevance research retrieval review revision revisions richard rijsbergen rocchio rules sahami salton santiago science second sensitive september seventeenth seventh sigir singer smart society song sprack srikant states symbolic system systems technology tennessee text that theoretical theory tong towards track transactions trec uncertainty using vardi very vldb voorhees watanabe williams with wong words york yoshida zaniolo
http://doi.acm.org/10.1145/1008992.1009006	9	Discriminative Models for Information Retrieval	advances algorithms allan american analysis annual application applied approach asis automatic based bayes bayesian berger better brin bringing build burges callan capturing case categorization cation cikm citation citeseer classi cliffs cmis combining comparison computational conference cooper craswell croft csiro dabney data datasets della dependencies design development digital discovery discriminative distribution distributions document effectiveness empirical englewood entropy entry estimation european event experiments exploiting exponential extraction features filtering formal from generation generative greiff hall harter hawking hiemstra home homepages html http huizinga icml ijcai imbalanced importance increase indexing inferring information involving item joachims john jones jordan journal kantor karger kernel keyword knowledge known kraaij lafferty language large learning lemur library linguistics literature logistic machine machines making malouf mani many mathematical maxent maximum mccallum method methods mining model modeling models motwani nallapati natural nding neural nigam nlplab ogilvie order overview page pagerank pages parameter part pattern pietra ponte porter practical prentice press principle prior probabilistic probabilities probability proceedings processing project python qrels queries query ranking ratnaparkhi recognition references regression relevance relevant representations research retrieval rijsbergen robertson salton scale scholkopf science sciences search searching sentence sigir smart smola smoothing society sons spaces sparck specialty speech staged stanford statistical stephen study support system systems tagger technical technologies technology teevan term terms testing text textual theory toolkit track training trans trec trecweb tree tutorial unbalanced using vapnik vector weighting westerveld wiley winograd with words workshop zhai zhang zhangle
http://doi.acm.org/10.1145/1008992.1009062	53	Web Taxonomy Integration through Co-Bootstrapping	aaai addison advanced agrawal algorithm algorithms alignment american anchor annual application approach artificial athena austin automated automatic automatically baeza bayardo bayes berkeley berners between blum boostexter boosting boostingbased bottom breckenridge bullet cambridge canada catalogs categorization chalupsky chimaera classification college collins colt combining commerce comparison computational computer concept concepts conference confidence context corpora cristianini data database databases decision development doan document domingos edbt electronic emnlp empirical entity environment estimation event examination exchange experiments explicit extending extracted facilitating feedback fensel fikes flairs florida fourteenth freund fusion generalization germany groh halevy hall hawaii hendler hierarchy hill hofmann hong honiden ichise ijcai improved induction information integrating integration intelligence interactive international introduction jannink joint journal knowledge kong konstanz labeled lacher language large lassila learning lectures leveraging line lncs machine machines madhavan madison maedche management mappings matching mccallum mcgraw mcguinness meir mendelson merge merging methods miningbased mitchell mitra models modern msri musen naive named national natural neto nigam nonlinear nonlocal ontologies ontology ontomorph overview park predictions prentice press principles proceedings processing prompt rated ratsch reasoning references relevance representation research retrieval ribeiro rice rocchio rule salton schapire sciences scientific seattle semantic semi sharing shawe sigdat sigir silver singer smart smola society sources springer springerverlag srikant stumme sunnyvale support symbolic system takeda taylor technology text theoretic theory through tool toronto training translation university unlabeled unsupervised using vector verlag very wesley west wide wiederhold wilder with workshop world yang yates york
http://doi.acm.org/10.1145/1008992.1009038	33	Information Retrieval using Word Senses: Root Sense Tagging Approach	ambiguity american analysis annual august berkeley california comparative conference croft deerwester development dumais eriments ertson foundations furnas harshman hofmann indexing information jones journal krovetz landauer language latent lexical management manning model natural pages part press probabilistic proceedings processing references research retrieval schutze science semantic society statistical systems walker
