Preventive Mechanisms Against Cyberbullying in Social Media Environments

Abstract

Cyberbullying has become more common on social media sites. Since people of all ages use social media frequently, it's really important to make these platforms safer from cyberbullying. This paper introduces a new deep learning model known as DEA-RNN, which helps find cyberbullying on Twitter. The DEA-RNN model mixes a type of Recurrent Neural Network (RNN) called Elman with a smart method called the Dolphin Echolocation Algorithm (DEA) that fine-tunes the RNN settings and speeds up training. We tested DEA-RNN carefully using a collection of 10,000 tweets and compared how well it worked against other leading algorithms like Bi-directional long short-term memory (Bi-LSTM), RNN, Support Vector Machine (SVM), Multinomial Naive Bayes (MNB), and Random Forests (RF). The results showed that DEA-RNN performed better in all situations. It was more effective than the other methods at spotting cyberbullying on Twitter. In one specific test, DEA-RNN achieved an average accuracy of 90.45%, precision of 89.52%, recall of 88.98%, F1-score of 89.25%, and specificity of 90.94%.