HateMM: A Multi-modal Dataset for Hate Video Classification
Authors/Creators
- 1. Indian Institute of Technology Kharagpur
- 2. Microsoft, India
Description
Hate speech has become one of the most significant issues in modern society, with implications in both the online and offline worlds. However, most of the work has primarily focused on text media, with relatively little work on images and even less on videos. Thus, early-stage automated video moderation techniques are needed to handle the videos that are being uploaded to keep the platform safe and healthy. Therefore, we curated approximately ~43 hours of videos from BitChute and manually annotated them as hate or non-hate, along with the frame spans that could explain the labeling decision.
This upload contains the labeled video dataset crawled from BitChute, as described in the paper 'HateMM: A Multi-modal Dataset for Hate Video Classification'.