In this analysis, I will demonstrate how PCA and K-Means clustering can be applied to credit risk data. In this data set, we do not have a target variable, which leads us to build an unsupervised machine learning model.
Briefly in this analysis you will read:
- EDA
- Data normalization and scaling techniques
- Handling outliers
- Missing data imputaing with KNN
- PCA technique and interpreting its results
- K-Means cluster analysis with PCA data
- Validation of K-Means cluster analysis
- Interpreting K-Means cluster analysis