Revisions that improve cohesion in multi-document summaries

Abstract

Extractive summaries produced from multiple source documents suffer from an array of problems with respect to text cohesion. In this preliminary study, we seek to understand what problems occur in such summaries and how often. We present an analysis of a small corpus of manually revised summaries and discuss the feasibility of making such repairs automatically. Additionally, we present a taxonomy of the problems that occur in the corpus, as well as the operators which, when applied to the summaries, can address these concerns. This study represents a first step toward identifying and automating revision operators that could work with current summarization systems in order to repair cohesion problems in multidocument summaries.

Generating Referring Expressions in Context: The GREC Task Evaluation Challenges

“…under the heading of syntax-based statistical mt as in work by Knight and colleagues [6,13]), and researchers in extractive summarisation have started to develop post-processing text regeneration techniques to improve the coherence and clarity of summaries [27,34,25].…”

Building a Language Model for Local Coherence in Multi-document Summaries Using a Discourse-Enriched Entity-Based Model

“…As stated in [3], some of these multi-document phenomena (that must be avoided or properly treated in summaries) are redundancy, complementarity and contradiction, which may happen because information from different source texts may be used to compose the summary. In several MDS approaches [1,3,20,23,25], the CST (Cross-document Structure Theory) [22] discourse model has been successfully applied to deal with these issues.…”

Automatic Summarization from Multiple Documents

“…In [ORL02] we find a very interesting taxonomy of reformulation and revision strategies, based on the pragmatic (i.e. actual) concerns these strategies aim to tackle.…”

“…One should not ignore the reformulation aspect of temporal ordering, which is a serious issue especially in multi-document summarization [ORL02,Rad00].…”

“…Otterbacher et al uses these concerns as a guide to problems where reformulation and revision can be applied [ORL02]. Thus, the aforementioned list of concerns indicates the set of problems reformulation should address.…”