Machine Learning Teaching Materials - Bayesian Classification

Implementation of Naive Bayes Classifier from scratch with comparison to scikit-learn's GaussianNB, demonstrated on the Iris dataset.

Contents

Bayesian Classification Notebook

• Key Concepts:
  • Bayes' Theorem fundamentals
  • Naive Bayes classifier mathematics
  • Gaussian probability density function
  • Log-probability optimization
• Practical Implementation:
  • Complete Naive Bayes classifier implementation
  • Data standardization with StandardScaler
  • Model evaluation metrics (Accuracy, Precision, Recall, F1, Jaccard)
  • Confusion matrix visualization
• Dataset Analysis:
  • Iris dataset exploration
  • Feature distributions and correlations
  • Pairplots and heatmap visualizations

Key Features

• From-scratch implementation of Gaussian Naive Bayes
• Side-by-side comparison with scikit-learn implementation
• Detailed explanation of evaluation metrics
• Comprehensive visualization suite:
  • Pairplots
  • Correlation heatmaps
  • Class distribution charts
  • Confusion matrices

Dependencies

• Python 3.7+
• Jupyter Notebook
• Required libraries:

  pip install numpy pandas matplotlib seaborn scikit-learn