Retrieval-oriented summarization

This paper presents an investigation into the utility of document summarization in the context of information retrieval. The investigation explores the use of both context-independent standard summaries and query-biased summaries. We developed a generic, a query-based, and a hybrid summarizer, each with differing amounts of document context. The generic summarizer used a blend of discourse information and information obtained through traditional surface-level analysis. We evaluate these summaries through summary-based retrieval models proposed by author on several TREC collections. We show that it is worse that retrieval documents with summaries directly. However, summary-based retrieval model, smoothing document model with summary model, can perform consistently across collections of realistic size, and significant improvements over document-based retrieval can be obtained in a fully automatic manner and without relevance information provided by human.