Enhancing anti-social behaviour detection on Twitter: A comparative approach to machine learning models
B Senthil Kumar, B. A. Nagashree, Sindu R Kashyapa, B S Rakshitha, M R Narasimha Murthy · 2025
X(Twitter) is a platform that can be used for faster sharing, concise updates, opinions, and engaging in various public discussions on a variety of concepts makes it a valuable tool for communication so the purpose of this research is to aid in incorporating different advanced techniques to filter out the hateful tweets in the form of text, to make this platform to be more resourceful for the society. In this research, we incorporated many methods to determine instances of cyber hate that were frequent on the X(Twitter) platform via several kinds of pre-processing approaches, including sentiment analysis, feature engineering, and Doc2Vec embedding. Following that data was meticulously pre-processed and was subjected to a variety of machine learning models, including MLP, AdaBoost, Decision Tree, SVM(Linear), Random Forest, and Logistic Regression models. Then we assessed the model&s;s effectiveness using various criteria like accuracy, precision, recall, and F1 Score. The comparative analysis demonstrated the SVM (Linear) model&s;s superior performance, surpassing all the competitors. This highlights the superiority of SVM over other machine learning methods executed in this research.