In a world when youngsters are connected to social medial all the time, it becomes difficult to insulate oneself from online vitriolic content. But now using latest technology, researchers and faculty at the International Institute of Information Technology-Hyderabad are working on a solution that seeks to protect them from such toxicity.
As a part of Project Angel, Vasudeva Varma from IIIT-H’s Information Retrieval and Extraction Lab, says researchers spoke to several female respondents across cities. The work also entailed online surveys, among other information gathering techniques. The idea, he says, is to use artificial intelligence, machine learning and neuro-lingusitic programming to trawl social media, scrape certain words, phrases and sentences, which are akin to bullying, and protect users from it.
Derogatory comments
“Spending a lot of time online takes away productivity and the joys of regular life offline. In addition to that, online world is toxic as people here become much more aggressive. One kind of toxic behaviour online is body shaming. It’s one the biggest reasons why girls get into depression. If somebody makes derogatory comments on their body, it is of course a sensitive matter for them,” Prof. Varma says.
It is a natural course of action for IIIT-H to embark on Project Angel. For, in the past, the premier institution has developed a technology that detects hate speech.
“This ‘Angel’ will sit on your social network – be it Instagram, Twitter or Facebook – and will understand what content you a producing, and the content you are consuming. The user has to ‘follow’ the ‘Angel’. It is an intelligent software that understands content. It will try to estimate if you’re getting into any ‘trouble’, meaning it can detect whether anybody is harassing you or body shaming or putting you down,” Prof. Varma explains.
The first phase entails the Angel ‘listening’ to the conversation that improves its understanding of the content. The second deals with the right interventions at the right time and would start communicating with the user, discouraging her from further engagement with a toxic message.
The third phase involves developing an online social network in which those in trouble can connect with counsellors or those who have dealt with online toxicity. “We are focusing on Instagram where this ‘Angel’ will be rolled out as it is a popular platform among teenage girls,” says Prof. Varma. The Angel will then be launched on other social media platforms such as Twitter and Facebook.
