Font Size: a A A

Research On Tibetan Web Community Discovery Based On Social Network Analysis

Posted on:2013-10-09Degree:MasterType:Thesis
Country:ChinaCandidate:X G ChenFull Text:PDF
GTID:2248330395970830Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
Recently, with the rapid development of Internet, the Internet has become the world’s largest information center as the information amount on the Internet increase larger and larger. The information on the Internet is so complicated that how to analyze its contents and dig out people’s needs has become a hot topic, and community discovery technology provides some solutions to this problem in some ways. It not only saves user’s time, but also improves user’s analysis efficiency. This paper applies the community discovery technique to the Tibetan Web, it provides technical means for us to understand the nature laws in the virtual world of the Tibetan web; Secondly, it makes analysis and statistics of the contents of the Tibetan web, and offers theoretical and technical support for us to understand and master the focuses of public opinions.Social network analysis is a kind of sociological research method. Sociology believes that society is not constituted by individuals but by the networks, the network includes nodes and the relationship between nodes. Social network analysis explores the structure and attributive characters through the analysis of relationships in networks, and achieves improvements to the network. This paper based on social network analysis theory to study on the Tibetan Web community discovery.This paper first takes advantage of the self-developed Web crawler to fetch the Tibetan Web pages, uses HtmlParser and regular expression to extract texts. On the basis of united coding of the texts to recognize entities and extract entity relations. This study makes use of the database storage technology to store entities and relations between entities, and builds topology through the social relationships in social network analysis (including individuals, groups or society) as well as structural analysis and matrix transformations. Through deleting the edges of max intermediary to split the topology, it makes central analysis to the founded community in order to identify the members of high importance.By the end this paper, it generally introduces the deficiencies and next studies need to be done.
Keywords/Search Tags:social network analysis, community discovery, graph cut, entity relation, Tibetan web
PDF Full Text Request
Related items