Informal SIGIR Test of Time Award
I have the fortune of attending ICML this year and hope to report on that next week. Like other conferences, ICML includes a Test of Time award “given to papers that time and hindsight proved to be of lasting value to the Machine Learning community.” This year, the award went to `Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data‘ by John Lafferty, Andrew McCallum, and Fernando Pereira.
As an exercise, I scanned the list of titles from SIGIR 2001 and created a poll to see which papers readers would nominate for an informal SIGIR 2011 Test of Time award. The poll can be found here. Three votes per person. Poll closes on July 24, 2011.
Update: Looks like I did not read the fine print of the polling site close enough and the polls will close on July 24, 2011 or when 100 votes have been received, whichever comes first. Currently at 63 votes total. I did specify “informal”, didn’t I?
Update: The poll closed over the weekend and the top three papers captured 50% of the votes,
- “Relevance based language models”, Victor Lavrenko, W. Bruce Croft (24.39%; citations: 252/ACM,618/G)
- “A statistical learning learning model of text classification for support vector machines”, Thorsten Joachims (15.85%; citations: 60/ACM,215/G)
- “A study of smoothing methods for language models applied to Ad Hoc information retrieval”, Chengxiang Zhai, John Lafferty (13.41%; citations: 281/ACM,709/G)
The best paper at SIGIR 2001 was “Temporal summaries of new topics”, James Allan, Rahul Gupta, Vikas Khandelwal (1.22%; citations: 43/ACM,132/G). I cannot find an easy to get the most cited paper at the conference.
