Pre-service mathematics teachers' performance prediction using a modified Mann-Tseng forward-backward splitting algorithm

Abstract

Classifying educational data into a particular category remains challenging due to the massive and extensive number of variables within the dataset. This paper emphasizes a new algorithm for variational inclusion problems with the classification of pre-service mathematics teachers' performance in their method courses through a mathematics teacher education program as its application. First, we propose the modified Mann-Tseng forward-backward splitting algorithm based on inertial technique to speed up the convergence of the algorithm. Then, we prove the weak convergence theorem, we compare and demonstrate the efficacy and applicability of our classification schemes in extreme learning machine (ELM) with other machine learning methods; support vector machine (SVM), logistic regression, boosted trees. Moreover, we compare our algorithm with other algorithms in the same ELM. The application is based on the genuine educational data provided in this paper.