Student dropout is a pervasive problem in the education system, with a significant number of students failing to complete their studies due to various reasons such as academic difficulties, financial issues, personal problems, and more. Dropout not only affects the student’s academic and career prospects but also has significant financial and social costs for educational institutions and society. Identifying at-risk students and providing targeted interventions to prevent dropout has been a longstanding goal of educational researchers and practitioners. In recent years, machine Learning algorithms have emerged as a promising method for predicting student attrition and pinpointing the primary elements that contribute to it. Predicting student attrition and pinpointing the primary elements that contribute to it. This research paper offers an extensive investigation of the application of machine-learning algorithms-in-predicting student attrition. The study is based on dataset of student’s academic and demographic information collected from a major university in the Bangladesh. The dataset comprises information on over 400 students who enrolled in the university between 2015 and 2020, including their academic records, demographic characteristics, and enrollment history. Analyze datasets utilizing a variety of machine learning approaches such as support vector machines, random forests, logistic regression etc. Evaluate the performance of these algorithms using metrics such as accuracy, precision, recall, and F1 score. Our study finds that machine learning algorithms can effectively predict student dropout with high accuracy, precision, and recall. The best-performing algorithm is Random Forest with a precision of 0.78, recall of 0.78, and F1-score of 0.78. Logistic regression and KNN algorithms also perform reasonably well, with a precision of 0.75 and 0.76, respectively.