This page lists papers and other topics of interest for the weekly IR seminar. For Spring 2003, the seminar meets Mondays from 1:30-3:30 in CS 203.
The seminar has a mailing list. To join or leave the list, go to
http://ciir.cs.umass.edu/mailman/listinfo/irseminar. To send messages to people on the list, email
irseminar@ciir.cs.umass.edu.
Seminars are listed in reverse chronological order. Check the date carefully because some may be listed for the future.
Monday, May 12th
- Yang, Zhang, and Kisiel, "A scalability analysis of classifiers in text categorization." To appear in SIGIR 2003. PDF (Hard to view on the screen, but prints fine.)
- Teevan and Karger, "Empirical development of an exponential probabilistic model for text retrieval." Draft of paper to appear in SIGIR 2003. PS
- Would love to get another paper in here, but it looks like it won't be available in time.
- Last class of the semester
Monday, May 5th
Two quite different papers today. Was hoping to get a SIGIR paper from Robertson et al, but don't have it yet.
- Darwish and Oard, "Probabilistic Structured Query Methods." Draft of paper to appear in SIGIR 2003. PDF
- Sparck Jones, Robertson, Hiemstra, and Zaragoza, "Language modelling and relevance". Appears in Language Modeling for Information Retrieval, Croft and Lafferty, eds, 2003. PDF
Monday, April 28th
Wednesday, April 23rd (Monday class schedule)
- Pinto, McCallum, Wei, and Croft, "Table Extraction Using Conditional Random Fields." To appear in SIGIR 2003. PDF
Monday, April 21st
- Patriots Day, we do not meet
Monday, April 14th
- X.Li and Croft, "Time-based language models." CIIR technical report. PDF
- Allan, Wade, and Bolivar, "Retrieval and Novelty Detection at the Sentence Level." Draft of paper to appear in SIGIR 2003. PDF
Monday, April 7th
- Zhai, Cohen, and Lafferty, "Beyond Independent Relevance: Methods and Evaluation Metrics for Subtopic Retrieval." Preliminary draft of paper to appear in SIGIR 2003 (do not cite or distribute). PDF
- Second paper TBA.
Monday, March 31st
- Rath, Lavrenko, and Manmatha, "A statistical approach to retrieving historical manuscript images without recognition." ICCV submission. PDF
- Jeon, Lavrenko, and Manmatha, "Automatic image annotation and retrieval using cross-media relevance models." Draft of paper to appear in SIGIR 2003. PDF
Monday, March 24th
- Lawrie and Croft, "Generating Hierarchical summaries for web searches." Submission to SIGIR. PDF
- Nevill-Manning, Witten, and Paynter, "Lexically-generated subject hierarchies for browsing large collections." In Int'l Journal on Digital Libraries, 2(2-3):111-123, 1999. Citeseer reference (the on-line file is 17 pages).
Monday, March 17th
- Spring break, we do not meet
Monday, March 10
- Allan and Kumaran, "Stemming in the language modeling framework." SIGIR submission. PDF
- Nallapati and Allan, "Sentence-forest language model." CIIR tech report. PDF
Monday, March 3
- IR-279: (2003) Diaz, F. and Allan, J., "Browsing-based User Language Models for Information Retrieval," submitted to SIGIR '03 Conference, Toronto, Canada, July 28-August 1, 2003. PDF
- Widyantoro, D.H., Ioerger, T. and Yen, J. (1999) An Adaptive Algorithm for Learning Changes in User Interests. In Proceedings of the Eighth International Conference on Information and Knowledge Management CIKM'99, pp. 405-412, Kansas City, Kansas, November, 1999. PDF
Monday, February 24
- Cronen-Townsend, Corrada-Emmanuel, and Croft. "Predicting the quality of answer passage retrieval." SIGIR submission. PDF
- Corrada-Emmanuel, Croft, and Murdock. "Answer passage retrieval for question answering." SIGIR submission. PDF
Tuesday, February 18 is a Monday class schedule
- but unfortunately the snowstorm closed the university, so we didn't have class.
Monday, February 17
Monday, February 10
- Allan and Raghavan. "A probabilistic model of named entities: you are what they say you are." SIGIR submission. PDF
- Conrad and Utt. "A System for Discovering Relationships by Feature Extraction from Text Databases." SIGIR 1994. PDF
Suggestions for papers to consider
- Add papers here if you like
to top