Student Academic Performance Classification Using N-Valued Interval Neutrosophic Sets with Optimization Algorithms for Significant Feature Selection

Nahla Moussa¹, Cuauhtemoc Samaniego^2,*, Moustafa Mohamed Abouelnour³, Wael F. Ali⁴

¹College of Humanities and Sciences, Ajman University, UAE

²College of Business Administration, American University of the Middle East, Kuwait

³University of Khorfakkan, Sharjah, UAE

⁴American University in the Emirates, UAE

Emails: n.moussa@ajman.ac.ae; jose.reyna@aum.edu.kw; Moustafa.abdelmawla@ukf.ac.ae; wael.ali@aue.ae

Abstract

The most effectual tools for demonstrating uncertainty in decision-making issues are the neutrosophic set (NS) and its additions, like interval NS (INS), complex NS (CNS), and interval complex NS (ICNS). NS delivers an effectual and precise method for defining an imbalance of information as per the data features. In present times, students’ academic performances have been evaluated on the base of regular examinations or memory-related tests and by equating their performances to recognize the features for forecasting their academic excellence. The prediction of student academic performance is involved in Educational data mining (EDM), which mainly focuses on using data mining methods in the educational side. EDM progress models for finding data, which is a result of educational surroundings. This paper presents a Student Academic Performance Prediction Using N-Valued Interval Neutrosophic Sets and Optimization Algorithms (SAPP-NINSOA). The main intention of the SAPP-NINSOA technique is to provide a prevalent technology for predicting students’ academic performance using an advanced optimization algorithm. At first, the data pre-processing stage applies Z-score normalization to convert input data into a beneficial format. Besides, the secretary bird optimization algorithm (SBOA) to select the relevant features from input data has executed the feature selection process. Next, the proposed SAPP-NINSOA model designs the N‐Valued Interval Neutrosophic Sets (NVINS) method for the classification process. Finally, the arithmetic optimization algorithm (AOA) fine-tunes the parameter values of the NVINS model. An extensive range of experimentation was led to certify the performance of the SAPP-NINSOA technique. The simulation outcomes stated that the SAPP-NINSOA algorithm emphasized furtherance when compared to other existing systems.

Keywords: Student Academic Performance Prediction; Neutrosophic Sets; Feature Selection; N-Valued Interval; Arithmetic Optimization Algorithm