EduPro Student Segmentation and Personalized Course Recommendation System Using Machine Learning and Clustering Techniques

This work presents an AI-driven educational analytics system for learner segmentation and personalized course recommendation using machine learning and clustering techniques. The project integrates exploratory data analysis, feature engineering, clustering algorithms, ensemble learning models, and recommendation logic to analyze learner behaviour and predict educational preferences within online learning environments. The system was developed under the Unified Mentor educational analytics project framework as part of an undergraduate applied machine learning and data science study. A statistically realistic synthetic educational dataset was used to simulate learner demographics, transactional behaviour, instructor attributes, and course interaction patterns while addressing privacy and accessibility limitations associated with real-world educational data. Multiple machine learning models including Logistic Regression, KNN, SVM, Random Forest, XGBoost, LightGBM, and ensemble techniques were evaluated and compared. The project also includes learner segmentation using K-Means clustering, PCA-based visualization, silhouette analysis, and an interactive Streamlit dashboard for analytics and recommendation visualization.