Less Is More: Maximal Marginal Relevance as a Summarisation Feature

Authors:
Jan Frederik Forst;Anastasios Tombros;Thomas Roelleke
Affiliations:
Department of Computer Science Queen Mary, University of London,;Department of Computer Science Queen Mary, University of London,;Department of Computer Science Queen Mary, University of London,
Venue:
ICTIR '09 Proceedings of the 2nd International Conference on Theory of Information Retrieval: Advances in Information Retrieval Theory
Year:
2009

Citing 3
Cited 0

Advantages of query biased summaries in information retrieval

Proceedings of the 21st annual international ACM SIGIR conference on Research and development in information retrieval
The use of MMR, diversity-based reranking for reordering documents and producing summaries

Proceedings of the 21st annual international ACM SIGIR conference on Research and development in information retrieval
Creating and evaluating multi-document sentence extract summaries

Proceedings of the ninth international conference on Information and knowledge management

Quantified Score

Hi-index	0.00

Visualization

Abstract

Summarisation approaches aim to provide the most salient concepts of a text in a condensed representation. Repetition of extracted material in the generated summary should be avoided. Carbonell and Goldstein proposed Maximal Marginal Relevance as a measure to increase the diversity of documents retrieved by an IR system, and developed a summariser based on MMR. In this paper, we look at the viability of MMR as a feature in the traditional feature-based summarisation approach proposed by Edmundson.