http://doi.acm.org/10.1145/1008992.1009076	64	Merging Retrieval Results in Hierarchical Peer-to-Peer Networks	appear approach callan cased client computed conference content database document documents engine experiments hybrid information international kirsch knowledge learning lemur management merging multiple networks ogilvie over patent peer proc ranking references relevance results retrieval scores search semi supervised systems text toolkit topeer transactions using wherein
http://doi.acm.org/10.1145/1008992.1009053	45	A Nonparametric Hierarchical Bayesian Framework for Information Filtering	aaai accuracy advances algorithms american analysis annals antoniak antonio applications approach architecture arti association august automatic automating balabanovic bartlett based basu bayes bayesian becker bergstrom billsus blei book boosted breese cambridge canada categorization cation chain chinese cial classi claypool cohen collaborative combining communications computer concentration conference content curves data demographic density dept diagnosis dietterich digital dirichlet document editors edmonton eighteenth empirical ensemble environments erent escobar estimation experiments experts feedback fifteenth fifth figure framework francisco gaussian ghahramani giles gokhale grouplens hall heckerman hierarchical hierarichical hirsh horvitz hybrid iaai iacovou identifying improved inference information intel interesting international jordan journal june kadie kaufmann laborative large lawrence learning libaries ligence ligencen ltering lters machine machines maes margin markov melville memory methods miranda mixture mixtures model models mooney morgan mouth muramastsu murnikov nagara national neal nested netes netnews neural newspaper nite nonparametric online open pages parameter pazzani pennock personality platt popescul portland predictive prentice press probabilistic probabilities problems proc proceedings process processes processing rasmussen recommendation recommendations recommender recommending references relevance report resnick restaurant retrieval review riedl rocchio sampling sartin scholkopf schuurmans schwaighofer seattle shardanand shoham sigir sites smart smola social sparse statistical statistics supported sushak syskill system systems technical tenenbaum text thirteenth topic toronto tresp uncertainty ungar university using values washington webert west with word work workshop york zhang
http://doi.acm.org/10.1145/1008992.1009021	20	Resource Selection for Domain-Specific Cross-Lingual IR	acoustics adaptation alignment american annual arabic asian association august bilingual brown callan campaign carbonell chen chinese clef clir collection college combination combining come communication comparative computational conference conversational corpora corpus cross crosslanguage crosslingual cues darwish dependencies domain draft english entropy estimation europarl european evaluation evidence experiments exploiting fifth finland fork forum foster franz fraser frederking french futures general geng german girt grams hermann ieee ijcai improved information integrate international japan jiang journal june kando khudanpur kluck koehn language languages lemur lingual linguistics list machine maryland mathematics maximum mccarley meeting mercer mining model modeling models multi multilingual multiple nist norway notes november ntcir oard october ogilvie open overview pages parallel parameter park part peters pietra proc procedure proceedings processing publication query recognition references research resnik resources results retrieval road roadmap rogati rosenfeld roukos savoy science sept seventh seymore short sigir signal simard society sources special specific speech spoken statistical stemming stopword story strategies system take tampere task technology tenth text thesaurus third tokyo toolkit topic topics track translation translingual transparent trec trondheim unpublished using weischedel when with working workshop yang
http://doi.acm.org/10.1145/1008992.1009088	76	The Document as an Ergodic Markov Chain	american apart applied beaulieu burgess cautionary chains chater cognition concepts conducted conference context corpus course discourse distribution document ectiveness editors erent erty expansion experiment explorations feedback from gull hancock hoenkamp information journal language level livesay lund markov material memory method minimization models note oaksford okapi operators others ours oxford pages part press procedure proceedings processes query rational references relevance relevant represent resolution retrieval risk robertson science sentences sets sigir society space steyvers technology text than that topics trec unitary university used validation walker whole words zhai
http://doi.acm.org/10.1145/1008992.1009108	96	A Search Engine for Imaged Documents in PDF Files	alto anal analysis based character coding computer conf dial digital document documents doermann image imaged images indexing information libraries palo proc recog references retrieval retrieving shape smeaton spitz survey understanding using vision word workshop zhang
