The primary objective of this project is to analyze student performance data to identify patterns and predict future performance. By leveraging data analytics techniques, we aim to gain insights into factors influencing academic outcomes and develop predictive models to anticipate students' future performance.
We utilize two main datasets for our analysis:
-
UCI Machine Learning Repository's Student Performance Dataset: This publicly available dataset provides detailed information about student performance in various subjects. It serves as the primary source for our analysis.
- Dataset Link: UCI Machine Learning Repository's Student Performance
-
World Bank's Learning Poverty Global Database: We also explore the World Bank's Learning Poverty Global Database for a broader analysis of global educational indicators. This dataset offers insights into the prevalence of learning poverty and related factors on a global scale.
Our analysis involves the following key techniques:
- Correlation Analysis: We perform correlation analysis to identify relationships between performance in different subjects. This helps uncover associations between academic variables and informs our understanding of the factors influencing student performance.
- Regression Models: We explore regression models to predict student performance based on background factors available in the datasets. By building predictive models, we aim to forecast future academic outcomes and provide insights for educators and policymakers.
By combining these analysis techniques with comprehensive datasets, we strive to enhance our understanding of student performance dynamics and contribute to the development of evidence-based interventions in education.