An ID3 Algorithm for Performance of Decision Tree in Predicting Student’s Absenteeism in an Academic Year using Categorical Datasets

Objective: The main objective of higher educational institutes is to provide quality education to its students and to improve the quality of managerial decisions. Objective of this paper is to use the data mining techniques like decision tree induction, rule mining to predict the student’s behavior. Methods: We have used ID3 algorithm to predict the reason why students take leave from classes. The data collected from 123 students studying in an arts and science college located in semi-rural area in Villupuram district. Data collection is base on the questionnaire in five point scale method. Decision tree induction algorithm (ID3) was used to predict the reason and the decision tree was constructed using Entropy and Information Gain. Findings: The result obtained from Tanagra tool shows that job attribute plays a key role in predicting the absenteeism of students. That is lack of attendance is due to the job attribute. The results can be used to take managerial decisions. Application: Discovering the knowledge from the educational data helps to achieve the highest level of quality in higher education. The results shows that the decision tree induction algorithms can be used to predict the student’s behavior as well as to improve the managerial decision making process.