Skip to content

Keerthana-webdev/Fraud-Detection

Repository files navigation

Fraud Detection Prediction App

Description

This project is a machine learning-powered web application that detects fraudulent financial transactions. Built using Streamlit, it provides an interactive interface where users can input transaction details and instantly get predictions on whether the transaction is safe or fraudulent.

The model is trained using transaction data and deployed using a pipeline for seamless preprocessing and prediction.


Features

  • Interactive web app using Streamlit
  • Real-time fraud prediction
  • Pre-trained ML pipeline using joblib
  • User-friendly input interface
  • Instant results with clear visual indicators
  • Supports multiple transaction types

Tech Stack

  • Python
  • Pandas
  • Numpy
  • matplotlib
  • Seaborn
  • Scikit-learn
  • Streamlit
  • Joblib

Project Structure

FRAUD_DETECTION/
│── analysis_model.ipynb          # Model training & analysis notebook
│── fraud_detection_pipeline.pkl # Saved ML pipeline model
│── fraud_detection.py           # Streamlit web app

Installation & Setup (Run Locally)

1️⃣ Clone the Repository
git clone https://github.com/Keerthana-webdev/Fraud-Detection.git
cd fraud-detection

2️⃣ Install Dependencies
pip install -r requirements.txt
*(If requirements.txt not available, install manually:)*
pip install streamlit pandas scikit-learn joblib

3️⃣ Run the App
streamlit run fraud_detection.py

How to Use

  1. Select the transaction type
  2. Enter transaction details like:
    • Amount
    • Sender balance
    • Receiver balance
  3. Click on Predict
  4. View result:
    • ✅ Safe Transaction
    • 🚨 Fraudulent Transaction

Live Demo

https://fraud-detection-qnsjo8ho3tgzxnepurqtrg.streamlit.app/


Author

Keerthana S

About

A Machine Learning-based Fraud Detection Web App built using Streamlit that predicts whether a financial transaction is fraudulent based on transaction details.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors