Sustaining a Peaceful Social Media Environment with Natural Language Processing

Approximately 63% of the total global population are active social media users (Kemp, 2022). Nevertheless, social media platforms like Facebook, Twitter, and TikTok have become breeding grounds for cybercriminals to launch social engineering attacks by exploiting the trust of unsuspecting users and collecting their sensitive data. These attackers can use stolen information to manipulate individuals into revealing further confidential information or performing actions that can compromise their personal or professional security. Therefore, it is essential for users to exercise caution when sharing personal information on social media. Additionally, platforms can use intelligent tools to identify and alert users to potential leaks of sensitive information. This research investigates the feasibility of natural language processing (NLP) for developing such intelligent tools. Social engineering experiments will be performed on a publicly available dataset that contains tweets of the top 1000 Twitter celebrity accounts (Sakib, 2022). Output of this experiment will inform the development of an intelligent tool that can counteract the impact of such attacks. Funding through the RIMES scheme will provide a unique opportunity for the student to develop valuable skills in NLP whilst exploring a tangible facet of the research.

Project Aim
The aim is to assess how well NLP can recognise commonly known information about well-known figures, including their birth date and location, as well as their profession. This critical data will be used to reverse-engineer protection mechanisms that safeguards social media users.

This research was delivered as part of the RIMES project which gives students an opportunity to engage with a range of research environments. All projects were linked to the UN’s sustainable development goals, which provide a blueprint to achieve a better and more sustainable future for all. SustainNET supported the project from inception to completion.

References
Kemp, S. (2022, July 21). Digital 2022 July Global Statshot Report. DataReportal.
Sakib, A. S. (2022). Top 1000 Twitter Celebrity Tweets And Embeddings. Kaggle. https://www.kaggle.com/datasets/ahmedshahriarsakib/top-1000-twitter-celebrity-tweets-embeddings

Student Intern: Sandip Pradhan
Mentors: Dr Nonso Nnamoko and Prof. Yannis Korkontzelos
Award: £2,000