“An Advanced AI-Integrated Expert System for Academic Performance Forecasting Using Ensemble Learning and Educational Data Mining Techniques”

Abstract

Academic performance at the secondary school level plays a vital role in determining students’ future educational and career opportunities; however, traditional evaluation methods based on examinations and teacher observations often fail to capture the complex factors influencing student outcomes. This study focuses on the design and development of an expert system for predicting the academic performance of secondary school students in Raigad District by analyzing the present scenario of ICT implementation, evaluating student performance, and identifying key influencing parameters such as attendance, study habits, socio-economic background, and technological access. A mixed-method research approach is adopted, involving survey-based data collection from students, teachers, principals, and parents using stratified sampling techniques, followed by data preprocessing, statistical analysis using tools like SPSS and Excel, and implementation of machine learning algorithms including Decision Trees, Random Forest, Support Vector Machines, and ensemble methods. The proposed expert system integrates multi-dimensional data to generate accurate predictions and includes an impact factor analysis framework to identify significant determinants of academic performance. The expected outcomes of the study include improved prediction accuracy, early identification of academically at-risk students, enhanced decision-making support for educators, and better understanding of ICT’s role in education. The study concludes that the integration of expert systems and machine learning provides a comprehensive and reliable approach to academic performance prediction, enabling data-driven decision-making, personalized learning interventions, and improved educational outcomes in secondary schools.