|
- All good things must end, and we've now come to the end of this
project. It's been a fantastic opportunity to explore an important
new research problem, and to work with some fantastic people on that!
One thing we are doing here at the end of the project is working with
the Linguistic Data Consortium to deposit the annotations that we and
others have created for the Avocado Email Research Collection.
- Ph.D. Student Xin Qian has led the work two new papers that
explore managing conversational interaction for search that have
implications for search among sensitive content, working with
Professor Doug Oard. The first paper is for TREC 2021, where Xin
explored alternative ranking techniques in the Conversational
Assistance Track. The second will be presented at the 2022
iConference, on the use of conversational representations of
historical figures in museum exhibits. Because this work involved
searching informal content, privacy protection issues will surely
arise.
- We're celebrating the graduation of Ph.D. student Mahmoud Sayed,
whose dissertation topic was Search Among Sensitive Content!
- Our article on using machine learning to support review for
Freedom of Information Act exemptions has been accepted for a special
issue on Computational Archival Science of the ACM Journal on
Computing and Cultural Heritage that will likely be published in 2022.
This is joint work between Professor Jason Baron, Ph.D. student
Mahmoud Sayed, and Professor Doug Oard.
- Professor Doug Oard presented a Tutorial with Graham McDonald
from the University of Glasgow on Search Among Sensitive Content at
the 2021 European Conference on Information Retrieval.
- Professor Doug Oard presented a keynote address on The Importance
of Domain Knowledge for Computational Law at the SIGIR 2020 workshop
on Legal Intelligence .
- Professor Doug Oard and Ph.D. student Jyothi Vinjumur will be
presenting our 2018 TOIS paper on "Jointly Minimizing the Expected
Costs of Review for Responsiveness and Privilege" on video at the 2020
ACM SIGIR conference and discussing that paper with the participants.
SIGIR offers TOIS authors this opportunity as a way of increasing the
exposure of articles that are published in TOIS, but because of the
annual schedule of SIGIR there's a bit of a lag between publication
and the presentation.
- Ph.D. student Mahmoud Sayed will be presenting a paper from
pretty much the entire team on "A Test Collection for Relevance and
Sensitivity" at SIGIR 2020, which had been originally planned for
Xi'an China, but which will now be held online. Professor Doug Oard
and REU Student Will Cox plan to join him "there" to discuss the paper
with online participants.
- We're celebrating the graduation of REU student Jonah Rivera!
Will Cox is continuing with us for the summer, working on automated
content analysis for our new test collection.
- We've finished the initial annotations for our new relevance and
sensitivity test collection!
- Our full paper on "We Could, But Should We? Ethical
Considerations for Providing Access to GeoCities and Other Historical
Digital Collections" led by Professor Jimmy Lin had been scheduled for
presentation at the ACM CHIIR conference in Vancouver in mid-March of
2020, but the COVID-19 coronavirus had other ideas. So that paper is
available from ACM, but we didn't get the chance to present it.
- Professor Doug Oard and Ph.D. student Mahmoud Sayed contributed
to the SIGIR 2019 workshop report on "FACTS-IR: Fairness,
Accountability, Confidentiality, Transparency, and Safety in
Information Retrieval" that was published in the December 2019 issue
of SIGIR
Forum.
- We welcomed two new Research Experience for Undergraduate
(REU) students to the project in August, 2019: Will Cox and Jonah Rivera.
Will and Jonah are working with us on creation of a new email test
collection for relevance and sensitivity.
- Ph.D. student Mahmoud Sayed presented a talk on "Detecting
Sensitive Content" at the SIGIR Workshop on Fairness, Accountability,
Confidentiality, Transparency and Safety in Information Retrieval,
(FACTS-IR) in Paris, France in July 2019.
- Ph.D. student Mahmoud Sayed presented a full paper on "Jointly
Modeling Relevance and Sensitivity for Search Among Sensitive Content"
at SIGIR 2019 in Paris, France in July 2019.
