My Kaggle ML Projects

Explore my machine learning projects on Kaggle, where I apply various algorithms and techniques.

Titanic - Machine Learning from Disaster

• Exploratory Data Analysis (EDA)
• Generalized Linear Model (GLM): Accuracy = 94%
• Random Forest (RF): Accuracy = 98.32%

Anime Recommendations Database

• Exploratory Data Analysis (EDA)
• Item-Based Collaborative Filtering (IB_CF): Implemented with dummies [ALL_ids]
• Hybrid Collaborative Filtering (Hybrid_CF): Applied to [1000_ids]

House Prices - Advanced Regression Techniques

• Linear Model (LM) with Lasso: RMSE = 0.16370
• Random Forest (RF): RMSE = 0.15591
• Gradient Boosting (GB): RMSE = 0.14144

Store Sales - Time Series Forecasting

• Exploratory Data Analysis (EDA)
• Prophet with Regressors: RMSE = 0.43745
• ARIMA with Regressors: RMSE = 0.45330
• CatBoost with Fourier Series: RMSE = 0.51673

Wine Quality

• XGBoost Classifier with Streamlit UI

Mercedes-Benz Greener Manufacturing

• TensorFlow Model with Isomap and JS Encoder: k-fold mean RMSE = 0.06534