Y. Shakuntala1 , Mohammed Saif Rehan2 , Mohammed Aftab Talha3 , Abdullah Muneeb Siddiqui4 03 04 The increasing use of social media platforms has created new ways for people to communicate but it has also led to the rise of cyberbullying which affects users emotionally and psychologically This project focuses on developing a machine learningbased system to detect cyberbullying in usergenerated messages The system is designed to analyze text posts and classify them as offensive or nonoffensive using algorithms such as AdaBoost Multinomial Nave Bayes and Stochastic Gradient Descent SGD In addition to detecting abusive content a Support Vector Machine SVM model is used to determine the sentiment of the message as positive or negative The application provides an interactive platform where users can register log in and post messages along with images Each message is automatically evaluated and the results are displayed to the user If offensive language is detected the system increases the users offense count which is visible to the administrator Based on this count the admin has the authority to block users who repeatedly violate the rules The system uses a MySQL database to store user information posts and analysis results This approach reduces the need for manual monitoring and helps in maintaining a safer online environment The performance of the models is evaluated using accuracy and other metrics showing that machine learning can effectively identify harmful content The system serves as a useful tool for managing online interactions responsibly KeywordsCyberbullying detection social media analysis machine learning text classification Stochastic Gradient Descent SGD AdaBoost Multinomial Nave Bayes Support Vector Machine SVM sentiment analysis offensive content detection data preprocessing model evaluation metrics webbased application MySQL database user behavior monitoring Copyright (c) Journal of Science Engineering Technology and Management Science. All rights reserved 2026