Abstract:
|
There are many examples of dynamic text networks, and some are relevant to national security. This research extends previous work on the analysis of U.S. political blogs, and uses dynamic network modeling to improve topic discovery. A key innovation is a novel model for topic correlation that leverages community detection within the blogosphere. This is pertinent to security in that the methodology enables one to discover groups of like-minded bloggers for whom sentiment analysis shows a trend towards progressive radicalization.
|