Data-Driven Materials Discovery Using Machine Learning

     Understand materials data and build the first ML model

• Introduction to materials informatics
• Types of materials data:
• Introduction to a real-world materials dataset

  Hands-on Activities

• Load dataset in Google Colab
• Data cleaning and preprocessing
• Feature understanding: composition to features

📅 Day 2: Machine Learning for Property Prediction

  Build predictive models for material properties

• Regression models for materials discovery
• Linear Regression
• Random Forest
• Basics of feature engineering

Hands-on Activities

• Train an ML model to predict material properties
• Example targets: bandgap, conductivity, or strength
• Evaluate model performance using R² and MAE

📅 Day 3: Optimization + Interpretation + Research Output

   Make results research-ready

• Model improvement techniques
• Feature importance analysis
• Interpretation of results

Hands-on Activities

• Improve model performance
• Generate plots and comparison graphs
• Export results for reporting

Final Output

Model, results, plots, and a research-ready case study