LexRank: Graph-based lexical centrality as salience in text summarization

DMCA

LexRank: Graph-based lexical centrality as salience in text summarization (2004)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Dragomir R. Radev

Venue:	Journal of Artificial Intelligence Research
Citations:	266 - 9 self

BibTeX

@ARTICLE{Radev04lexrank:graph-based,
    author = {Dragomir R. Radev},
    title = {LexRank: Graph-based lexical centrality as salience in text summarization},
    journal = {Journal of Artificial Intelligence Research},
    year = {2004},
    pages = {457--479}
}

OpenURL

Abstract

We introduce a stochastic graph-based method for computing relative importance of textual units for Natural Language Processing. We test the technique on the problem of Text Summarization (TS). Extractive TS relies on the concept of sentence salience to identify the most important sentences in a document or set of documents. Salience is typically dened in terms of the presence of particular important words or in terms of similarity to a centroid pseudo-sentence. We consider a new approach, LexRank, for computing sentence importance based on the concept of eigenvector centrality in a graph representation of sentences. In this model, a connectivity matrix based on intra-sentence cosine similarity is used as the adjacency matrix of the graph representation of sentences. Our system, based on LexRank ranked in rst place in more than one task in the recent DUC 2004 evaluation. In this paper we present a detailed analysis of our approach and apply it to a larger data set including data from earlier DUC evaluations. We discuss several methods to compute centrality using the similarity graph. The results show that degree-based methods (including LexRank) outperform both centroid-based methods and other systems participating in DUC in most of the cases. Furthermore, the LexRank with threshold method outperforms the other degree-based techniques including continuous LexRank. We also show that our approach is quite insensitive to the noise in the data that may result from an imperfect topical clustering of documents. 1.

Keyphrases

text summarization graph-based lexical centrality graph representation sentence importance threshold method degree-based method rst place imperfect topical clustering recent duc textual unit natural language processing particular important word relative importance sentence salience stochastic graph-based method continuous lexrank duc evaluation centroid pseudo-sentence connectivity matrix degree-based technique eigenvector centrality adjacency matrix important sentence centroid-based method extractive t relies new approach detailed analysis intra-sentence cosine similarity several method similarity graph

DMCA