Final project of IBM's course https://www.coursera.org/learn/machine-learning-with-python on coursera
This classification project aims to analyze and predict loan default risk using machine learning algorithms such as K-Nearest Neighbors (KNN), Support Vector Machines (SVM), Decision Trees, and Logistic Regression.
With a dataset of 346 customers containing historical loan information, the project categorizes borrowers into risk groups based on the likelihood of loan default. By comparing the performance of each algorithm, the project seeks to identify the most accurate and reliable method for loan default risk classification.
The insights gained from this project can help financial institutions enhance their risk management strategies, make informed lending decisions, and optimize their overall loan portfolio.
A simple comparison between KNN,SVM,Decision Tree and Logistic Regression models on a given data set of loans records.
| Algorithm | Jaccard | F1-score | LogLoss |
|---|---|---|---|
| KNN | 0.7407 | 0.7144 | NA |
| Decision Tree | 0.7592 | 0.7618 | NA |
| SVM | 0.7592 | 0.6959 | NA |
| LogisticRegression | 0.7777 | 0.7089 | 0.4947 |