HYDERABAD: With uncivil behaviours and hate speech becoming a norm on the internet, a team from IIIT-Hyderabad has developed an automated system using Artificial Intelligence chatbots that can detect hate speech in tweets.
Developed at IIIT-Hyderabad’s Informational and Retrieval Extraction Lab (IREL) by Vasudeva Varma, professor and dean (R&D), his students Pinkesh Badjatiya, Shashank Gupta and adjunct faculty Manish Gupta. The project took close to a year to materialise.
The project was presented at the International World Wide Web (WWW) 2017 at Perth earlier this month where their poster on ‘Deep Learning for Hate Speech Detection in Twitter’ was voted as the best poster presentation among 166 submissions from around the world.
WWW is one of the most prestigious web-conference where several policies regarding the Internet are formulated.
“The conference is attended by academia and research wings of large companies like Facebook, Google, Microsoft, etc. The fact that we’ve been able to make it through the rigorous review process to have our poster presentation voted as the best is a great validation of our research work,” said Prof Varma.
As of now, they are able to detect abusive language, sexist and racist speech and flag offensive content. This not just helps in automatically filtering such content, but also in analysing public sentiment to get to the root of the problem through user-generated content.
How does it work?
To detect hate speech, they use a popular approach in machine learning called ‘supervised learning’. Essentially a computer algorithm is fed many examples of text from each form of hate, which can be categorised as ‘racist’ or ‘sexist’ tweets.
The algorithm is designed in such a way that it learns as it sees the data and after the algorithm terminates, the program recognises racism or sexism in the text, if there is one. The algorithm uses neural networks, popularly called ‘Deep Learning’. Inspired by the human brain, these algorithms try to simulate how humans learn from examples.