Presentation Title

Temporal Analysis of Radical Dark Web Forum Users

Presenter Information

Patrick Lam
Brian Beck
Darrick Fletcher

Location

House of Learning Library, 3rd floor

Start Date

18-3-2016 12:00 PM

End Date

18-3-2016 6:00 PM

Abstract

Extremist groups have turned to the internet and social media as a means of sharing information amongst one another. This research project analyses forum posts, and finds people who show radical tendencies through the use of natural language processing and sentiment analysis. The forum data being used is from seven Islamic forums on the dark web which are made available for security research. The research project uses OpenNLP as a POS (Parts of Speech) tagger. The POS tagger isolates keywords and nouns that can be utilized with the sentiment analysis program, SentiStrength, to determine polarity of the post. The post is scored as either positive or negative. Each user is assigned a “radical score” which is composed of the following: average sentiment, volume of negative posts, severity of negative posts, and duration of negative postings. These scores are then divided into monthly radical scores for each user. Once these time clusters are mapped, the change in opinions of the users may be interpreted as rising or falling levels of radicalism. Each user is then compared on a timeline to other radical users and events to determine possible connections or relationships. This project attempts to visualize such connections or relationships with radical tendencies so that big data of forum posts can be understood intuitively. By visualizing a forum for an overall change in attitude, unrest and possible radical actions or terrorism can be detected.

Department

Computing Science

Faculty Advisor

Andrew Park

This document is currently not available here.

Share

COinS
 
Mar 18th, 12:00 PM Mar 18th, 6:00 PM

Temporal Analysis of Radical Dark Web Forum Users

House of Learning Library, 3rd floor

Extremist groups have turned to the internet and social media as a means of sharing information amongst one another. This research project analyses forum posts, and finds people who show radical tendencies through the use of natural language processing and sentiment analysis. The forum data being used is from seven Islamic forums on the dark web which are made available for security research. The research project uses OpenNLP as a POS (Parts of Speech) tagger. The POS tagger isolates keywords and nouns that can be utilized with the sentiment analysis program, SentiStrength, to determine polarity of the post. The post is scored as either positive or negative. Each user is assigned a “radical score” which is composed of the following: average sentiment, volume of negative posts, severity of negative posts, and duration of negative postings. These scores are then divided into monthly radical scores for each user. Once these time clusters are mapped, the change in opinions of the users may be interpreted as rising or falling levels of radicalism. Each user is then compared on a timeline to other radical users and events to determine possible connections or relationships. This project attempts to visualize such connections or relationships with radical tendencies so that big data of forum posts can be understood intuitively. By visualizing a forum for an overall change in attitude, unrest and possible radical actions or terrorism can be detected.