The Journal of Machine Learning Research
Knowledge sharing in virtual communities – a review of the empirical research
International Journal of Web Based Communities
Dark web forums portal: searching and analyzing Jihadist forums
ISI'09 Proceedings of the 2009 IEEE international conference on Intelligence and security informatics
Topic-based social network analysis for virtual communities of interests in the Dark Web
ACM SIGKDD Workshop on Intelligence and Security Informatics
ICDMW '11 Proceedings of the 2011 IEEE 11th International Conference on Data Mining Workshops
Collecting and analyzing the presence of terrorists on the web: a case study of jihad websites
ISI'05 Proceedings of the 2005 IEEE international conference on Intelligence and Security Informatics
Extending market basket analysis with graph mining techniques: A real case
Expert Systems with Applications: An International Journal
Leveraging social network analysis with topic models and the Semantic Web extended
Web Intelligence and Agent Systems - Web Intelligence and Communities
| Hi-index | 0.00 |
A hot research topic is the study and monitoring of online communities. Of course, homeland security institutions from many countries are using data mining techniques to perform this task, aiming to anticipate and avoid a possible menace to local peace. Tools such as social networks analysis and text mining have contributed to the understanding of these kinds of groups in order to develop counter-terrorism applications. A key application is the discovery of sub-communities of interests which main topic could be a possible homeland security threat. However, most algorithms detect disjoint communities, which means that every community member belongs to a single community. Thus, final conclusions can be omitting valuable information which leads to wrong results interpretations. In this paper, we propose a novel approach to combine traditional network analysis methods for overlapping community detection with topic-model based text mining techniques. Afterwards, we developed a sub-community detection algorithm that allow each member belong more than one sub-community. Experiments were performed using an English language based forum available in the Dark Web portal (islamicAwakening).