A study of statistical query expansion strategies for sentence retrieval
David E. Losada · 2008
retrieval of sentences that are relevant to a given in- formation need is a challenging passage retrieval task. In this context, the well-known vocabulary mismatch prob- lem, present in most Information Retrieval processes, arises severely because of the fine granularity of the task. Short queries, which are usually the rule rather than the excep- tion, come to aggravate the problem. Consequently, effec- tive sentence retrieval methods tend to apply some form of query expansion, usually based on pseudo-relevance feed- back. Nevertheless, there are no extensive studies compar- ing different expansion strategies for sentence retrieval prob- lems. In this work we aim to fill this gap. We start from a set of retrieved documents in which relevant sentences have to be found. In our experiments we test different term selec- tion strategies and we also check whether expansion before sentence retrieval can yield reasonable performance. This is particularly novel because expansion techniques for sentence retrieval are often applied after a first retrieval of sentences and there are no comparative results available between ex- pansion before and after sentence retrieval. This compari- son is valuable not only for testing distinct expansion-based methods but also because there are important implications in time efficiency.