Student Performance Clustering

My Contribution

• Performed exploratory data analysis on student performance, including missing values, duplicates, distributions, correlations, and outliers.
• Cleaned the dataset and encoded categorical features for machine learning.
• Created academic performance levels by binning mathematics, reading, and writing scores.
• Built a K-Means clustering model to group students based on performance and characteristics.
• Selected the cluster configuration using the Elbow Method and Silhouette Score.
• Applied Principal Component Analysis (PCA) for dimensionality reduction and cluster interpretation.
• Exported the clustering results, including the Cluster_PCA label, for the classification stage.
• Trained and evaluated Decision Tree and Random Forest models to predict cluster labels.
• Performed hyperparameter tuning using Grid Search, Randomized Search, and Bayesian Optimization.
• Developed a Streamlit dashboard to present data exploration, cluster profiles, visualizations, and model evaluation.