Piyush Gupta

Developed a predictive model using Multiple Linear Regression to estimate house prices based on features like area and number of bedrooms. The workflow included data preparation, model training, and performance evaluation using MAE, MSE, and R² Score. Successfully deployed the model to predict prices for new property data with accurate, data-driven results.

Built and deployed an end-to-end ML web app using Random Forest Regression to predict real-time flight ticket prices. Integrated Flask for backend, HTML/CSS for a clean interface, and used Pickle to save the model, encoders, and scaler. Users can input flight details like airline, route, stops, timings, and instantly get a price estimate right in the browser.

Developed an end-to-end ML pipeline to predict salaries of tech professionals based on features like age, gender, education, experience, and job title. Implemented multiple regression algorithms including Linear, Ridge, Lasso, ElasticNet, KNN, Decision Tree, Random Forest, and Gradient Boosting, with StandardScaler for preprocessing and GridSearchCV/RandomizedSearchCV for hyperparameter tuning. Achieved a best R² score of 0.889, with performance evaluated using MAE, MSE, RMSE, and R².

Implemented clustering-based ML models to segment mall customers using features like spending behavior, income, age, and gender. Applied StandardScaler for preprocessing, PCA for dimensionality reduction, and clustering algorithms including K-Means, DBSCAN, Hierarchical, and Agglomerative Clustering. Incorporated Isolation Forest for anomaly detection, with PCA-transformed cluster visualization. Built a Flask web app for real-time cluster prediction from user input, and saved models, scalers, and PCA objects for deployment.