Automated Fake Social Media Account Detection Tool

Overview

The Automated Fake Social Media Account Detection Tool is a comprehensive solution that identifies and flags fake profiles across platforms such as Instagram, Facebook, Twitter, and LinkedIn. Combining a machine learning backend with an intuitive frontend interface, this project provides users with a seamless way to analyze and report suspicious accounts.

---

Frontend

Features

• Machine Learning Integration: Interacts with backend APIs to leverage trained ML models for detecting fake accounts.
• User-Friendly Interface: Built with React.js, the frontend offers a clean and intuitive user experience.
• Scalable Architecture: Optimized for performance and scalability, ensuring smooth operation even with high traffic.
• Navigation and Dashboard: Provides quick access to tools like account verification, reporting, and an admin panel.
  • Dashboard Statistics:
    • Accounts Verified
    • Fake Accounts Detected
    • Reports Processed
    • Success Rate

Installation and Usage

1. Clone the repository:

   git clone https://github.com/pnnv/fotodile
   cd fotodile

2. Install dependencies:

   npm install
   npm install lucide-react@0.263.1

3. Start the development server:

   npm run dev -- --open

---

Backend

Features

Machine Learning Algorithms

• Utilizes classification models like Random Forest, AdaBoost, Decision Tree, Logistic Regression, and SVM for high-accuracy predictions.
• Processes account data to classify profiles as real or fake.

Feature Engineering

• Scrapes and analyzes key account features:
  • Profile Picture: Checks for presence or absence.
  • Username Patterns: Examines length, format, and keywords.
  • Follower-to-Following Ratio: Compares numerical metrics.
  • Account Activity: Measures post frequency and engagement levels.
  • Verification Status: Identifies whether the account is verified.

Centralized Reporting System

• Flags detected fake accounts and coordinates actions like suspension or deletion with social media platforms.

API Integration

• Built with Flask and FastAPI to handle data processing and predictions.
• Deployed on platforms like Heroku or AWS for scalability.

Workflow

1. Input Form: Users submit social media account links for analysis.
2. Data Scraping: Extracts account features using Python-based HTML parsing.
3. Feature Vector Generation: Converts extracted data into structured inputs for ML models.
4. ML Classification: Predicts the likelihood of an account being fake.
5. Results Visualization: Outputs results in a tabular format with interactive visualizations.
6. Reporting: Sends flagged accounts to a central agency for further action.

Technical Architecture

• Frontend: React.js
• Backend: Flask and FastAPI
• Machine Learning: Algorithms include Random Forest, AdaBoost, Logistic Regression, Decision Tree, KNN, and SVM.
• Deployment: Hosted on Heroku or AWS, with GitHub for version control.

Installation

1. Clone the repository:

   git clone https://github.com/pnnv/fotodile
   cd fotodile

2. Install dependencies:

   cd backend
   pip install -r requirements.txt

3. Run the application:

   python app.py

---

Contribution

Contributions are welcome! Follow these steps:

1. Fork the repository.
2. Create a new branch:

   git checkout -b feature-branch-name

3. Commit your changes and push the branch:

   git push origin feature-branch-name

4. Open a pull request.

---

License

This project is licensed under the MIT License. See the LICENSE file for details.

---

Contact

For queries or support, reach out through the Issues section.