Session search image

InfoSense Group

InfoSense is an Information Retrieval (IR) research group in the Department of Computer Science, Georgetown University. We research on Information Retrieval, Text Mining, Natural Language Processing, and Machine Learning.

Now our main focus is on dynamic IR and privacy-preserving IR.


You can explore this website to know more about our group.

TREC Dynamic Domain Track Privacy-preserving IR workshop 2015
Dynamic Domain

TREC Dynamic Domain Track

We propose a new track focused on domain-specific search tasks in which professional searchers explore complex content spread across a corpus. To help such users, we need retrieval algorithms that can dynamically adjust as the user makes sense of the entities and relationships mentioned in the corpus. While TREC hosts evaluations in several domains, we propose to create domain-agnostic evaluation protocols for studying retrieval systems that "hang in there" and evolve along with the user's own understanding.

The goal of the TREC Dynamic Domain Track is to evaluate retrieval systems that help users rapidly discover information about entities in a specific content domain. (introduction ppt)

Session search image

Session Search

Complex information needs often require more than one queries in a search session. Given a series of queries that a user enters in the same session, how do the earlier queries, returned search results, and click information intertact with the user and the targeted search goal? This research investigates techniques in query formulation, query expansion, user interations, and relevance feedback for in-depth understanding of user behaviors in search sessions and better modeling of search activities with complex information needs.

This research has participated in TREC 2012 Session track evaluation and won the 2nd position in whole session search (RL2-RL4). It is published in SIGIR 2013.

Online information image

Online Information Exposure Detection

This research investigates effective natural language processing technologies to quickly identify components and/or attributes of a user's online public profile that may reduce the user's privacy. Through identifying potential risks of sharing information in social media, the system facilitates a better understanding and warns of one's vulnerability on the Web.

This is a joint project with Dr. Lisa Singh (PI) and Dr. Micar Sherr (Co-PI). The research is sponsored by National Science Foundation.

We organize the first Privacy-Preserving IR workshop PIR 2014 co-located with SIGIR 2014 at Gold Coast, Australia.

Personalized, Dynamic Browsing Hierarchy Construction image

Personalized, Dynamic Browsing Hierarchy Construction

A concept hierarchy is a set of concepts and relations between those concepts. Most hierarchies, such as Yahoo! Directory and Library of Congress Subject Headings, are large and complex. Some situations, however, call for light-weight concept hierarchies that are user-specific and task-specific. For example, a user wants to gather information from the online search engines to plan a multi-day family trip to Disneyland. He would like to quickly sort though large amounts of relevant materials to make decisions. This research examines concept hierarchy construction, a mechanism to create dynamic and personalized concept hierarchies for Web search.

Contextual Suggestion image

Contextual Suggestion

Based on a given context including city, date, time, season, and a user's personal interest profile, a contextual system makes suggestions about places to go, to eat, and to have fun. The research emphasizes on designing and implementing effective and efficient systems to search unlimitedly on the Open Web, intelligently identify and merge interesting results based on comprehensive understanding of user profiles and contexts. Novel and advanced retrieval techniques and machine learning algorithms are designed to tackle the new challeges presented in this new area.

This research has participated in TREC 2012 Contextual Suggestion track evaluation and won the 2nd position.

  • Tweets

News

  • 2017 Oct
    Sicong successfully defended his Ph.D. thesis! Congratulations!!
  • 2017 Oct
    Zhiwen won the Best Student Paper Award at ICITR 2017 for our paper "Investigating Per Topic Upper Bound for Session Search Evaluation". Congratulations!
  • 2017 Oct
    Our group has 2 full papers, 1 short papers and 1 tutorial presented at ICTIR 2017, Amsterdam. Nice work team!
  • 2017 May
    Welcome Amy, Shuchen, and Chong to join our group!
  • 2016 June
    Grace's Book on "Dynamic IR Modeling" is published!
  • 2016 June
    Welcome Jianxian and Lingqiong to join our group!
  • 2016 May
    Congratulations!! Hongkai's thesis is successfully defended!
  • 2016
    Welcome Yunyun Chen, Angela Yang, Susan Bhattarai, Joshua Kang, Simeon Meerson and Christopher Zawora join our research group!
  • 2015
    Welcome Hongkai Wu join our research group!
  • 2014
    Welcome Jie Zhou join our research group!
  • 2014
    Grace introduced Dynamic Domain Track in the 2014 TREC workshop. (download ppt here).
  • 2014
    Grace will co-organize the TREC 2015 Dynamic Domain Track with John Frank and Ian Soboroff.
  • 2014
    Welcome Shiqi Liu join our research group!
  • 2014
    Our group has 1 full paper, 1 short paper and 1 demo paper accepted by SIGIR 2014. Congratulations, Jiyun, Sicong and Chris!
  • 2014
    Grace will give the Dynamic Information Retrieval Modeling Tutorial in SIGIR 2014, with Marc Sloan and Jun Wang.
  • 2014
    Grace co-chairs the first Privacy-Preserving IR workshop PIR 2014 co-located with SIGIR 2014.
  • 2014
    Grace co-chairs SIGIR 2014 Doctoral Consortium with Shane Culpepper.
  • 2013
    Our group won the 1st position in TREC 2013 Contextual Suggestion (ClueWeb). Congratulations, Jiyun!
  • 2013
    Our group won the 1st position in TREC 2013 Contextual Suggestion (ClueWeb). Congratulations, Jiyun!
  • 2013
    New class for Fall 2013 - COSC289 Multimedia Processing. Take it if you want to know how to play with images, audios, and video files by only using simple programming skills such as Arrays and File operators.
  • 2013/6
    Our group won the 1st position in patent prior art retrieval evaluation in CLEF-IP 2013. Congratulations, Jiyun!
  • 2013/4
    Our group has 1 full, 1 short, and 1 demo paper accepted by SIGIR 2013. Congratulations, Dongyi, Sicong, and Jiyun!
  • 2013/4
    Our full paper "Utilizing Query Change for Session Search" is accepted by SIGIR 2013. Congratulations, Dongyi and Sicong!