In this analysis, I will demonstrate how PCA works in different tasks and how much time and resources we save in our daily analysis. To do so, I will make a basic EDA and then build a PCA model. I will first apply the PCA output on K-Means clustering analysis and then test the same output on logistic regression analysis.
Briefly in this anaylis:
- Basic EDA
- Data normalization and scaling
- Correlation matrix
- Smote oversampling technique
- Interpreting PCA results
- K-Means clustering with PCA outcomes
- Accuracy rates across PCA components
- Accuracy rates with and without PCA