Tackling Cyberbullying on Social Media a Machine Learning for Effective Detection

Over the past few years, social media and online social networks (OSN) have seen a sharp rise in popularity. However, the main issues with social media sites and online networks are security and privacy. However, attention must be paid to the grave issue of cyberbullying (CB) on social media platforms. An intentional, tenacious, and forceful reaction The phrase "cyberbullying" (CB) refers to behaviors on information and communication technology (ICT) platforms, such as social media, the internet, and mobile devices. Deep learning (DL) methods are needed for the identification and classification of CB in social networks in order to counter this trend. An novel method that combines deep learning and feature subset selection for social networks is called feature subset selection with deep learning based CB detection and categorization (FSSDL-CBDC). Three steps make up the suggested FSSDL-CBDC method: preprocessing, classification, and feature selection. combining a system for automatically filtering social media accounts with a A deep learning method system for automatically identifying and classifying cyberbullying on social media platforms is given in this study. Proactive steps to protect user safety and wellness are becoming more and more crucial due to the prevalence of online abuse and cyberbullying. With cutting-edge techniques for text and audio-visual data analysis on social media, the suggested system attains a 99.983% accuracy rate. Convolutional neural networks (CNNs) and recurrent neural networks (RNNs) are two categories of deep learning models. All things considered, the experimental findings showed that the FSSDL-CBDC technique outperformed the other options in several areas

Internet privacy · Social media · World Wide Web · Advanced Malware Detection Techniques · Bullying, Victimization, and Aggression · Computer Science · Environmental Science · Hate Speech and Cyberbullying Detection · Psychology · Artificial Intelligence