Email Spam Shield 🛡️

A machine learning application that detects spam emails using hybrid features and multiple classifiers.

📋 Overview

This project implements a spam email classification system using multiple machine learning models and feature extraction techniques. The application provides a user-friendly interface built with Streamlit for analyzing emails and determining whether they are spam or legitimate.

Features

• Multiple Feature Extraction Techniques:

  • TF-IDF Vectorization
  • Word2Vec Embeddings
  • Hybrid feature combination

• Multiple Classifiers:

  • Logistic Regression
  • Naive Bayes
  • Support Vector Machine (SVM)
  • Random Forest

• Interactive UI:

  • Real-time email analysis
  • Visual representation of results
  • Keyword analysis
  • Performance metrics visualization

🚀 Getting Started

Prerequisites

• Python 3.9+
• pip

Installation

1. Clone the repository:

   git clone https://github.com/your-username/email-spam-shield.git
   cd email-spam-shield

2. Create a virtual environment:

   python3 -m venv venv
   source venv/bin/activate  # On Windows, use: venv\Scripts\activate

3. Install the required packages:

   pip install -r requirements.txt

Usage

1. To run the Streamlit application:

   python3 -m streamlit run app.py

2. To retrain the models using the Jupyter notebook:

   jupyter notebook spamemailclassification.ipynb

🧠 How It Works

The spam classification system works in three main steps:

1. Text Preprocessing:

   • Converting to lowercase
   • Removing special characters
   • Tokenizing into words
   • Removing stop words
   • Lemmatizing words

2. Feature Extraction:

   • TF-IDF: Converts text into numerical values based on word frequency
   • Word2Vec: Neural network model that learns word associations
   • Hybrid Features: Combines both methods for enhanced performance

3. Classification:

   • Multiple machine learning models analyze the features
   • The model with the best F1-score is selected as the final classifier

📊 Performance

The system has been tested on the Enron email dataset, with the following performance metrics:

Model	Accuracy	Precision	Recall	F1-Score
Logistic Regression	~92%	~89%	~85%	~87%
Naive Bayes	~88%	~82%	~91%	~86%
SVM	~93%	~90%	~87%	~88%
Random Forest	~91%	~88%	~84%	~86%

Note: Actual performance may vary based on the most recent training.

📚 Project Structure

email-spam-shield/
├── app.py                       # Streamlit application
├── spamemailclassification.ipynb # Training notebook
├── requirements.txt             # Dependencies
├── best_spam_classifier.pkl     # Best model
├── tfidf_vectorizer.pkl         # TF-IDF model
├── scaler.pkl                   # Feature scaler
├── w2v_model.pkl                # Word2Vec model
└── enrondataset.csv             # Dataset

🤝 Contributing

Contributions, issues, and feature requests are welcome! Feel free to check the issues page.

📜 License

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

🙏 Acknowledgements

• Enron Email Dataset
• Streamlit for the interactive UI framework
• NLTK for natural language processing tools
• Scikit-learn for machine learning models