http://doi.acm.org/10.1145/1008992.1009020	19	Translating Unknown Queries with Web Corpora for Cross-Language Information Retrieval	across algorithm altavista american among analysis anchor annual approach artificial automatic based bilingual categorization center cheng chien chinese choquette chuang church coling collocations comparable computational conference construction contiguous corpora corpus correspondences croft cross crosslanguage darpa dias dictionary diekema digital durand english equivalence evaluating experiments exploring extraction finding from fung gale german grefenstette guillore harman hatzivassiloglou henzinger huang hull identification identifying ieee information inproc intelligence interests introduction isabelle issue joint journal keyword kilgarriff kupiec kwok language languages large lavrenko lecture lexical lexicons libraries lingual linguistics localmaxs lopes marais mckeown meeting melamed mining model models morics multilingual multiword natural nguyen nonparallel notes noun ntcir oard overview parallel phrase pircs probabilistic proc queries query querying rapp references relevance report research resnik retrieval review science search sigir silva silverstein simard sixth smadja society special speech statistical subject systems technical technology teng terms text texts transactions transitive translating translation translational translations trec tree units unknown unrelated users using very voorhees wang weischedel word words workshop yang
http://doi.acm.org/10.1145/1008992.1009098	86	Answer Models for Question Answering Passage Retrieval	algorithm analysis annual answer answering approach association based baseline best came cation class classi clues computational conference correspond croft experimental expressions fact from hovy improvement information language learning likelihood linguistics measurements meeting metzler modeling nist numbers over page pages paper passages patterns performance ponte potential proceedings publication query question questions quoted ranking ravichandran references results retrieval right sabboutin same shows sigir since slight special statistical summarize surface system table take tenth text this together trec tuned tuning using
http://doi.acm.org/10.1145/1008992.1009128	116	Exploiting Hyperlink Recommendation Evidence In Navigational Web Search	adcs amsterdam anatomy australia brin brisbane canb craswell dels ehugoz elds engine eriments erra ertextual ertson evidence extension fame fortune graph hawking home http indegree independent information jagopalan kumar large maghoul microsoft multiple nding page pagerank predicting proceedings query raghavan references research scale search simple stata structure systems taylor tomkins transactions unpublished upstill weighted wiener zaragoza
http://doi.acm.org/10.1145/1008992.1009009	11	Length Normalization in XML Retrieval	abolhassani academic amati amitay annals another answering applied approach approaches based berger bootstrap buckley callan cance carmel chapman comparisons components computer conference content contributions croft cross database divergence document documents duisburg ective ectiveness editors efron eighth entry ercim erent erty evaluation first fourth fragments from fuhr generation grei grossjohan hall harman hidden hiemstra hill http hyrex importance inex inference informatik information initiative introduction jackknife jlovic journal juruxml kamps kazai kluwer kraaij lalmas language leek length list look maarek malik management mandelbrod markov marx mass mcgill mcgraw measuring methods miha miller mitra model modeling models modern morgan most ninth normalization novelty ogilvie overview page pages parametric performance pohlmann practice preproceedings prior probabilistic probabilities proceedings processing projects publishers queries query question randomness references relevance relevant retrieval retrieve retrieving rijke rijsbergen rnsson salton savoy schwartz science search searching second series seventh sigir signi sigurb singhal smart smoothing soboro statistical statistics structured study system systems technology tests text theory thesis tibshirani tijah track transactions translation trec twelfth twente twenty university using vert voorhees vries westerveld what wilbur wilkinson with workshop york zhai
http://doi.acm.org/10.1145/1008992.1009086	74	A Session-Based Search Engine	ability after algorithm along american based cadiz canada change chien cikm clicked combination conference contextual current cutrell decay divergence document documents dumais during each engine exploiting factor feature feedback given history huang importance information interactive interface international issue jancke journal knowledge lafferty list logs management model oyang page pages past personal proceedings queries query rank ranked ranking receiving reduce references relevant retrieval returns robbins sarin science search seen session shen sigir society stuff suggestion summaries summary system technology term their they title titles toronto user users view visualization visualize volume weight where with zhai
