scispace - formally typeset
Search or ask a question
Topic

Multi-document summarization

About: Multi-document summarization is a research topic. Over the lifetime, 2270 publications have been published within this topic receiving 71850 citations.


Papers
More filters
Journal ArticleDOI
TL;DR: This paper introduces for the first time the features of the human visual system within the summarization framework itself to allow for the emphasis of perceptually significant events while simultaneously eliminating perceptual redundancy from the summaries.
Abstract: The enormous growth of video content in recent times has raised the need to abbreviate the content for human consumption. Thus, there is a need for summaries of a quality that meets the requirements of human users. This also means that the summarization must incorporate the peculiar features of human perception. We present a new framework for video summarization in this paper. Unlike many available summarization algorithms that utilize only statistical redundancy, we introduce for the first time the features of the human visual system within the summarization framework itself to allow for the emphasis of perceptually significant events while simultaneously eliminating perceptual redundancy from the summaries. The subjective and objective evaluation scores have evaluated the framework.

38 citations

Proceedings Article
09 Oct 2010
TL;DR: This paper proposes an A* search algorithm to find the best extractive summary up to a given length, which is both optimal and efficient to run, and proposes a discriminative training algorithm which directly maximises the quality of the best summary.
Abstract: In this paper we address two key challenges for extractive multi-document summarization: the search problem of finding the best scoring summary and the training problem of learning the best model parameters. We propose an A* search algorithm to find the best extractive summary up to a given length, which is both optimal and efficient to run. Further, we propose a discriminative training algorithm which directly maximises the quality of the best summary, rather than assuming a sentence-level decomposition as in earlier work. Our approach leads to significantly better results than earlier techniques across a number of evaluation metrics.

38 citations

Proceedings Article
01 Mar 2003
TL;DR: An ontology database is built for analyzing the main topics of the article andRank the paragraphs based on the relevance between main topics and each individual paragraph to choose desired proportion of paragraphs as summary.
Abstract: In this paper, we compare two methods for article summarization. The first method is mainly based on term-frequency, while the second method is based on ontology. We build an ontology database for analyzing the main topics of the article. After identifying the main topics and determining their relative significance, we rank the paragraphs based on the relevance between main topics and each individual paragraph. Depending on the ranks, we choose desired proportion of paragraphs as summary. Experimental results indicate that both methods offer similar accuracy in their selections of the paragraphs.

38 citations

Proceedings Article
18 Aug 1985
TL;DR: A procedural, rule-based approach is proposed which is implemented in a prototype experimental of System operating in the specific domain of text summarization and utilizes world knowledge on the subject domain contained in an encylopedia.
Abstract: The paper deals with the problem of evaluation importance of descriptive texts and proposes a procedural, rule-based approach which is implemented in a prototype experimental of System operating in the specific domain of text summarization. Importance evaluation is performed through a set of rules wich are used to assign importance values to the different parts of a text and to resolve or explain conflicting evaluations. The system utilizes world knowledge on the subject domain contained in an encylopedia and takes into accont a goal assigned by the user for specifying the pragmatic aspects of the understading activity. In the paper some examples of the system operation are presented by following the evaluation of a small sample text.

38 citations

Proceedings ArticleDOI
28 Oct 2008
TL;DR: This research introduces a method to make extractions based on three factors of Readability, Cohesion and Topic relation to create a summary of text summarization.
Abstract: Currently vast amounts of textual information exist in large repositories such as Web. To processes such a huge amount of information, automatic text summarization has been of great interests. Unlike many approaches which focus on sentence or paragraph extraction, in this research, we introduce a method to make extractions based on three factors of Readability, Cohesion and Topic relation. We use Harmony Search-based sentence selection to make such a summary. Once the summary is created, it is evaluated using a fitness function based on those three factors. The evaluation of the algorithm on a test collection is also presented in the paper. Our results indicate that the extracted summaries by our proposed scheme have better precision and recall than the other approaches.

38 citations


Network Information
Related Topics (5)
Natural language
31.1K papers, 806.8K citations
85% related
Ontology (information science)
57K papers, 869.1K citations
84% related
Web page
50.3K papers, 975.1K citations
83% related
Recurrent neural network
29.2K papers, 890K citations
83% related
Graph (abstract data type)
69.9K papers, 1.2M citations
83% related
Performance
Metrics
No. of papers in the topic in previous years
YearPapers
202374
2022160
202152
202061
201947
201852