No abstract

Data mining in education deals with formulating strategies for students with the aim to increase the parameters affecting the learning and employability. It also helps the educational institutes in maintaining their reputation as it is directly linked to the student's grades. We need to identify the parameters involved in learning and the relationship among those parameters. In EDM, feature selection (FS) is one of the most important and needed method in EDM, as it removes the features which have no direct link with the student's performance. For example, the date of birth of a student does not impact his/her performance. In this paper, an attempt has been made to improve the performance of the classifiers for undergraduate students at Maharaja Agrasen Institute of Technology. We have applied several techniques of data mining to make some rules that increase the learning and employability of students. The results of our study have shown a significant increase in accuracy, recall, precision and F-measure for naïve Bayes and decision tree classifiers.