http://doi.acm.org/10.1145/1008992.1009137	125	Implicit Queries (IQ) for Contextualized Search	access algorithm analyze analyzes author automatically based birnbaum brin budzik cadiz calendar chang composing conference context copyright cutrell czerwinski dantzich demonstrate dumais dziadosz email files finds free from hammond held henzinger identifies implicit important includes index information international items jancke journal july just knowledge maes mail management messages milch news other owner pages personal plug proceedings queries query reading references related retrieval rhodes robbins robertson sarin search seen sheffield sigir south stuff system systems that tiernan time user using version visualizing what which wide words world wrappers written yorkshire
http://doi.acm.org/10.1145/1008992.1008996	1	Evaluating High Accuracy Retrieval Techniques	abdessamad access adam addison advances algorithm algorithms almost ambiguity american analysis annual answering approach approaches aravind association august automatic based beckwith berger boris bruce buchholz case categorization cation challenges christian clarke classi cognitive coling combining communications computational computer conference cormack corpus craswell croft cronen daelemans database deepak dell derek detailed development disambiguate disambiguation distinguishing document donald eacl echihabi editor eduard eighth eleventh emnlp empirical euroconference evaluation expanding expansion experiments extraction fact fellbaum formation frank from general george gerald giles global grishman gross harabagiu hawking hermjakob hovy hozumi hull ieee inference informaion information international introduction january jasis jersey jimmy jinxi john joshi journal katherine katz keyword kisman laboratory lafferty language lavrenko lawrence learning lexical lexically line linguistics local lovins lynam machines magazine management mandala meaning mechanical meng methods metzler miller model modeling models multitext natural network nineteenth ninth nist november overview pages parsing parts passage patterns pedersen people performance ponte predicting presentation press princeton proceedings processing providence quantifying query question questions ralph ranlp ravichandran rebecca recent redundant references reformulation related relations relevance remove report research resource retreival retrieaval retrieval richard rila sabine salton sanda sanderson scie science scienti searching selection semantic sense senses seventeenth shallow shapaqa sigir sixteenth society somerset speech springer srinivas statistical statistics stemming steve study submitted super supertags support surface system systems takenobu tanaka technical techniques tenth text tokunaga townsend track transformation translation trec tutorial university untagged using vector vectors verlag voorhees walter want wesley what wide wilbur with word wordds wordnet words world yang yiming zhang zhou
http://doi.acm.org/10.1145/1008992.1009134	122	Filtering for Personal Web Information Agents	acapulco adapt adaptive advantage agent agents algorithm analysis arti australia autonomous billsus browsing callan cation chen cial classi classify coding conference cuments data demands development dissemination document documents does dublin eled eliminate endent erformance erforms ersonal ertson examined eyond feedback figure found four gale generally hard higher howe identi information intel inter interesting international ireland joint learning lewis ligence ligent lightweight lisb little ltering machine marginal melb methods mexico minmax minneap more need olis optimization ositive ourne outp paired parameterizations pazzani performance portugal presented proc proceedings proves quickly ranges references requiring research results retrieval revising sample searching second sequential setting sigir sites somlo stabilize storage store sycara symposium table test tested text tfidf than that threshold thresholds topic topics training uctuate unlab user using vectors viable webmate while within worse would yields
http://doi.acm.org/10.1145/1008992.1009093	81	Topic Prediction Based on Comparative Retrieval Rankings	amount analysis anchormap annual answer appears approaches based basic being blind both bruce changing clarity conclusion conference croft cronen designed developed development difference directly does effective effectiveness entire essential exactly feedback fifth followup from getting gives good have here however hundred identi impact importance important improve improved improves improving individual information international issue jobs just largest like made make measure measures model more most much nearly necessarily nitely nothing number open optimizing optimum other outperforms overall pages parameters particular perform performance poorly positive predict predicting predictive proceedings query question random reason references research results retrieval runs same scores several sigir simpler small some steve still such suggest surprisingly system terms than that these this those tools topic topics townsend trying twenty used wants well were which while will with works worse yields zhou
