A novel hybridization of classification trees and artificial neural networks for selection of students in a business school
In recent years, business schools face a common problem of selecting quality students for their Master of Business Administration (MBA) programs so that the target placement percentage is achieved. Selecting a wrong student may increase the number of unplaced students. Also, more the number of unplaced students more is the negative impact on the institute’s reputation. Business school authorities would therefore always want to ensure that they admit the right set of students to their MBA program. In this article, we used supervised learning techniques to model and select the optimal academic characteristics of students to enhance their placement probability. We propose a novel hybrid model based on classification tree (CT) and artificial neural network (ANN) which we call as hybrid CT–ANN model, to analyse business school data. A comparative study of various supervised models with our proposed model using different performance measures is also presented. Our finding shows that the proposed hybrid CT–ANN model achieves greater accuracy in predicting students’ placement than conventional supervised learning models.
Chakraborty, Tanujit; Chattopadhyay, Swarup; and Chakraborty, Ashis Kumar, "A novel hybridization of classification trees and artificial neural networks for selection of students in a business school" (2018). Journal Articles. 1362.