Spam Email Checker

This is a machine learning-based web application that classifies emails as SPAM or NOT SPAM using natural language processing (NLP). The application is built with Flask and includes a simple web interface for users to input email text and receive predictions.

---

File Structure

project-folder/
├── app.py                # Main Flask application
├── dataset/              # Folder containing the dataset
│   └── email.csv         # Dataset file (spam email data)
├── spam_model.pkl        # Trained ML model
├── tfidf_vectorizer.pkl  # Trained TF-IDF vectorizer
├── templates/            # Folder containing HTML templates
│   └── index.html        # Web interface template
└── README.md             # Project documentation

---

Getting Started

Prerequisites

• Python 3.7+
• Required Python packages:
  • Flask
  • scikit-learn
  • pandas
  • joblib

Install the required packages:

pip install flask scikit-learn pandas joblib

---

Setup Instructions

1. Clone the repository or download the project files.

   git clone https://github.com/yourusername/spam-email-checker.git
   cd spam-email-checker

2. Place your dataset (email.csv) in the dataset/ folder.

3. Train the model (if not already trained):

   • Open a Python script or notebook.
   • Load email.csv for training.
   • Save the trained model as spam_model.pkl and the TF-IDF vectorizer as tfidf_vectorizer.pkl in the project root directory.

   If you already have these files (spam_model.pkl and tfidf_vectorizer.pkl), skip this step.

4. Run the Flask application:

   python app.py

5. Open your web browser and navigate to http://127.0.0.1:5000.

---

Using the Application

Webpage Screenshot

Below is a screenshot of the web interface:

1. Enter email text in the input field on the web interface.
2. Click the Check button.
3. View the result: The app will display whether the email is classified as SPAM or NOT SPAM.

---

Dataset

The dataset (email.csv) should contain email text and labels (e.g., 0 for NOT SPAM and 1 for SPAM). If you're using your own dataset, ensure it is preprocessed appropriately.

---

Model Training

The model is trained using:

• TF-IDF Vectorizer: Converts email text into numerical features.
• Logistic Regression: Used for binary classification.

Training Steps:

1. Preprocess the text by removing special characters, converting to lowercase, and tokenizing.
2. Vectorize the text using TF-IDF.
3. Train the Logistic Regression model.
4. Save the model and vectorizer using joblib.

---

Technologies Used

• Python
• Flask: For building the web application.
• scikit-learn: For machine learning and text processing.
• HTML/CSS: For the front-end web interface.

---

Future Enhancements

• Improve model accuracy by exploring advanced NLP techniques (e.g., word embeddings or deep learning).
• Add more user-friendly features to the web interface.
• Deploy the application to a cloud platform (e.g., Heroku, AWS, or Google Cloud).

---

License

This project is open-source and available under the MIT License.

---

Author

Virendra Ghule