Multi-Documents Summarization Based on TextRank and its Application in Online Argumentation Platform
Authors:
- 1 July 2018
- journal article
- research article
- Published by IGI Global Scientific Publishing in International Journal of Data Warehousing and Mining
- Vol. 14 (3) , 69-89
Abstract
In an Online Argumentation Platform, a great deal of speech messages are produced. To find similar speech texts and extract their common summary is of great significance for improving the efficiency of argumentation and promoting consensus building. In this article, a method of speech text analysis is proposed. Firstly, a heuristic clustering algorithm is used to cluster the speech texts and obtain similar text sets. Then, an improved TextRank algorithm is used to extract a multi-document summary, and the results of the summary are fed back to experts (i.e. participants). The method of multi-document summarization is based on TextRank, which takes into account the position of sentences in paragraphs, the weight of the key sentence, and the length of the sentence. Finally, a prototype system is developed to verify the validity of the method using the four evaluation parameters of recall rate, accuracy rate, F-measure, and user feedback. The experimental results show that the method has a good performance in the system.Keywords
This publication has 35 references indexed in Scilit:
- Enabling Large-Scale Deliberation Using Attention-Mediation MetricsComputer Supported Cooperative Work (CSCW), 2012
- A survey on clustering algorithms for wireless sensor networksInternational Journal of Space-Based and Situated Computing, 2011
- A next generation emerging technologies roadmap for enabling collective computational intelligence in disaster managementInternational Journal of Space-Based and Situated Computing, 2011
- Topic-based web site summarizationInternational Journal of Web Information Systems, 2010
- Adaptive estimated maximum-entropy distribution modelInformation Sciences, 2007
- Multi-document summarization of news articles using an event-based frameworkAslib Proceedings, 2006
- Text clustering and summary techniques for CRM message managementJournal of Enterprise Information Management, 2004
- Computer supported argumentation and collaborative decision making: the HERMES systemInformation Systems, 2001
- Authoritative sources in a hyperlinked environmentJournal of the ACM, 1999
- On the acceptability of arguments and its fundamental role in nonmonotonic reasoning, logic programming and n-person gamesArtificial Intelligence, 1995