A new rank correlation coefficient for information retrieval

By Emine Yilmaz, J. Aslam, Stephen E. Robertson. Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 2008

298 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Score Aggregation Techniques in Retrieval Experimentation (2009, methodology): “To broaden the scope of our investigation we also plan to explore both further aggregation mechanisms, such as the MEDRANK approach of Fagin et al. (2003); and also other rank correlation approaches, including theτAP topweighted approach of Yilmaz et al. (2008).”
Towards Meaningful Statements in IR Evaluation: Mapping Evaluation Measures to Interval Scales (2021, background): “Ferro [35] has shown that, even if the absolute correlation values are different, removing or not the lower quartile runs produces the same ranking of measures in terms of correlation; similarly, he has shown that both τ and AP correlation τap [105] produce the same ranking of measures in terms of c”
Preferences on a Budget: Prioritizing Document Pairs when Crowdsourcing Relevance Judgments (2022, background): “63 for Kendall’s τ ; we can also see that the effectiveness of the most effective systems is poorly estimated, as confirmed by the low τAP [32] scores.”
Effective User Interaction for High-Recall Retrieval: Less is More (2018, background): “Themostcommonmeasuresofrankingperformance are Kendall’s rank correlation coefficient, τ , and Yilmaz et al.’s τ AP [37].”
Test Collection Based Evaluation of Information Retrieval Systems (2010, background): “[290] produced a new correlation coefficient τap that addresses this failing.”