Analyzing Domestic Violence and Online Harassment During COVID19

In this project, award-winning Red Dot Foundation partnered with Omdena to apply various AI and NLP tools to better understand domestic violence and online harassment trends during COVID19.

The problem: A rise in domestic violence

COVID19 has shown a marked increase in cases of domestic violence and online harassment. Data substantiating domestic violence from government resources are only available in summary form. Incidents increases are largely reported via calls and hence making data and subsequent mapping difficult. Also, online harassment is on the rise and that information is available anecdotally. Both problems need urgent attention!

Work from Omdena’s AI again domestic violence project

The solutions

Based on initial social media and keyword searches provided by Red Dot, Omdena’s collaborators built an NLP classifier model by mining thousands of multi-lingual posts from Twitter, Reddit, YouTube, and news articles as well as government reports and Google trends. The team delivered the model, a labeled dataset, and Tableau workbook.

The impact

With this additional data, Red Dot was able to file public interest litigation with the Supreme Court of India. Their case was referred to the Ministry of Home Affairs and Omdena’s results formed part of the evidence base for their request to declare DV services as essential and issue guidelines.