http://doi.acm.org/10.1145/1008992.1009036	32	Parameterized Generation of Labeled Datasets for Text Categorization Based on a Hierarchical Directory	advances aggressive algorithms analysis appear application approach approaches association automated automatic based benchmark bennett bicknell blake bloomsbury broad buckley budanitsky burges cambridge categories categorization cation chakrabarti chen cikm classi cohen collection combination comp competitive computational concept conf conference content darpa data database databases daviddlewis describe devices directories distance distribution documents duda dumais ecml edition editor editors electronic electrotech engineering evaluating evaluation experimental extracted feature features february fellbaum finin fire flannery forum from gabrilovich ghani goblet gonzalo grobelnik harman harry hart headings heckerman herscovici hersh hickam hierarchical hierarchy hirst horvitz html http hypertext icml indicators inductive information interactive jair jasis ject jiis jmlr joachims john jones joshi kaufmann kernel know knowledge labrou lang language large learning lection ledge leone lewis lexical library linguistics lter maarek machine machines make making many markovitch master measures medical medicine meng merz mesh meta methods mining mladenic mlearn mlrepository mobile models morgan naacl national natural nature netnews newbold newsweeder nigam numerical ohsumed ontology oriented other ottawa pages pattern pennock personalized petruschka platt pocket potter practical press probabilistic proc programs punera quinlan rada ranking recipes redundant references relevant reliability repository representations research resnik resources results retrieval reuters rose rowling sahami santamaria scale scene schoelkopf scott sebastiani second selection semantic senses september sequences sigir sigkdd similarity slattery smola sons speech springer statistical structure study support surveys svms symbolic systems taxonomy test testcollections teukolsky text theory thesaurus thesis tipster topics university using vapnik vector verdejo verlag vetterling volume wang wide wiley with word wordnet workshop world yahoo yang
http://doi.acm.org/10.1145/1008992.1009030	27	Learning to Cluster Web Search Results	accounts adaptive advances advice africa afsc aftermath agrawal allan almuajaha alternet america amherst amnesty analysis animal animals annual apple application arab arabbay arabic atari atariage audio august autobytel automobile background bahlmann based beach birds breaking brief browsing bruce budapest burges business california canada care career carlynx cars casta cats center cheats chien chin chinese christianity chuck city classic classified club clubs cluster clustering codes collection collections college columbus combining coming complete computer concepts concerns conference conflict congress constant council countries country cover crisis croft culture cutting data database databases defenders defending definitions demonstration department destination development directory discovering discovery division document drivers dynamic earthtimes east economy eighth elements employers employment engine england enthusiasts ethnologue etzioni evaluation everything examples experience extraction face fact factory facts feasibility feature federal figure find finding flag flags flipdog florida forum france free freedom friedman from game gather geography global google grouper guardian guide guides hastie hearst herald hierarchical hints history home homepage house http human hungary hussein hypothesis improving index information institute intelligent interaction interactive interface international internjobs internship interviews introduction iraq iraqi ireland jaguar jaguars jdrcnwa joachims jobs jobstar jobweb june justice kahn kansas kansascity karger kernel keyphrase kids kits knowledge land large lawrie learning lent leouski letter letters leuski library links list lkopf lonely lovers luton macnn magazine mail main major makes making maps massachusetts masterpiece methods miami middle mining mission model models money monster muscle nations neurocolt newport news next nola north northern october onca online operation overview owner page pages panthera peace pedersen pederson people perry phliadelphia pittsburgh planet policy political polytechnic post posters posting practical preparing press prints proceedings professional purchase quality query race racing ranked rarity reach reexamining references referral regression rensselaer repair reparation report reports research resin resource resources result results resume resumes resumezapper retrieval riao rights saddam sale sample samples scale scatter schlkopf science scientist search seattle security series service services shamanism sheets shopping sigir site sites smola snow special species specific spirit springer srikant state statistical studies summarization support susan technical techniques text this tibshirani tiger time times tips today topic toronto tourist travel tree trends tutorial twelfth type united university unlimited unofficial update vector verlag very visa visas vivisimo warning watch white wide wildlife winners with witness words world write writers writing yahoo yahooligans york your zamir zensearch zurich