作者: Kevyn Breca Collins-Thompson , Paul Nathan Bennett , Karthik Raman
DOI:
关键词: Search engine 、 Classifier (UML) 、 Web query classification 、 Computer science 、 Query expansion 、 Information retrieval
摘要: Various technologies described herein pertain to predicting intrinsically diverse sessions and retrieving information for such sessions. Search results retrieved by a search engine responsive executing query are received. A classifier can be employed determine whether the is or not based on one more features of session interaction properties. The when included in an directed towards task, where disparate queries respective subtasks task. An objective function evaluated at least upon compute optimized value determined diverse. presented display screen according