House Price Prediction System

Build and deploy a linear regression model to predict residential property prices using real-world housing data. Apply gradient descent optimization, feature engineering, and model evaluation metrics to create a robust prediction pipeline. Compare model performance across different regression techniques and deploy the final model as a usable application.

Credit Risk Assessment Using Ensemble Methods

Develop a credit risk classification system using Decision Trees, Random Forests, XGBoost, and LightGBM to assess loan default probability. Implement and compare bagging, boosting, and stacking ensemble strategies to maximize prediction accuracy. Evaluate models using precision, recall, F1-score, and ROC-AUC to select the best-performing approach.

Customer Segmentation & Image Classification

Apply K-Means and DBSCAN clustering algorithms to segment customers based on behavioral data, enabling targeted marketing strategies. Additionally, build an SVM-based image classifier on the MNIST dataset using kernel methods and PCA for dimensionality reduction. Analyze clustering quality and classification accuracy to derive actionable business insights.