Email Spam Classifier

Overview

This project is an Email Spam Classifier that uses the Naive Bayes algorithm to distinguish between spam and non-spam emails. The model is trained on a labeled dataset and achieves high accuracy, making it effective for real-world applications.

The project demonstrates the end-to-end process of building a machine learning model, from data preprocessing and feature extraction to model training, evaluation, and deployment.

---

Features

• Spam Detection: Classifies emails as spam or non-spam.
• High Accuracy: Achieves impressive performance on the test data.
• Preprocessing Pipeline: Includes text cleaning, tokenization, and feature extraction.
• Streamlit Deployment: Provides a user-friendly web interface for real-time spam detection.

---

Technologies Used

The following technologies and libraries were used in this project:

• Programming Language: Python
• Machine Learning Algorithm: Naive Bayes (MultinomialNB from scikit-learn)
• Libraries:
  • scikit-learn: For implementing the Naive Bayes classifier and evaluating the model.
  • NumPy: For numerical computations.
  • Pandas: For data manipulation and analysis.
  • NLTK (Natural Language Toolkit): For text preprocessing (e.g., tokenization, stopword removal).
  • Matplotlib & Seaborn: For visualizing results and data distributions.
  • Streamlit: For deploying the model as an interactive web application.

---

Dataset

The dataset used in this project contains labeled emails with two categories:

• Spam: Emails that are irrelevant or unsolicited.
• Non-Spam (Ham): Legitimate emails.
• Dataset Download link: https://www.kaggle.com/datasets/uciml/sms-spam-collection-dataset

Preprocessing Steps:

1. Text Cleaning: Removal of special characters, numbers, and punctuation.
2. Tokenization: Splitting text into individual words.
3. Stopword Removal: Removing common words that do not contribute to classification.
4. Vectorization: Converting text into numerical features using techniques like TF-IDF or Count Vectorization.

---

Implementation

1. Install Dependencies

Ensure you have the required libraries installed:

pip install scikit-learn numpy pandas nltk matplotlib seaborn streamlit

2. Training the Model

The Naive Bayes algorithm was chosen due to its simplicity and effectiveness for text classification tasks. The model was trained using scikit-learn's MultinomialNB implementation.

3. Evaluation Metrics

The model's performance was evaluated using:

• Accuracy
• Precision

4. Results

The classifier achieved the following metrics:

• Accuracy: 95.93%
• Precision: 100%

---

Deployment

The project includes a Streamlit application for deploying the spam classifier as a web app. Users can input email text, and the model will classify it as spam or non-spam in real time.

Run the Streamlit App

1. Navigate to the project directory.
2. Run the following command:

streamlit run app.py

3. Open the provided URL in your web browser to access the application.

---

Usage

Clone the Repository

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

Run the Project

1. Place your dataset in the appropriate folder.
2. Train the model and make predictions by running:

python spam_classifier.py

3. Deploy the Streamlit app:

streamlit run app.py

---

Visualization

Key visualizations included in the project:

• Word Cloud: Displays the most frequent words in spam and non-spam emails.
• Confusion Matrix: Highlights the model's performance on the test data.

---

Future Work

• Enhancing the Streamlit app with additional features, such as file upload for bulk email classification.
• Experimenting with advanced algorithms like Support Vector Machines (SVM) or random forest or deep learning models.

---

Contributing

Contributions are welcome! Feel free to open an issue or submit a pull request.

---

Contact

For any inquiries, please contact: Shubham jain
Email: [email protected] GitHub: Shubham-Jain52