This year’s ACM SIGIR Test-of-Time Award went to a paper by Thorsten Joachims, Laura Granka, Bing Pan, Helene Hembrooke, and Geri Gay. "Accurately interpreting clickthrough data as implicit feedback" (2005) was one of the first papers to rigorously explore the behavioral biases in implicit feedback from user behavior, giving insight into how to properly use machine learning methods to learn from such data. In particular, the paper combined insights and experimental methodology from the behavioral social sciences with the theory underlying machine learning algorithms. The award recognizes the works’s impact on information retrieval research, as well as on how search engines and other online systems use machine learning today.
The SIGIR Test of Time Award recognizes research that has had long-lasting influence, including impact on a subarea of information retrieval research, across subareas of information retrieval research, and outside of the information retrieval research community (e.g. non-information retrieval research or industry). The winning paper is selected from the set of full papers presented at the main SIGIR conference 10-12 years before.