https://github.com/eva-kaushik/fraud-detection-mlops
An AI-powered application leveraging machine learning for detecting fraudulent activities in financial transactions with MLOps.
https://github.com/eva-kaushik/fraud-detection-mlops
financial-analysis fraud-detection fraud-prevention h2o-automl h2oai mlops tensorflow
Last synced: 3 months ago
JSON representation
An AI-powered application leveraging machine learning for detecting fraudulent activities in financial transactions with MLOps.
- Host: GitHub
- URL: https://github.com/eva-kaushik/fraud-detection-mlops
- Owner: Eva-Kaushik
- Created: 2024-03-26T10:26:22.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2024-03-26T11:14:06.000Z (over 2 years ago)
- Last Synced: 2025-06-26T18:43:52.598Z (about 1 year ago)
- Topics: financial-analysis, fraud-detection, fraud-prevention, h2o-automl, h2oai, mlops, tensorflow
- Language: Jupyter Notebook
- Homepage:
- Size: 15.6 KB
- Stars: 0
- Watchers: 1
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Fraud-Detection-MLOps
## Fraud Detection System
## Overview
The Fraud Detection System is an AI-powered application designed to detect fraudulent activities in financial transactions. It leverages machine learning techniques to analyze historical data and predict the likelihood of a transaction being fraudulent.
## Features
- **Data Collection**: Collects data from NSFAS and University of Limpopo CSV files.
- **Data Preprocessing**: Cleans, preprocesses, and merges the data, handling missing values and performing feature engineering.
- **Model Training**: Trains a neural network classifier using TensorFlow/Keras.
- **Model Evaluation**: Evaluates the trained model's performance using accuracy, precision, recall, and F1 score.
- **SMS Confirmation**: Generates a random confirmation code and sends it via SMS to the user for confirmation.
## Requirements
- Python 3.x
- Pandas
- Scikit-learn
- Imbalanced-learn
- Keras
- TensorFlow
- H2O.ai
- Other dependencies listed in `requirements.txt`
## Usage
Installation: Install the required packages listed in `requirements.txt`.
```bash
pip install -r requirements.txt
Data Preparation: Ensure that the NSFAS and University of Limpopo CSV files are available and correctly located. Update the file paths in the code if necessary.
Running the Application: Execute the main script Fraud Detection.py.
bash
python fraud_detection.py
Result Analysis: Check the output for model evaluation metrics and the confirmation code sent via SMS.