german-credit-dataset

Some old code.

A Consideration Point of Fraud Detection in Bank Loans - Project Aug 2018

Boosting is one of the methods contained in Ensemble Learning. This technique makes it possible to combine several models into one more robust model.

This project analyzes how applicant characteristics such as credit history, income, education, employment status, and property area relate to loan approval outcomes, presenting evidence-based insights through visualizations and Chi-Square tests.

Cost Sensitive Learning in German Credit Data

Machine learning-based credit risk prediction and risk banding, with threshold tuning, XGBoost modeling, and actionable policy simulations

Gini calculation on "German Credit Dataset" by Kaggle using R.

This repository provides some group fairness metrics to Machine Learning classifier of German Credit Scoring Dataset. It computes demographic parity, equal opportunity and equalized odd for the sensitive variable gender.

In this repository, we will explore linear models and compare different methods of learning linear classifiers and regression functions.

Explore how bias can be detected and mitigated in real-world datasets

A comprehensive machine learning solution for predicting credit risk using XGBoost with SHAP explainability. Features an interactive Streamlit interface for real-time risk assessment and transparent decisionmaking. Built on the German Credit Dataset with advanced feature engineering, hyperparameter tuning and productionready model interpretability.