Prediction of Student Academic Performance using Data Mining and Warehousing Techniques

Abstract

In educational institutions, managing and predicting student performance is vital in enhancing academic outcomes. This project introduces a Student Performance Prediction Management System (SPPMS) that leverages the power of data mining and warehousing techniques to forecast student performance. The system integrates various data sources, including academic records, demographic information, and extracurricular activities, to build comprehensive student profiles. Data preprocessing techniques are applied to clean, transform, and integrate heterogeneous data into a centralized data warehouse. The predictive modelling phase employs data mining algorithms such as classification, regression, and clustering to extract meaningful patterns from the integrated data. These patterns are utilized to develop predictive models capable of forecasting student performance indicators such as grades, exam scores, and academic outcomes. This project is conceived with the vision of leveraging advanced analytics to gain actionable insights into factors influencing student academic success, facilitating informed decision-making for educators, administrators, and stakeholders.