top of page

Highly active or connected on social network, hmmm!... could you be cyber bullied? Let's find out by whom, a stranger or your friend?

This research focuses on using social network analysis and content textual features to detect the cyber bullying cases.

 

First, based on related literature from social science and psychology, we built three hypotheses for the differences between cyber bullying and non-bullying cases.

 

Then, by adding 1.5 ego-network graphs from two users in a conversation, we can build and learn the relationship graph of these two users.

 

After gathering the social network data, such as relationship tie, number of friends, and social network centrality, we find that multiple social characteristics are statistically different between the cyber bullying and non-bullying groups, thus supporting the results found in previous psychological studies.

 

Finally, the cyber bullying detection models have been significant improved by adding social network features compared with only using textual features. 

bottom of page