Query type classification using query history
Chen Hongtao, Hua Zhou, Yang Fangchun · China-Ireland International Conference on Information and Communications Technologies (CIICT 2007) · 2007
It is crucial for search engines to discriminate the query type and discard the irrelevant Web pages in the search result. In this paper, a query type classification is proposed in two processes. First, a Web classifier is recommended using Naive Bayes Network to classify all the Web pages. Second, a query type classifier is trained by calculating the classification probabilities of all the clicked Web pages of each of the query words in the training data set. Considering that most history query words within a search session may express the users' searching goal, a query modification model is proposed and the queries are extended by the history query words to improve the precision of our classification algorithm.