On the usefulness of query features for learning to rank

Learning to rank studies have mostly focused on query-dep-endent and query-independent document features, which en-able the learning of ranking models of increased effective-ness. Modern learning to rank techniques based on regres-sion trees can support query features, which are document-independent, and hence have the same values for all docu-ments being ranked for a query. In doing so, such techniques are able to learn sub-trees that are specific to certain types of query. However, it is unclear which classes of features are useful for learning to rank, as previous studies leveraged anonymised features. In this work, we examine the useful-ness of four classes of query features, based on topic classifi-cation, the history of the query in a query log, the predicted performance of the query, and the presence of concepts such as persons and organisations in the query. Through ex-periments on the ClueWeb09 collection, our results using a state-of-the-art learning to rank technique based on regres-sion trees show that all four classes of query features can significantly improve upon an effective learned model that does not use any query feature