Data analysis and transformation

Now that we have processed our data, it is ready for analysis. We will be carrying out descriptive and exploratory analysis in this section, as mentioned earlier. We will analyze the different dataset attributes and talk about their significance, semantics, and relationship with the credit risk attribute. We will be using statistical functions, contingency tables, and visualizations to depict all of this.

Besides this, we will also be doing data transformation for some of the features in our dataset, namely the categorical variables. We will be doing this to combine the category classes which have similar semantics and remove the classes having very less proportion by merging them with a similar class. Some reasons for doing this include preventing the overfitting of our predictive models, which we will be building in Chapter 6, Credit Risk Detection and Prediction – Predictive Analytics, linking semantically similar classes together and also because modeling...