A New Hybrid Model to Predict the Performance of Trainee Teachers Based on Clustering and Classification

Abstract

This article explores how artificial intelligence, particularly machine learning, can be used to assess and predict the performance of trainee teachers in Morocco. Considering the country's focus on integrating technology into education and the challenges posed by the COVID-19 pandemic, the authors propose a novel hybrid model that combines clustering and classification algorithms. This model aims to understand the Information and Communication Technologies (ICT) skills of trainees from various backgrounds and predict their performance after training at a Moroccan center. It should be noted that, in this study, we used trainee data in compliance with ethical principles and confidentiality protocols. All data collected were anonymized in order to protect the identity of the participants and guarantee their confidentiality. The study investigates whether a gap exists in the digital literacy of trainees based on their prior degrees and analyzes their progress after training. By applying the hybrid model, the research identified distinct groups of trainees, including high achievers and a mixed group with varied performance. The findings suggest that while a trainee's digital skills may be influenced by their prior institution, the training program effectively improves their ICT skills and allows them to achieve success. The clustering algorithm used prior to classification provides a better understanding of the data and improves the classification rate. The experimental results provide valuable information for scientists looking to take advantage of new clustering techniques and classification for a variety of applications in data analysis. The paper further explores the impact of AI in education, details the proposed model, and discusses the results alongside potential avenues for future research