Exploring friendships in blogosphere by finding 1.5-cliques

In this work, we show that finding 1.5-cliques is a good idea for exploring friendships in Blogosphere. In todaypsilas blog service environment, friend relationships are usually unidirectionally established, without the confirmation from the other party. Conventional methods find either too small or too few cliques in forming friend groups, since bidirectional links are not completely built under such circumstances. To solve this problem, the idea of finding 1.5-cliques is proposed, and is used for exploring friend groups from a social network, Wretch (www.wretch.cc), as a case study. The effectiveness of this idea is valued from the perspective of the densities of the explored groups. By investigating the data we have been continually collecting in the past year, this method is demonstrated to be pretty good.

[1]  Jennifer Jie Xu,et al.  Mining communities and their relationships in blogs: A study of online hate groups , 2007, Int. J. Hum. Comput. Stud..

[2]  Jiawei Han,et al.  Data Mining: Concepts and Techniques, Second Edition , 2006, The Morgan Kaufmann series in data management systems.

[3]  Wan-Jui Lee,et al.  Finding Friend Groups in Blogosphere , 2008, 22nd International Conference on Advanced Information Networking and Applications - Workshops (aina workshops 2008).

[4]  Stanley Wasserman,et al.  Social Network Analysis: Methods and Applications , 1994 .

[5]  Stanley Wasserman,et al.  Social Network Analysis: Methods and Applications , 1994, Structural analysis in the social sciences.

[6]  Jian Pei,et al.  Data Mining: Concepts and Techniques, 3rd edition , 2006 .

[7]  Qiang Yang,et al.  Latent Friend Mining from Blog Data , 2006, Sixth International Conference on Data Mining (ICDM'06).

[8]  Fengyu Liu,et al.  Web Classification Based on Latent Semantic Indexing , 2006 .