EyeSpy - AI-Powered Fake News Detection

A cutting-edge multimodal system for detecting fake news and manipulated content using artificial intelligence.

Features

• Text Analysis: Advanced NLP algorithms detect linguistic patterns and misinformation indicators.
• Image Forensics: (Coming Soon) Computer vision techniques to identify digital manipulation and deepfakes.
• Real-time Processing: Lightning-fast analysis with confidence scores.
• Modern UI: Sleek, responsive frontend with dark theme.

Backend Technology Stack

Here's a brief overview of the backend technologies and libraries used in EyeSpy:

• Flask: A lightweight web framework for Python used to build the core API that serves the fake news detection model.
• Flask-CORS: An extension for Flask that handles Cross-Origin Resource Sharing (CORS), making it possible for the frontend to communicate with the backend API.
• scikit-learn: A powerful and easy-to-use machine learning library for building, training, and deploying the fake news classification model.
• Joblib: A set of tools to provide lightweight pipelining in Python. In this project, it's used for saving and loading the trained scikit-learn model and vectorizer.
• PyMongo: The official Python driver for MongoDB, enabling the application to connect to and interact with the database for storing analysis results and training data.
• Pandas: A versatile data analysis and manipulation library, primarily used for loading and processing the datasets during the model training phase.
• Numpy: A fundamental package for scientific computing in Python, used by scikit-learn for numerical operations.
• Pillow: The Python Imaging Library fork, included for future development of image-based analysis and manipulation features.
• Gunicorn: A robust and production-ready web server (WSGI) for deploying the Flask application, ensuring scalability and performance.

Key Components

• Docker: Simplifies the setup of the development environment by containerizing the MongoDB database. The docker_setup.bat script automates the process of starting the database container.
• MongoDB: A NoSQL database used to store the news articles dataset, analysis history, and other application data. The application is configured to connect to a local MongoDB instance, but also supports cloud-based MongoDB Atlas.
• News Datasets: The project includes Fake.csv and True.csv datasets, which are used to train the fake news detection model. These datasets are processed by the scripts in the backend folder.
• Streamlit: An open-source app framework for Machine Learning and Data Science projects. In EyeSpy, modelrun.py uses Streamlit to provide a simple and interactive interface for testing the trained model with custom text inputs.

How to Run

Prerequisites

• Docker
• Python 3

Instructions

1. Start the backend and database:

   Open a terminal and run the following command from the EyeSpy/backend directory:

   ./docker_setup.bat

   This will start a MongoDB container, set up the database, and run the Flask backend server.

2. Start the frontend:

   Open another terminal and run the following command from the EyeSpy/frontend directory:

   python serve.py

   This will start a simple Python web server for the frontend.

3. Access the application:

   Open your web browser and go to http://localhost:8080.

File Structure

C:\USERS\SURAJ\DOWNLOADS\EYESPY\EYESPY

    architecture_diagram.md
    
+---backend

        app.py
        cloud_setup.py
        database.py
        database_extension.py
        dataset_loader.py
        docker_setup.bat
        fake_news_model.ipynb
        model.jb
        modelrun.py
        requirements.txt
        run_server.bat
        run_with_mongodb.bat
        setup_database.py
        vectorizer.jb
        
+---Datasets

        Fake.csv
        True_news.csv
        
+---frontend
        index.html
        serve.py
        test.html

Testing the Model

To test only the model, you can run the modelrun.py file (with streamlit) after running the fake_news_model.ipynb file to generate the model and vectorizer files.

Dependencies

• Flask
• Flask-CORS
• numpy
• scikit-learn
• joblib
• Pillow
• pymongo
• pandas
• gunicorn

All Python dependencies are listed in backend/requirements.txt.

License

MIT License - Feel free to use and modify for your projects.