IIT-H algorithm to detect e-bullying in mixed text

  • | Friday | 11th January, 2019

HYDERABAD: Students of the International Institute of Information Technology, Hyderabad have developed machine learning algorithms in English-Hindi mixed coded text to detect cyber-bullying on social media platforms. The language interchange involves complex grammar , and the terms ‘code-switching’ and ‘code-mixing’ are used to describe it. Majority of the work so far in this field has been done for English only. But the presence of multilingual speakers on social media platforms like Twitter , Facebook fills the internet with text which is a mixture of many languages and difficult to tackle using the present models for analysis and classification. While algorithms to detect abusive words in English language already exists, the current work of IIT-H would help in cases where multilingual speakers who switch back and forth between languages when writing informally like Hindi written in Latin script and resort to cyber-bullying.24-year old Vinay Singh of IIIT-H says, “cyberbullying and aggression has increased.

Read Full Article Here

If You Like This Story, Support NYOOOZ



Your support to NYOOOZ will help us to continue create and publish news for and from smaller cities, which also need equal voice as much as citizens living in bigger cities have through mainstream media organizations.

Stay updated with all the Hyderabad Latest News headlines here. For more exclusive & live news updates from all around India, stay connected with NYOOOZ.

Related Articles