Combining social network analysis and sentiment analysis to explore the potential for online radicalisation

From Brede Wiki
Jump to: navigation, search
Conference paper (help)
Combining social network analysis and sentiment analysis to explore the potential for online radicalisation
Authors: Adam Bermingham, Maura Conway, Lisa McInerney, Neil O'Hare, Alan F. Smeaton
Citation: Combining social network analysis and sentiment analysis to explore the potential for online radicalisation  : 231-236. 2009
Editors:
Publisher: IEEE
Meeting: 2009 International Conference on Advances in Social Network Analysis and Mining
Database(s): Google Scholar cites
DOI: 10.1109/ASONAM.2009.31.
Link(s): http://www.computing.dcu.ie/~nohare/papers/DCU_asonam09.pdf
Search
Web: DuckDuckGo Bing Google Yahoo!Google PDF
Article: Google Scholar PubMed
Restricted: DTU Digital Library
Services
Format: BibTeX

Combining social network analysis and sentiment analysis to explore the potential for online radicalisation reports a text mining study on YouTube comments in relation to radical islamic internet use. They use sentiment analysis and social network analysis.

[edit] Methods

They downloaded 122'011 comments and over 13'000 profiles from YouTube. For sentiment analysis they used SentiWordNet which assigns positivity and negativity to synsets of WordNet. They also used part-of-speech tagging with the Stanford Maximum Entropy Part of Speech Tagger as well as the RitaWN for word stemming. Sentiments was characterized with respect to topics.

In social network analysis they described the betweenness, the network density and the average communication speed with respect to gender.

[edit] Results

Not all YouTube users write their gender.

Of the users that reported their gender females had higher network density and higher average communication speed.

For sentiment analysis they show, e.g., that "Mubarak" has more positivity than negativity.

Personal tools