Prediction of Academic Performance of Alcoholic Students Using Data Mining Techniques

Alcohol consumption by students has become a serious issue nowadays. Addiction to alcohol leads to the poor academic performance of students. This paper describes few algorithms that help to improve the efficiency of academic performance of students addicted to alcohol. In the paper, we are using one of the popular Data Mining technique-"Prediction" and finding out the best algorithm among other algorithms. Our project is to analyze the academic excellence of the college professionals by making use of WEKA toolkit and R Studio. We implement this project by making use of alcohol consumption by student datasets provided by kaggle website. It is composed of 395 tuples and 33 attributes. A classification model is built by making use of Naive Bayes and ID3. Comparison of accuracy is done between R and WEKA. The prediction is performed in order to find out whether a student can be promoted or demoted in the next academic year when previous year marks are considered.