Abstract
We consider the prediction of possible student dropouts from an undergraduate Computer Science program in a higher educational institution. To this end, we collect our own data from the students who started their degree in years 2016 and 2017. After preprocessing and cleaning we are left with 366 participants. To predict graduations and dropouts from the program, four different binary classifiers, namely, Naive Bayes, Support Vector Machine, Logistic Regression, and Artificial Neural Network models were considered. The average performances of the four are reported to be 96%, 89%, 88%, and 95%, respectively. The studies of the similar kind are very useful in terms of advising the Computer Science majoring students how their performances to the date determine their graduation probabilities.