Solving classification problem by Logistic Regression model implementation; Linear transformation techniques application for dimensionality reduction:
“Unsupervised” PCA (Principal Component Analysis) algorithm
“Supervised” LDA (Linear Discriminant Analysis) algorithm
Model evaluation results with PCA:
- Accuracy: 0.966666667
- Precision (“exactness)”: 0.933333333
- Recall (“precision”): 1
- F1 Score (“compromise” between precision and recall): 0.965517241
Model evaluation with LDA showed no incorrect results.