Classification of Cyberbullying Detection in Social Networking with Audio using Machine Learning Approach
Bandari Saichandana, Pille Kamakshi · International Journal on Recent and Innovation Trends in Computing and Communication · 2023
Every day, more people use the internet and social media, which leads to an increase in cyberbullying vulnerabilities. By transmitting, posting, and spreading damaging, false, and bad stuff online, it is taking place. For those impacted, it causes psychological and emotional issues. Therefore, the development of automated tools for cyberbullying identification and prevention is essential. Recent research on identifying cyberbullying has largely focused on text-based analysis. The two most significant media in cases of cyberbullying are text and audio. In this paper, a machine learning model for detecting cyberbullying in two types of social data, namely text and audio is presented. This paper is focused on detecting majors form of Cyberbullying: cyberbullying dataset on Twitter and classify them as containing Cyberbullying or not. This paper used datasets, namely, ‘cyberbullying Dataset’. In this paper, the implementation can be done with machine learning algorithms such as logistic regression, naive bayesian classifier, and support vector machine. Also, these three algorithms were compared and evaluated with performance metrics like accuracy, precision, recall, and f1 score. The main aim is to detecting cyberbullying messages on any type of social media platform.