Machine Learning Project for Predictive Maintenance using Sensor Data

This project focuses on building a complete Machine Learning pipeline for Predictive Maintenance using real-world sensor data.

The workflow includes:

- Data Collection and Exploration

- Data Cleaning and Preprocessing

- Feature Engineering and Scaling

- Clustering using K-Means to discover hidden patterns

- Labeling data based on clusters

- Classification using Support Vector Machine (SVM)

- Hyperparameter tuning using Grid Search and Cross-Validation

- Regression analysis for predicting tool wear

- Data Visualization and Analytics

Tools & Technologies:

Python, Pandas, NumPy, Scikit-learn, Matplotlib, Seaborn

Key Results:

- High classification accuracy using SVM

- Effective clustering of machine states

- Accurate regression predictions for tool wear

This project demonstrates my ability to handle end-to-end Machine Learning workflows and deliver data-driven insights.