https://github.com/shoaib1522/credit-card-fraud-analysis

"A comprehensive project on Credit Card Fraud Detection combining Exploratory Data Analysis, Machine Learning, and an interactive Streamlit web app for real-world applicability."
https://github.com/shoaib1522/credit-card-fraud-analysis

eda epochs mlp-classifier neural-network numpy pandas-python perceptron-learning-algorithm python regression sckiit-learn seaborn streamlit streamlit-webapp tensorflow

Last synced: 4 months ago
JSON representation

"A comprehensive project on Credit Card Fraud Detection combining Exploratory Data Analysis, Machine Learning, and an interactive Streamlit web app for real-world applicability."

• Host: GitHub
• URL: https://github.com/shoaib1522/credit-card-fraud-analysis
• Owner: shoaib1522
• License: mit
• Created: 2025-01-07T17:55:40.000Z (6 months ago)
• Default Branch: main
• Last Pushed: 2025-01-07T18:12:12.000Z (6 months ago)
• Last Synced: 2025-03-01T22:28:22.042Z (4 months ago)
• Topics: eda, epochs, mlp-classifier, neural-network, numpy, pandas-python, perceptron-learning-algorithm, python, regression, sckiit-learn, seaborn, streamlit, streamlit-webapp, tensorflow
• Language: Jupyter Notebook
• Homepage: https://www.kaggle.com/code/muhammadshoaibahmad/credit-card-fraud-eda
• Size: 64.3 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        

# Credit Card Fraud Detection and EDA Analysis

## 📌 Overview

This repository contains the work for my **Term Project** focused on **Exploratory Data Analysis (EDA)** and building a **Machine Learning model** to detect credit card fraud. The dataset used includes transaction data with a highly imbalanced distribution of fraudulent and legitimate transactions.

The project involves:

- Performing **EDA** to uncover patterns and insights from the data.

- Implementing a **web application** using Streamlit to interactively explore data and make predictions.

- Creating a **Jupyter Notebook** for in-depth analysis and visualization.

- Building and training a **Machine Learning Model** (Multi-Layer Perceptron) to classify transactions as fraudulent or legitimate.

---

## 📁 Repository Structure

```plaintext

├── .gitignore                     # Files to ignore in version control

├── App.py                         # Streamlit application for fraud detection

├── LICENSE                        # MIT license for the repository

├── README.md                      # Project documentation

├── credit-card-fraud-eda.ipynb    # Jupyter Notebook with EDA and analysis

├── credit_card/                   # Contains processed data or relevant resources

├── credit_card2/                  # Additional resources or files

├── credit_card3/                  # Additional resources or files

```

---

## 💻 Key Features

### 1. **Exploratory Data Analysis (EDA)**

- Visualizations of transaction distributions and trends.

- Analysis of the imbalanced class distribution.

- Insights into transaction amounts and time.

### 2. **Streamlit Web App**

- Interactive user interface for:

  - Viewing transaction data and distributions.

  - Predicting fraudulent transactions based on user inputs.

  - Visualizing Precision-Recall Curve and evaluation metrics.

### 3. **Machine Learning**

- Implemented a **Multi-Layer Perceptron (MLP)** model using TensorFlow and Keras.

- Data preprocessing includes scaling and train-test splitting.

- Evaluation using metrics such as **Precision-Recall Curve** and **Area Under Curve (AUC)**.

### 4. **Jupyter Notebook**

- Detailed EDA and visualization.

- Step-by-step walkthrough of data preprocessing and model development.

---

## 🚀 How to Run the Project

### Clone the Repository

```bash

git clone https://github.com/shoaib1522/Credit-Card-Fraud-Analysis.git

cd Credit-Card-Fraud-Analysis

```

### Prerequisites

- Install Python (>=3.8)

- Install required libraries:

  ```bash

  pip install -r requirements.txt

  ```

### Run the Streamlit App

```bash

streamlit run App.py

```

### Jupyter Notebook

- Open `credit-card-fraud-eda.ipynb` using Jupyter or any notebook editor.

---

## 🗂️ Dataset

The dataset used in this project contains transaction data from a Kaggle competition. The key features include:

- **Time**: Seconds elapsed between this transaction and the first transaction in the dataset.

- **Amount**: Transaction amount.

- **Class**: Target variable (1 = Fraud, 0 = Legitimate).

*Note*: Due to size constraints, the dataset file is not included in the repository. You can download it from [Kaggle's Credit Card Fraud Dataset](https://www.kaggle.com/mlg-ulb/creditcardfraud).

---

## 📊 Visualizations and Insights

Here are some key visualizations included in the project:

- **Class Distribution**: Highlights the imbalance between legitimate and fraudulent transactions.

- **Transaction Amount Distribution**: Visualizes the distribution of transaction amounts.

- **Precision-Recall Curve**: Demonstrates the model's performance in detecting fraud.

---

## 🛠️ Tools and Libraries

- **Python**

- **Streamlit**

- **Pandas**

- **NumPy**

- **Matplotlib & Seaborn**

- **TensorFlow & Keras**

- **Scikit-learn**

---

## 📜 License

This project is licensed under the MIT License. See the [LICENSE](LICENSE) file for details.

---

## 📢 Author

Developed by **Shoaib Mughal** as part of the **Data Science Term Project**.

Feel free to explore, contribute, and provide feedback! If you have any questions or suggestions, feel free to reach out.

---