📧 Spam SMS/Email Classifier (with Streamlit App)

A powerful and accurate Machine Learning-based solution that detects whether a given SMS or email is spam or not. This project uses Logistic Regression and TF-IDF vectorization, built using Python, Scikit-learn, Pandas, and Streamlit.

---

📂 Repository Structure

Spam-SMS-Email-Classifier/
│
├── data/
│   └── spam.csv                  # Raw dataset (SMS labeled as ham/spam)
│
├── pkl/
│   ├── model.pkl                 # Trained ML model (Logistic Regression)
│   └── vectorizer.pkl            # TF-IDF vectorizer
│
├── app.py                        # Streamlit app for message prediction
├── spam_classifier_notebook.ipynb  # Jupyter notebook with full model pipeline
├── README.md                     # Project documentation
├── requirements.txt              # Python dependencies
└── .gitignore

---

📊 Features

• Detects whether input message is Spam or Not
• Uses Logistic Regression (option to change model)
• Data preprocessing & text cleaning
• Spam class balancing using upsampling
• TF-IDF feature extraction (top 5000 features)
• User-friendly Streamlit Web App
• Ready for deployment (Render/HuggingFace)

---

📁 Dataset

• Dataset: spam.csv
• Classes:
  • ham: Legitimate messages
  • spam: Fraud, loan, prize messages

---

🚀 How to Run Locally

1. Clone the repository

git clone https://github.com/DevWaqarAhmad/Spam-SMS-Email-Classifier.git
cd Spam-SMS-Email-Classifier

2. Create a virtual environment

python -m venv env

3. Activate the environment

• On Windows:

env\Scripts\activate

4. Install dependencies

pip install -r requirements.txt

5. Run Streamlit app

streamlit run app.py

---

🧠 Model Details

• Vectorizer: TfidfVectorizer with max_features=5000
• Classifier: LogisticRegression (max_iter=1000)
• Balanced Dataset: Using sklearn.utils.resample to upsample spam class
• Accuracy Achieved: ~97-98%

---

🧪 Sample Predictions

Message	Prediction
"Claim your free $5000 loan now. No documents required!"	Spam
"Hey, meeting is at 3 PM today. Don't be late."	Not Spam

---

🖼️ Screenshots

✅ Not Spam Example	🚫 Spam Example

These screenshots show how the Streamlit app classifies input text messages.

🌍 Live Deployment (Coming Soon)

Deployed app link will be added here once hosted on Render/HuggingFace

---

👨‍💻 Author

Waqar Ahmad
📧 Email: devwaqarahmad@gmail.com
🌐 GitHub: DevWaqarAhmad

---

📃 License

This project is licensed under the MIT License.

---

🤝 Contributing

Pull requests are welcome! For major changes, please open an issue first to discuss what you would like to change.

---