A case study on Credit Risk Analysis using Taiwanese Banking Data

Author(s): Harshit Deepak Bhavnani, Shreyansh Suman Bardia

Abstract

This system-description paper essentially works towards aiding the financial industry in the sub-domain of credit risk by evaluating numerous machine learning algorithms in addition to neural networks, thereby observing that ensemble-based classifiers outperformed neural networks, and the best performance was demonstrated by the XGBoost classifier – a decision-tree-based ensemble machine learning algorithm that uses a gradient boosting framework that predicted with an accuracy score of 82.0833% and a precision score of 80.3417%. This research also led to an extensive survey of the socio-economical condition of Taiwan in order to understand the relationship between the features present in the dataset and the results obtained. With this research, we also verified that the claims made by several researchers stating that gradient boosting and random forest algorithm is well suited for credit risk while neural networks may not give impressive results also hold true in this case.

The Chartered Data Scientist Designation

Achieve the highest distinction in the data science profession.

Explore more from Association of Data Scientists

Become ADaSci Chapter Lead

As a chapter lead, you will have the opportunity to connect with fellow data professionals in your area, share knowledge and resources, and work together to advance the field of data science.