Perceiving Highest Faculty Performance Using Classification Algorithms

Abstract

The main objective of this study was to predict who among the faculty has the highest performance in terms of students’ evaluation and head of the department (HOD) evaluation using data mining. The authors proposed three different classification techniques such as Decision tree, Naïve Bayes, and KNN algorithms. These were used to build classifier models that will determine the faculty performance. Their performances were compared over a dataset composed of responses of students and evaluation coming from the HOD using accuracy, precision, and recall. According to the results, the Nave Bayes classifier demonstrated the lowest accuracy over other classifiers that separate the most important variables, "Strongly Agree" and "Agree," with an accuracy of 91.67 percent based on student perception. This result can be a big help to the faculty and administration in decision making in what ways the faculty members can improve their performances where it is most needed.