Towards CST-Enhanced Summarization

In this paper, we propose to enhance the process of automatic extractive multi-document text summarization by taking into account cross-document structural relationships as posited in Cross-document Structure Theory (CST). An arbitrary multidocument extract can be CST-enhanced by replacing lowsalience sentences with other sentences that increase the total number of CST relationships included in the summary. We show that CST-enhanced summaries outperform their unmodified counterparts using the relative utility evaluation metric. We also show that the effect of a CST relationship on an extract depends on its type.

Authors

• Zhu Zhang
• and Dragomir R. Radev University of Michigan

Keywords

No keywords are indexed for this paper.