- Sputnik International
World
Get the latest news from around the world, live coverage, off-beat stories, features and analysis.

US Government Develops System To Automatically Detect Cyberbullying

© AP Photo / David GoldmanChris Boston looks over a screen shot on his computer of the phony Facebook account that was set up in his daughter Alex's name Thursday, April 26, 2012, at their home in Acworth, Ga.
Chris Boston looks over a screen shot on his computer of the phony Facebook account that was set up in his daughter Alex's name Thursday, April 26, 2012, at their home in Acworth, Ga. - Sputnik International
Subscribe
A grant of around $120,000 was granted to Rutgers University by the National Science Foundation to create a system that automatically detects hurtful online speech.

“This project aims to define new approaches for automatic detection of cyberbullying by integrating the relevant research in social sciences and computer science,” the grant reads.

According to the NSF, cyberbullying has become a “critical social problem, with over 40% of teenagers in the US having reported being cyberbullied,” leading many to psychiatric and psychosomatic disorders.

The idea is that the system will to search for keywords as well as study the social relationships between teenagers who send and receive mean messages.

The program also aims at “employing social intervention mechanisms” that help prevent future cyberbullying activities.

Illinois schools and universities can demand students’ online passwords and punish them for bullying outside of school hours, under HB 4207. - Sputnik International
Illinois Law Demands Students Hand Over Social Media Passwords

Data on cyberbullying will also “be made available to the larger research community,” according to the grant.

The project will start in July and is set to last for two years, with the goal of improving cyberbullying detectors.

“By analyzing the social relationship graph between users and deriving features such as number of friends, network embeddedness, and relationship centrality, the project will validate (and potentially refine) multiple theories in social science literature and assimilate those findings to create better cyberbullying detectors,” the grant said. 

“The project will yield new, comprehensive models and algorithms that can be used for cyberbullying detection in automated settings.”

Assistant Professor Vivek K. Singh of Rutgers’ School of Communication and Information, who has done research on Twitter hashtags, will be leading the project.

The Obama administration has made tackling cyberbullying a priority, and so has the Department of Health and Human Services (HHS), which runs a website to stop cyberbullying that encourages reporting mean online behavior to the police.

Newsfeed
0
To participate in the discussion
log in or register
loader
Chats
Заголовок открываемого материала