TrustFinance

Smart Credit Risk Evolution is a web application that evaluates the credit risk of loan applicants. Using a machine learning model, it predicts whether a loan application should be approved and provides the probability of approval based on various applicant details.

Table of Contents

• Project Overview
• Features
• Installation
• Usage
• Model Training
• API Endpoints
• Technologies Used
• Contributing
• License

Project Overview

Smart Credit Risk Evolution leverages a logistic regression model to assess the risk associated with loan applicants. The system provides a frontend for users to input their details and receive a prediction on their loan application status along with the probability of approval.

Features

• Input personal and financial information through a web interface.
• Receive real-time predictions and approval probabilities.
• User-friendly interface built with Streamlit.
• Backend powered by Flask and a logistic regression model.

Installation

Prerequisites

• Python 3.8+
• pip (Python package installer)

Setup

1. Clone the repository:

   git clone https://github.com/yourusername/smart-credit-risk-evolution.git
   cd smart-credit-risk-evolution

2. Install dependencies:

   pip install -r requirements.txt

3. Prepare the Model:

   Ensure you have the dataset (Loan_default.csv) in the appropriate directory. Run the Jupyter Notebook (model_training.ipynb) to train and save the model:

   jupyter notebook model_training.ipynb

   This will generate a prediction_model.pkl file in the specified directory.

4. Configure the Flask server:

   Update the model_path in the Flask script to the location of your prediction_model.pkl.

   model_path = r'C:\Users\hp\Documents\AILML Hackathon\prediction_model.pkl'

5. Run the Flask server:

   python flask_server.py

6. Run the Streamlit app:

   streamlit run streamlit_app.py

Usage

• Open the Streamlit app in your web browser. By default, it runs at http://localhost:8501.
• Input your personal and financial information into the provided form.
• Click the "Predict" button to receive your loan approval prediction and probability.

Model Training

The model is trained using a logistic regression algorithm on a dataset of loan applications. The training process involves:

• Preprocessing the data by converting categorical variables to numerical values.
• Splitting the data into training and testing sets.
• Training the logistic regression model.
• Saving the trained model for use in the Flask server.

Example Code for Model Training

import pandas as pd
import numpy as np
import joblib
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression

# Load the dataset
df = pd.read_csv("Loan_default.csv")

# Preprocess the dataset
df = df.drop(columns=['LoanID'])
df.replace({"Education": {"Bachelor's": 0, 'High School': 1, "Master's": 2, 'PhD': 3}}, inplace=True)
df.replace({"EmploymentType": {'Part-time': 0, 'Unemployed': 1, 'Self-employed': 2, 'Full-time': 3}}, inplace=True)
df.replace({"MaritalStatus": {'Married': 0, 'Divorced': 1, 'Single': 2}}, inplace=True)
df.replace({"HasMortgage": {'No': 0, 'Yes': 1}}, inplace=True)
df.replace({"HasDependents": {'No': 0, 'Yes': 1}}, inplace=True)
df.replace({"LoanPurpose": {'Business': 0, 'Home': 1, 'Education': 2, 'Other': 3, 'Auto': 4}}, inplace=True)
df.replace({"HasCoSigner": {'No': 0, 'Yes': 1}}, inplace=True)

# Split the data
X = df.drop(columns=['Default'], axis=1)
Y = df['Default']
X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.1, stratify=Y, random_state=2)

# Train the model
model = LogisticRegression()
model.fit(X_train, Y_train)

# Save the model
joblib.dump(model, 'prediction_model.pkl')

API Endpoints

POST /predict-model

Description: Receives applicant data and returns the prediction and probability. Request Body: JSON object containing applicant features.

{
    "age": 30,
    "income": 50000,
    "loanAmount": 20000,
    "creditScore": 700,
    "monthsEmployed": 36,
    "numCreditLines": 5,
    "interestRate": 5,
    "loanTerm": 60,
    "dtiRatio": 0.4,
    "education": 1,
    "employmentType": 3,
    "maritalStatus": 0,
    "hasMortgage": 1,
    "hasDependents": 0,
    "loanPurpose": 1,
    "hasCoSigner": 0
}
{
    "prediction": "Approved",
    "probability": 0.85
}

Technologies Used

• Frontend: Streamlit
• Backend: Flask
• Machine Learning: scikit-learn, joblib
• Data Processing: pandas, numpy

Contributing

Contributions are welcome! Please fork the repository and submit a pull request with your changes. Ensure that your code adheres to the project's coding standards and includes appropriate tests.