Cyberbullying, a great trouble enabled by using virtual communique systems, offers massive barriers to the intellectual and emotional fitness of people, particularly young adults and teenagers. The rise of social media and the potential to stay anonymous on line has substantially elevated its presence and have an effect on global. As a reaction, researchers have resorted to employing device gaining knowledge of techniques for the motive of identifying and reducing the impact of the problem. This takes a look at examines cutting-edge research on the detection of cyberbullying, with a selected emphasis on the utilization of device studying strategies for analyzing Twitter facts. The facts pretreatment tiers, which contain the removal of stop words, stemming, and lemmatization, are defined intensive. The analysis of the dataset provides valuable insights into the distribution and properties of the accrued statistics. Following the cleansing process, any duplicate values are removed, and the length of the textual content is chosen for education functions. The dataset is partitioned into separate education and trying out sets, and the elegance distribution is equalized the usage of SMOTE. The overall performance of 4 system mastering algorithms-Logistic Regression, Decision Tree, Naïve Bayes, and Random Forest-become compared. Random Forest outperformed the other methods in terms of precision, recall, F1 score, and accuracy. Confusion matrices offer further understanding into the performance of a model. In summary, this work emphasizes the significance of machine learning in tackling cyberbullying and proposes potential areas for further investigation.