Usenet is a world-wide distributed discussion system, and it is one of the representative resources on Internet. The structure of news-group on Usenet forms gradually along with the evolution of the news-group and could provide helpful information for the users. In this paper, we present a method to evaluate the impact factors of participators and threads based on the structure of newsgroup. Some analysis and experimental results on real data sets are also provided. The impact factors could provide useful intrinsic information of the newsgroup and can be used in some applications to help access information more efficiently on the Usenet. The method can be also applied on other discussion systems, such as web forums, bulletin board systems, and so on.
[1]
Michael F. Schwartz,et al.
Discovering shared interests using graph analysis
,
1993,
CACM.
[2]
Carl D. Meyer,et al.
Deeper Inside PageRank
,
2004,
Internet Math..
[3]
Jon Kleinberg,et al.
Authoritative sources in a hyperlinked environment
,
1999,
SODA '98.
[4]
Amin Saberi,et al.
Exploring the community structure of newsgroups
,
2004,
KDD.
[5]
Ramakrishnan Srikant,et al.
Mining newsgroups using networks arising from social behavior
,
2003,
WWW '03.
[6]
Matthew Richardson,et al.
Mining the network value of customers
,
2001,
KDD '01.