Term-weighting approaches in automatic text retrieval
Information Processing and Management: an International Journal
Machine Learning
Clustering user queries of a search engine
Proceedings of the 10th international conference on World Wide Web
IEEE Internet Computing
ACM SIGIR Forum
Support Vector Machines for Text Categorization
HICSS '03 Proceedings of the 36th Annual Hawaii International Conference on System Sciences (HICSS'03) - Track 4 - Volume 4
Query type classification for web document retrieval
Proceedings of the 26th annual international ACM SIGIR conference on Research and development in informaion retrieval
Understanding user goals in web search
Proceedings of the 13th international conference on World Wide Web
Evaluating implicit measures to improve web search
ACM Transactions on Information Systems (TOIS)
Automatic identification of user goals in Web search
WWW '05 Proceedings of the 14th international conference on World Wide Web
An introduction to ROC analysis
Pattern Recognition Letters - Special issue: ROC analysis in pattern recognition
Determining the informational, navigational, and transactional intent of Web queries
Information Processing and Management: an International Journal
Automatic query type identification based on click through information
AIRS'06 Proceedings of the Third Asia conference on Information Retrieval Technology
The intention behind web queries
SPIRE'06 Proceedings of the 13th international conference on String Processing and Information Retrieval
Deriving query intents from web search engine queries
Journal of the American Society for Information Science and Technology
Hi-index | 0.00 |
In this paper we introduce a high-precision query classification method to identify the intent of a user query given that it has been seen in the past based on informational, navigational, and transactional categorization. We propose using three vector representations of queries which, using support vector machines, allow past queries to be classified by user's intents. The queries have been represented as vectors using two factors drawn from click-through data: the time users take to review the documents they select and the popularity (quantity of preferences) of the selected documents. Experimental results show that time is the factor that yields higher precision in classification. The experiments shown in this work illustrate that the proposed classifiers can effectively identify the intent of past queries with high-precision.