Ranking Learning Algorithm for Likert Scale (RLALS) for Prediction of Student Perceptions about Curriculum, Teaching-Learning and Research

The Likert scale is an important aspect of collecting data for various situations. The data is in ordinal form, so performing analysis and prediction requires a special kind of algorithm. In this paper, a ranking learning algorithm for the Likert Scale (RLALS) is proposed to predict ordinal data. The data from the education domain is collected for experimentation. The data related to the feedback process in the context of curriculum, teaching-learning, and research was collected from 339 students with 12 different parameters. The proposed algorithm is compared with a well-known logistic regression model. The accuracy of the proposed model before feature engineering and after feature engineering is better than logistic regression. The accuracy of the proposed model before feature engineering is 68.63%, while after feature engineering, it is 89.24%.