国内首个迁移学习赛题 中国平安前海征信“好信杯”迁移学习大数据算法大赛 FInSight团队作品(算法方案排名第三)

[Project repo] Improving business with a credit risk model

天池大数据竞赛 千里马大赛 风险识别与预测赛题 Top5

Predicting how much loan will be approved

Credito - Credit Risk Analysis using XGBoost Classifier with RandomizedSearchCV for loan approval decisions.

This dataset contains information on default payments, demographic factors, credit data, history of payment, and bill statements of credit card clients in Taiwan from April 2005 to September 2005. Submitted to GUVI (IITM)

Credit Risk Modeling is a fintech-focused project that enhances traditional credit scoring by introducing key financial metrics like Debt-to-Income Ratio, Total Financial Accounts, and Total Savings. It calculates scaled credit scores over 3 and 6-month periods to provide a comprehensive assessment of customer creditworthiness. The project helps fi

Loan Defaulter's Prediction using Statistical Analysis

Replication of the paper "Credit risk evaluation model with textual features from loan descriptions for P2P lending", Zhang et al. (2020), published in Electric Commerce Research and Applications

A collection of the scripts and notebooks I wrote as part of my Data Science Bootcamp capstone project

Built a Logistic Regression model to predict loan risk, focusing on credit risk management with precision, recall, and F1-score evaluation

Using multiple supervised machine learning algorithms to measure credit risk

This Credit Risk Assessment agent leverages advanced machine learning techniques, including Chain of Thought (CoT) reasoning and Reinforcement Learning (RL), to evaluate credit risk. The project aims to provide more transparent, effective, and explainable solutions to the complex task of assessing creditworthiness.

We create a model using the gradient boosting algorithm to cut down on the noise and improve performance. This work was done during an informal project under Prof. Yaganti while studying at BITS.