Fake News Detection

Artificial Intelligence for Fake News Detection

This project utilizes machine learning models to classify news articles as either real or fake. It applies Natural Language Processing (NLP) techniques to preprocess text, extract features, and train multiple classification models to achieve high accuracy.

Table of Contents

1. Dataset and Preprocessing
2. Feature Engineering
3. Data Visualization
4. Model Training
5. Evaluation and Performance
6. Prediction on New Data

---

Dataset and Preprocessing

The dataset used (final_en.csv) consists of news articles labeled as real (0) or fake (1). To enhance accuracy, preprocessing includes:

• Converting text to lowercase
• Tokenization and lemmatization using spaCy
• Removing stop words, common, and rare words

The text features are extracted from both the title and text columns to form a processed dataset.

Feature Engineering

After preprocessing, we use the TF-IDF Vectorizer to transform text into numerical vectors. Key steps include:

• Setting n-gram range (1,2)
• Selecting a maximum of 5000 features
• Removing English stop words

Data Visualization

To gain insights into the dataset, we use:

• Seaborn count plot to visualize label distribution
• Word clouds to show common words in fake and real news articles

Model Training

We train multiple machine learning models to compare performance:

• Logistic Regression
• Naïve Bayes
• Random Forest
• Support Vector Machine (SVM)
• XGBoost
• Decision Tree
• AdaBoost
• LightGBM
• Multilayer Perceptron (MLP Neural Network)

Each model is trained on 80% of the dataset and tested on the remaining 20%.

Evaluation and Performance

The models are evaluated based on accuracy and a classification report. The results are displayed in descending order of performance to determine the most effective model.

Prediction on New Data

A function predict_news() allows predictions on new articles. Example usage:

new_news = ["Breaking news: Stock market crashes due to global tension!",
            "Scientists discover a new method to reverse climate change."]
predictions = predict_news(models, vectorizer, new_news)
for model, pred in predictions.items():
    print(f"{model} Predictions: {pred}")

This project provides a solid foundation for detecting fake news using AI models. Further enhancements could involve deep learning techniques or incorporating external fact-checking sources for improved accuracy.