期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2017
卷号：5
期号：11
页码：16492
DOI：10.15680/IJIRCCE.2017.0511026
出版社：S&S Publications
摘要：In our system, we can learn social circles in ego-networks which are based on multi-view networkstructure. We can classify information about the similar data or similar information. Here we can detect ego-networkbased on social circle. In an automatic social circle detection in ego-networks is a fundamentally important task forsocial network analysis.in this paper, we know,how to detect circles by leveraging multiple views of the networkstructure. For detection of this leveraging multiple views of the network structure, we crawl ego networks from Twitterand model them by six views, including user relationships, user interactions and user content.Friendship is the one viewwhich is used in social circle detection. In this system characterizes the friend relation between alters by a similaritymatrix where alters follow each other on Twitter. It is a most common view for social circle detection. Its only checkthe twitter users follow each other or not but it don’t check the tweets of user. In our system we use SentimentClassification of tweets using NLP (Natural Language Processing). It helps to find the accurate friend relation betweenalerts.We apply multi-view spectral clustering techniques to detect circles on these ego-networks.In this paper we canused a modified multi-view spectral clustering techniques over a single-view clustering methods. We integrate thishow the bound may be affected by several network characteristics.How the different network characteristics affected ona social network.
关键词：Social Circle Detection; Data Crawling; Sentiment Analysis; Multi-View Spectral Clustering
