Distance Measures In Query Space: How Strongly To Use Feedback From Past Queries

PROCEEDINGS OF THE IEEE/WIC/ACM INTERNATIONAL CONFERENCE ON WEB INTELLIGENCE: WI 2007(2007)

引用 4|浏览0
暂无评分
摘要
Feedback on past queries is a valuable resource for improving retrieval performance on new queries. We introduce a modular approach to incorporating feedback information into given retrieval architectures. We propose to fusion the original ranking with those returned by rerankers, each of which trained on feedback given for a distinct, single query. Here, we examine the basic case of improving a query's original ranking q(test) by only using one reranker: the one trained on feedback on the "closest" query q(train). We examine the use of various distance measures between queries to first identify q(train) and then determine the best linear combination of the original and the reranker's ratings, that is: to find out which feedback to learn from, and how strongly to use it. We show the cosine distance between the term vectors of the two queries, each enriched by representations of the top N originally returned documents, to reliably answer both questions. The fusion performs equally well or better than a) always using only the original ranker or the reranker b) selecting a hard distance threshold to decide between the two, or c) fusioning results with a ratio that is globally optimized, but fixed across all tested queries.
更多
查看译文
关键词
feedback information,original ranker,distance measures,cosine distance,query qtrain,past query,single query,query space,past queries,original ranking,original ranking qtest,hard distance threshold,new query,neurofeedback,web services,ontology,testing,global optimization,semantic web,information retrieval,support vector machines,information extraction,information processing,feedback
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要