The tool uses an artificial intelligence (AI) algorithm to extract critical information from social media platforms. This information can then be cross-checked to determine the authenticity of the social media post. The tool will help pass on critical data to aid the rescue and relief operations in the event of a disaster. The solution can detect fake news and even alert users during the time of disasters through deep learning algorithms.
Currently, the algorithm is limited only to disasters in terms of identifying fake news or rumors. The accuracy level of this AI algorithm developed by IIT students is an impressive 90%. The report states that only 2% of the tweets mined by the algorithm contained information pertinent to disasters. The algorithm mines these tweets and sends the relevant posts to relief operations in the disaster zones.
The Qatar Computing Research Institute (QCRI) has worked with IIT Kharagpur students on the project. The solution can extract critical information from social media platforms like Facebook, Twitter, and even instant chat messaging platforms like WhatsApp. The tool also specifies the type of critical and actionable information such as resource needs, availability, and information of affected people.
Saptarshi Ghosh, Assistant Professor in the Department of Computer Science and Engineering and IIT Kharagpur heads the team that has developed this AI tool. The project is supported by IIT Kharagpur’s Institute Scheme for Innovative Research and Development grant, Microsoft Research India and ITRA, Media Labs Asia, and Department of Electronics and Information Technology.
The institute has also proposed a research project to the Department of Science and Technology of IIT Bombay. The team plans to develop a web-based system to aid post disaster relief operations. The IIT-K started with a pilot on Twitter and WhatsApp during the 2015 Nepal Earthquake and the 2016 Chennai floods and demonstrated an effective use case of this tool.