Information Retrieval and Search Behavior

Information retrieval research examines how systems find and rank documents in response to human queries, drawing on signals ranging from the statistical patterns in text to the clicks users leave behind as implicit judgments of relevance. Getting this right matters enormously: the difference between a useful and a frustrating search experience often comes down to subtle choices in how a system models language, weighs evidence, or interprets ambiguous intent. Active work is pushing on how large language models can serve as richer representations of both documents and queries, and on how behavioral signals like clickthrough data can be used to train ranking algorithms without requiring expensive human annotations. A persistent open question is how to evaluate retrieval quality fairly when user needs are diverse, context-dependent, and often only partially expressed in the query itself.

Works

37,630

Total citations

388,677

Keywords

Information RetrievalSearch EnginesUser BehaviorLearning to RankQuery AnalysisRelevance Feedback