Text REtrieval Conference (TREC) trec7

Presentation to the Text REtrieval Conference (TREC-7) November 09-11, 1998 Gaithersburg, MD, USA
TREC home Publications home Presentations home

Each slide contains one or all of the following navigational symbols at the top of each slide. Please use these symbols to navigate within the presentation. go to first slide go to last slide go back one slide go forward one slide return to this index page view text version of current slide

Overview of the Seventh Text REtrieval Conference TREC-7 Program Committee TREC-7 Track Coordinators TREC-7 Participants TREC Goals TREC-7Ad Hoc Task Sample TREC-7 Topic TREC Topic Lengths Creating Relevance Judgments Assessor picture Overlap of Submitted Results Make-Up of Judgment Pool Unique Contributions to Relevant Set Ad Hoc Task Best Automatic Ad Hoc TREC-6 vs. TREC-7 TREC-7 vs.TREC-6 Topic Difficulty Ad Hoc Topic Lengths Differences in Topic Creation Best Manual Ad Hoc Cross Language Track Number Relevant by Language X to EFGI Results English to English-French Results Filtering Track Filtering Results High Precision Track High Precision Results Interactive Track Interactive Track Results Query Track Query Track Query Track Results SDR Track SDR Track Results VLC Track VLC Track Results Trends in TREC IR Evaluation Methodology IR Evaluation Methodology Retrieval Evaluation 101 Uninterpolated R-P Curve for Single topic Interpolated R-P Curves for Individual Topics Runs Ranked by Different Measures Correlations Between Rankings Evaluating Ranked Retrieval Document Level Measures Number Relevant Average Precision Average Precision Average Precision Tools compare.c Bead Plots Evaluation Methodology
Last updated: Date created: Tuesday, 01-Aug-00 trec@nist.gov

Text REtrieval Conference (TREC-7)