- Professor Doug Oard gave a talk on "The Future of Ubiquitous
Spoken Content" at the Good Systems Seminar at the University of Texas
at Austin in April, 2019.
- Professor Doug Oard is helping out as the Program Committee chair
for the First ACM SIGIR/SIGKDD Africa Summer School on Machine
Learning for Data Mining and Search (AFIRM) in Cape Town, South Africa
in January, 2019.
- Professor Doug Oard gave a talk on "Searching Intermixed Shareable
and Sensitive Content" at Wadhwani AI in Mumbai, India in December,
2018.
- Professor Doug Oard and former Ph.D. student Dr. Jyothi Vinjumur
gave three talks at a full-day seminar on E-Discovery at the National
Law School of India in Bangalore, including one on "Jointly Minimizing
the Expected Costs of Review for Responsiveness and Privilege" and one
on "An Information Scientist's Perspective on U.S. E-Discovery
Practice", in December, 2019.
- Professor Doug Oard gave a talk on Searching Intermixed Shareable
and Sensitive Content at Wadhwani AI in Mumbai, India in December,
2018.
- Professor Doug Oard gave an invited talk on "When You Can't
Review Everything, What Then?" at the Towards Assistive Digital
Sensitivity Review (TASER) Workshop at the University of Glasgow in
the UK in December, 2018.
- We organized an Open Forum on Safe
Search Among Sensitive Content at the 2018 Society of American
Archivists Conference in August, 2018.
- Professor Doug Oard gave a talk on this project at Kyushu
University in Japan in July, 2018.
- iSchool Ph.D. student Jyothi Vinjumur successfully defended her
dissertation in April, 2018!
- We went live on the Web with a demo version of our interactive
system for training classifiers to recognize sensitive email in March,
2018. Check it out at http://clip-sasc.umiacs.umd.edu/.
The demo version works with the Enron email collection, but you can
also download it from here and run it on your own
email. Unzip, start it as described in the readme, index an mbox collection (e.g., as a "takeout" file from gmail), and you should be off and running.
- Professor Doug Oard attended the NTCIR Workshop in Tokyo Japan
in December, 2017 to explore the design of a privacy task for the
Lifelogging track of NTCIR 14, one of the world's major information
retrieval evaluation venues.
- Ph.D. Student Yulu Wang successfully defended her dissertation in
November, 2017!
- Professor Doug Oard presented at talk on this project at the
University of New Hampshire in September, 2017.
- Professor Katie Shilton presented a paper at the Digital
Humanities Conference's First International Workshop on Privacy
Sensitive Collections for Digital Scholarship in Montreal, Canada in
August, 2017.
- Professor Doug Oard gave a public presentation on "Creating Search
Engines that Know What Not to Find" at Search Engines Amsterdam in the
Netherlands in June, 2017.
- Professor Doug Oard presented a paper on "When is it Rational to
Review for Privilege?" at the ICAIL 2017 DESI Workshop in London in
June, 2017.
- Professor Doug Oard presented a talk on this project at the
University of Glasgow (UK) in June, 2017.
- Ph.D. Student Mossaab Bagdouri created our first system for
detecting sensitive content in email in June, 2017.
- Computer Science Ph.D. Student Mossaab Bagdouri successfully defended his
dissertation in May, 2017!
- Professor Doug Oard gave a talk on this project at the National
Institute of Informatics in Tokyo, Japan in December, 2016.
- Professor Doug Oard gave an invited talk on "The Other Side of
the Coin: Proactive Language Technologies for Protecting Sensitive
Content" at the AAAI Fall Symposium on Privacy and Language
Technologies in Arlington, VA in November, 2016.
- Professor Doug Oard gave a talk on Effectively Searching
Among Sensitive Content at the University of Florida in
September, 2016.
- Professor Doug Oard inaugurated the new project with a talk
on Effectively Searching Among Sensitive Content at the
University of Central Florida in September, 2016.
|
Project Pages:
|