Summary of Linear Discriminant Analysis in Credit Scoring: a Transparent Hybrid Model Approach, by Md Shihab Reza et al.
Linear Discriminant Analysis in Credit Scoring: A Transparent Hybrid Model Approach
by Md Shihab Reza, Monirul Islam Mahmud, Ifti Azad Abeer, Nova Ahmed
First submitted to arxiv on: 5 Dec 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper explores the development of credit scoring approaches using machine learning (ML) and deep learning (DL) techniques. While complex models provide more accurate predictions, they often compromise interpretability, which is crucial for decision fairness in credit scoring. To address this issue, the authors implement Linear Discriminant Analysis (LDA) as a feature reduction technique to reduce model complexity. The study compares 6 ML models, 1 DL model, and a hybrid model with and without LDA, finding that the XG-DNN hybrid model outperforms others with an accuracy of 99.45% and F1 score of 99%. To interpret model decisions, the authors apply explainable AI techniques, such as LIME (local) and Morris Sensitivity Analysis (global). The research demonstrates how feature reduction techniques can be used without affecting model performance or explainability, making it useful for resource-constrained settings. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about using computers to help decide whether someone is a good credit risk. It uses special kinds of math called machine learning and deep learning to make these decisions. The problem is that these methods can be very complicated and hard to understand, which isn’t fair. To solve this issue, the authors used a technique called Linear Discriminant Analysis (LDA) to simplify things without affecting how well they work. They tested different models and found that one hybrid model worked the best with an accuracy of almost 100%. The authors also showed how to explain why these models made certain decisions. This research is important because it helps make sure credit scoring is fair and efficient. |
Keywords
» Artificial intelligence » Deep learning » F1 score » Machine learning
