Many educational institutions put a high value on monitoring and supporting university freshmen. The high proportion of students who do not receive their free education exemplifies this, as do the full economic and social costs associated with them. Higher education institutions’ challenge is developing and improving policies that will increase student retention, particularly in the beginning years. In this paper, we propose a methodology and a specific classification algorithm to find a better way to trace the success and dropout of the students, which would help universities to take the necessary steps for their respective students. We have used decision tree, logistic regression, naive Bayesian, and KNN algorithms and compared them to get better prediction results from them. The data set we used is information on undergrad degree students of Polytechnic Institute of Portalegre, Portugal and the result we found from the model was able to predict the students’ future situation of their degree. This will help the education institution and other concern institutions to make early decisions about students for their betterment.