📧 Spam Mail Detection using Machine Learning

This project builds a Spam Mail Detection model using Natural Language Processing (NLP) and Machine Learning techniques.

The model classifies messages as:

• Spam (0)
• Ham (1)

It uses TF-IDF Vectorization and Logistic Regression for classification.

---

🚀 Technologies Used

• Python
• NumPy
• Pandas
• Scikit-learn
• Matplotlib
• Seaborn

---

🧠 Machine Learning Workflow

1. Load dataset
2. Data cleaning & handling null values
3. Label encoding (Spam = 0, Ham = 1)
4. Train-test split (80-20)
5. Text feature extraction using TF-IDF
6. Model training using Logistic Regression
7. Model evaluation (Accuracy, Confusion Matrix)
8. Custom message prediction

---

📊 Model Performance

• Training Accuracy: ~97%
• Testing Accuracy: ~96%

(Accuracy may slightly vary depending on random_state.)

---

🧪 Example Prediction

Input: "Congratulations! You have won a $1000 gift card. Click the link now!"

Output: Spam

---

📂 Project Structure

spam-mail-detection/ │ ├── spam_mail_detection.ipynb ├── mail_data.csv └── README.md

---

🎯 Key Learnings

• Text preprocessing
• TF-IDF feature extraction
• Logistic Regression implementation
• Model evaluation techniques
• Real-world text classification workflow

---

👨‍💻 Author

Atiur Rahaman
AIML Student | Machine Learning Enthusiast