Fake News Detection Using Machine Learning

This project is a simple implementation of a fake news detection model using Python, scikit-learn, and the PassiveAggressiveClassifier. The model can classify news articles as either REAL or FAKE based on their content.

Dataset

The project uses two datasets:

• true.csv: Contains true news articles.
• fake.csv: Contains fake news articles.

Project Structure

• true.csv: CSV file containing true news articles.
• fake.csv: CSV file containing fake news articles.
• index.py: The main Python script for training and testing the model.
• model.pkl: The saved machine learning model (optional, created after running the script).
• vectorizer.pkl: The saved TF-IDF vectorizer (optional, created after running the script).
• README.md: Project documentation.

Installation

1. Clone the repository:

   git clone https://github.com/surya2821/machine.git

2. Install the required Python packages:

   pip install pandas numpy scikit-learn

3. Add your datasets (true.csv and fake.csv) to the project directory.

Usage

1. Run the index.py script to train the model:

   index.py

2. The script will:

   • Combine the true.csv and fake.csv datasets.
   • Train a PassiveAggressiveClassifier to detect fake news.
   • Display the model's accuracy and confusion matrix.
   • Save the trained model and vectorizer to model.pkl and vectorizer.pkl.

3. To predict whether a new piece of text is real or fake, use the predict_fake_news function in the script:

   example_text = "Your article text here"
   prediction = predict_fake_news(example_text)
   print(f"Prediction: {prediction}")

Example Output

After running the script, you should see output similar to:

Datasets loaded successfully.
Datasets combined.
Dataset shuffled.
(44898, 5)
                 title  ... label
0  You Can Smell It: Donald Trump Jr....  ...  REAL
1  Is It An NFL Sunday If Odell Beckham...  ...  FAKE

Accuracy: 92.74%
Confusion Matrix:
[[5897  303]
 [ 334 5829]]
Model and vectorizer saved.