The Automated Estimation of Content-Terms for Query-Focused Multi-document Summarization

Query-focused multi-document summarization aims to produce a summary in response to a user query. We present an approach based on estimation of content-terms to address this task. In the process of estimating content-terms, we make full use of the relevant feature and the information richness feature for assigning importance to each of them. With summary content-terms being identified correctly, the candidate sentences are ranked and best sentences are selected to form the summary. Experiments on DUC 2005 and 2006 are performed and the ROUGH evaluation results show that the proposed approach is encouraging.