《海量数据挖掘-王永利》Temporal Feedback for Tweet Search with Non-Parametric Density Estimation.pdfVIP

Temporal Feedback for Tweet Search with Non-Parametric Density Estimation Miles Efron1, Jimmy Lin2, Jiyin He3, and Arjen de Vries3 1 Graduate School of Library and Information Science, University of Illinois, Urbana-Champaign 2 The iSchool, University of Maryland, College Park 3 Centrum Wiskunde Informatica, Amsterdam, The Netherlands mefron@, jimmylin@, jiyinhe@, arjen@ ABSTRACT This paper investigates the temporal cluster hypothesis: in search tasks where time plays an important role, do relevant documents tend to cluster together in time? We explore this question in the context of tweet search and temporal feedback: starting with an initial set of results from a baseline retrieval model, we estimate the temporal density of relevant documents, which is then used for result reranking. Our contributions lie in a method to characterize this temporal density function using kernel density estimation, with and without human relevance judgments, and an approach to integrating this information into a standard retrieval model. Experiments on TREC datasets con?rm that our temporal feedback formulation improves search e?ectiveness, thus providing support for our hypothesis. Our approach outperforms both a standard baseline and previous temporal retrieval models. Temporal feedback improves over standard lexical feedback (with and without human judgments), illustrating that temporal relevance signals exist independently of document content. Categories and Subject Descriptors: H.3.3 [Information Storage and Retrieval]: Information Search and Retrieval— Relevance feedback Keywords: temporal clustering; cluster hypothesis; relevance feedback 1. INTRODUCTION Twitter has become an indispensable communications platform through which hundreds of millions of users around the world witness breaking news events. They can participate in the global conversation in real time, 140 characters at a time. To access relevant content in microblogs, people often turn to search. And naturally, ti