Ranking expansion terms with partial and ostensive evidence

In this paper we examine the problem of ranking candidate expansion terms for query expansion. We show, by an extension to the traditional F4 scheme, how partial relevance assessments (how relevant a document is) and ostensive evidence (when a document was assessed relevant) can be incorporated into a term ranking function. We then investigate this new term ranking function in three user experiments, examining the performance of our function for automatic and interactive query expansion. We show that the new function not only suggests terms that are preferred by searchers but suggests terms that can lead to more use of expansion terms.