R: Mining Spatial, Text, Web, and Social Media Data

Practical project around dimensionality reduction

We are going to apply dimensionality reduction procedure, both model-based approach and principal component-based approach, on the dataset to come up with less number of features so that we can use those features for classification of the customers into defaulters and no-defaulters.

For practical project, we have considered a dataset default of credit card clients.csv, which contains 30,000 samples and 24 attributes or dimensions. We are going to apply two different methods of feature reduction: the traditional way and the modern machine learning way.

Attribute description

The following are descriptions for the attributes from the dataset:

• X1: Amount of the given credit (NT dollar). It includes both ...