Fake News Detection ML Model and Web App

Overview

This project aims to detect fake news using machine learning techniques and provide a web application interface for users to interact with the model. The model achieves an accuracy of 94.7% using logistic regression and lemmatization.

Dataset

The dataset used for training the model is called WELFake. It consists of 72,134 news articles, with 35,028 labeled as real and 37,106 labeled as fake. The dataset is a combination of four popular news datasets (Kaggle, McIntire, Reuters, BuzzFeed Political) to prevent overfitting and improve model generalization. Each entry in the dataset contains a serial number, title, text, and label (0 for fake news and 1 for real news). It is published in IEEE Transactions on Computational Social Systems (doi: 10.1109/TCSS.2021.3068519).

Backend

The backend of the project is implemented using Django. Here's an overview of the directory structure:

fake_news_detection_project/
│
├── fake_news_detection_project/
│ ├── settings.py
│ ├── urls.py
│ └── ...
│
└── news_detection/
├── migrations/
├── models.py
├── urls.py
├── views.py
└── ...

• dataset/News_dataset.csv: The dataset used for training the machine learning model.
• ML Model/: The trained machine learning model.
• README.md: The file you are currently reading.
• requirements.txt: The list of dependencies required to run the web application.

Prerequisites

To run this web application locally, you need to have the following dependencies installed:

• Django
• joblib
• scikit-learn
• nltk

You can install these dependencies by running the following command:

pip install -r requirements.txt

The fake_news_detection_project directory contains the Django project settings and configurations. Inside this directory, a Django app named news_detection is created to handle the functionalities related to fake news detection.

To save the model, it should be placed inside the news_detection folder. You can modify the settings of the model in news_detection/models.py.

Instructions

1. Ensure you have Python and Django installed on your system.
2. Clone this repository.
3. Navigate to the project directory.
4. Run the Django development server.

   python manage.py runserver

Access the web application through the provided URL. ##Credits:-

• Dataset: WELFake (IEEE Transactions on Computational Social Systems, doi: 10.1109/TCSS.2021.3068519)
• Framework: Django

Feel free to explore the code and contribute to improving the project!