In this thesis, we examine the problem of mining antagonistic communities from social networks. In social networks, people with opposite opinions normally behave differently and form sub-communities each of which containing people sharing some common behaviors. In one scenario, people with opposite opinions show differences in their views on a set of items. Another scenario is people explicitly expressing whom they agree with, like or trust as well as whom they disagree with, dislike or distrust. We defined the indirect and direct antagonistic groups based on the two scenarios. We have developed algorithms to mine the two types of antagonistic groups. For indirect antagonistic group mining, our algorithm explores the search space of all the possible antagonistic groups starting from antagonistic groups of size two, followed by searching antagonistic groups of larger sizes. We have also developed a divide and conquer strategy to ensure our algorithm runs on large databases. We have conducted experiments on both synthetic datasets and real datasets. The results show that our approach can efficiently compute indirect antagonistic groups. Our experiments on the four real datasets show the utility of our work in extracting interesting information from real datasets. For direct antagonistic group mining, we have combined several existing algorithm blocks to mine the patterns. We found significant differences in behaviors of users showing antagonistic relationships and those showing friendship relationships.
|School:||Singapore Management University (Singapore)|
|Department:||School of Information Systems|
|School Location:||Republic of Singapore|
|Source:||MAI 49/04M, Masters Abstracts International|
|Keywords:||Antagonistic community, Apriori, Community finding, Community mining, Frequent itemset mining, Social network|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be