Students' Projects

Rumour Detection From Social Media
Department Electronics & Communication
Year 2020-21
Budget 14000
Students Nisarg Trivedi, Meet Velani
Mentor Prof K. R. Trivedi, Prof. Rahul Patel
Abstract In the Era of Digital Media, majority of people gather news online. Earlier newspaper and TV news channels were the main source of news events, but now with the increasing and easy use of internet in mobile people easily get news online faster than other sources. With use of internet in mobile, social media like twitter, Facebook, or whatsapp are the main platform used by almost all mobile users. On social networks everybody is free to obtain and share information, anywhere at any time. So, breaking news spread very fast in social media. With breaking news, sometimes rumour also spread quickly in social media i.e., information that are unverified at the time of posting, which may causes serious damage to the government, markets, society and Stakeholders also. Therefore, there is necessity of effective system for detecting rumours as early as possible before they widely spread. Effective system should consist of four components: Rumour detection, rumour tracking, stance classification, and veracity classification. Lots of work has been done in later component while very less work in component rumour detection. So, now we should work on rumour detection. Most of existing methods detects a priori rumours, i.e., predefined rumours. So it is required to have automated rumour detection method which detects new emerging rumours effectively and as early as possible.