LexRank: Graph-based Centrality as Salience in Text Summarization

By Günes Erkan, Dragomir R. Radev. 2004

317 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/8937f6ecd5e19b85c2875bad3e10e320fd15623b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Self reinforcement for important passage retrieval (2013, methodology): “tant passage retrieval/summarization area, models like TextRank [18] or LexRank [5] are still used as baselines or as the foundation of state-of-the-art work [3, 28].”
Scientific Paper Summarization Using Citation Summary Networks (2008, methodology): “Lexrank is proved to work well in multi-document summarization (Erkan and Radev, 2004).”
A Survey of Text Summarization Techniques (2012, methodology): “Graph-based approaches have been shown to work well for both singledocument and multi-document summarization [25, 61].”
Multiple documents summarization based on evolutionary optimization algorithm (2013, methodology): “LexRank (Erkan & Radev, 2004): The LexRank defines sentence salience based on graph-based centrality scoring of sentences.”
Benchmarking Large Language Models for News Summarization (2023, background): “CNN/DM includes articles from the CNN and DailyMail websites as the source article and adapt the bullet point highlights that come with the website articles as reference summaries.”
A Supervised Approach to Extractive Summarisation of Scientific Papers (2017, methodology): “outperform these methods, including graph-based methods which take account of global context: LexRank (Radev, 2004) and TextRank (Mihalcea and Tarau, 2004); probabilistic methods in KLSum (KL divergence summariser, Haghighi and Vanderwende (2009)); methods based on singular value decomposition with ”
Random-Walk Term Weighting for Improved Text Classification (2007, unknown): “While there are several random-walk algorithms that have been proposed in the past, we focus on only one such algorithm, namely PageRank [2], as it was previously found successful in a number of applications, including Web link analysis [2], social networks [8], citation analysis, and more recently ”
Dependency-Based Construction of Semantic Space Models (2007, background): “Examples include contextual spelling correction (Jones and Martin 1997), summarization (Erkan and Radev 2004; Barzilay 2003) and question answering (Lin and Pantel 2001).”
The Impact of Frequency on Summarization (2005, methodology): “Table 7 shows the scores of summaries produced by SumBasic, LexRank, and DEMS, just based on initial sentence weighting (with no re-ranking for SumBasic or duplication removal for LexRank or DEMS), as well as the scores for the complete summarizers with duplication removal.”
Learning Attitudes and Attributes from Multi-aspect Reviews (2012, background): “Matrix factorization techniques [18] use nothing but numeric scores to learn latent ‘aspects’ that best explain users’ preferences, while clustering and topic-modeling approaches use nothing but review text to model ‘aspects’ in terms of words that appear in document corpora [4], [9], [11].”
Enhancing Single-Document Summarization by Combining RankNet and Third-Party Sources (2007, methodology): “Graph-based methods have also been employed (Erkan and Radev, 2004a; Erkan and Radev, 2004b; Mihalcea, 2005; Mihalcea and Tarau, 2005; Mihalcea and Radev, 2006).”
A Survey of Text Summarization Extractive Techniques (2010, background): “An Abstractive summarization [32][33] attempts to develop an understanding of the main concepts in a document and then express those concepts in clear natural language.”
Automatic summarising: The state of the art (2007, background): “2003; Farzinder and Lapalme 2004), magazine articles (Moens and Dumortier 2000), and technical articles (McKeown et al. 1998, Elhadad and McKeown 2001; Saggion and Lapalme 2000, 2002). Web pages, source in Radev et al. (2001a), have complex structures with items that may need excluding (Berger and M”
Random-Walk Term Weighting for Improved Text Classification (2006, methodology): “While there are several random-walk algorithms that have been proposed in the past, we focus on only one such algorithm, namely PageRank [2], as it was previously found successful in a number of applications, including Web link analysis [2], social networks [8], citation analysis, and m ore recently”
Fill in the BLANC: Human-free quality estimation of document summaries (2020, methodology): “model (based on (Erkan and Radev, 2004)).”