• Publications
  • Influence
A support vector method for optimizing average precision
TLDR
We present a general SVM learning algorithm that efficiently finds a globally optimal solution to a straightforward relaxation of MAP that produces statistically significant improvements in MAP scores. Expand
  • 656
  • 97
  • PDF
Learning diverse rankings with multi-armed bandits
TLDR
We present two online learning algorithms that directly learn a diverse ranking of documents based on users' clicking behavior. Expand
  • 401
  • 46
  • PDF
How does clickthrough data reflect retrieval quality?
TLDR
We present a sequence of studies investigating this relationship for an operational search engine on the arXiv.org e-print archive. Expand
  • 287
  • 36
  • PDF
Evaluating the accuracy of implicit feedback from clicks and query reformulations in Web search
TLDR
This article examines the reliability of implicit feedback generated from clickthrough data and query reformulations in World Wide Web (WWW) search. Expand
  • 519
  • 32
  • PDF
Query chains: learning to rank from implicit feedback
TLDR
Using query chains, we generate new types of preference judgments from search engine logs, thus taking advantage of user intelligence in reformulating queries. Expand
  • 549
  • 31
  • PDF
Large-scale validation and analysis of interleaved search evaluation
TLDR
Interleaving is an increasingly popular technique for evaluating information retrieval systems based on implicit user feedback. Expand
  • 165
  • 19
  • PDF
Improving personalized web search using result diversification
TLDR
We present and evaluate methods for diversifying search results to improve personalized web search and evaluate the effectiveness of these methods. Expand
  • 258
  • 17
  • PDF
A Theoretical Framework for Conversational Search
TLDR
This paper studies conversational approaches to information retrieval, presenting a theory and model of information interaction in a chat setting. Expand
  • 138
  • 16
  • PDF
Personalizing web search using long term browsing history
TLDR
We present a personalization approach that builds a user interest profile using users' complete browsing behavior, then uses this model to rerank web results. Expand
  • 194
  • 15
  • PDF
TREC Complex Answer Retrieval Overview
TLDR
This notebook gives an overview of activities, datasets, and results of the second year of TREC Complex Answer Retrieval. Expand
  • 57
  • 15
  • PDF