Paper: Query classification using topic models and support vector machine

ACL ID W12-3304
Title Query classification using topic models and support vector machine
Venue Annual Meeting of the Association of Computational Linguistics
Session Student Session
Year 2012
Authors

This paper describes a query classification system for a specialized domain. We take as a case study queries asked to a search engine of an art, cultural and history library and clas- sify them against the library cataloguing cate- gories. We show how click-through links, i.e., the links that a user clicks after submitting a query, can be exploited for extracting informa- tion useful to enrich the query as well as for creating the training set for a machine learn- ing based classifier. Moreover, we show how Topic Model can be exploited to further enrich the query with hidden topics induced from the library meta-data. The experimental evalua- tions show that this system considerably out- performs a matching and ranking classifica- tion approach, where queries (and categories) were also enri...