CYBERBULLYING DETECTION IN SOCIAL NETWORKS USING MULTIMODAL TECHNIQUES

Abstract

The issue of cyberbullying on social media is on the rise, and it is imperative that we develop more effective methods to identify it. To improve results, a multi-modal method combines text, pictures, and user behavior. Computer vision and natural language processing are both capable of detecting incorrect text and images. Machine learning systems can detect patterns indicative of cyberbullying by analyzing user connections. This approach improves detection accuracy by considering a wide range of materials. By absorbing data from their environments, deep learning systems enhance categorization accuracy. Stopping online harassment is a breeze with real-time tracking. In order to put an end to various forms of cyberbullying using this approach, continuous instruction is required. With this strategy, internet users will be better protected. In order to improve detection, future research should make use of state-of-the-art AI systems.