Combatting Cybercrimes: Leveraging Natural Language Processing for Detection in Social Media

Social media platforms facilitate user connection and enable the development and exchange of information via safe, digital channels. The increased level of social connectedness may lead to major physical and emotional trauma as well as online abuse, cyberbullying, harassment, cybercrime, and trolling. Women and children are particularly vulnerable to these types of abuse. The research attempts to derive significant insights from massive volumes of textual data by utilising machine learning, sentiment analysis, and language modelling. It looks for questionable trends, derogatory language, and malicious intent hidden in social media exchanges. The goal of this project is to create an approach that uses natural language processing and machine learning (ML) methods like Random Forest, Naive Bayes, decision trees, and support vector machines to effectively identify bullying and threat posts on social media. An overview of the goals, approaches, and possible contributions of this research to the urgent problem of cybercrime in social media is provided in this abstract.