Spam or Ham Classifier

A simple Naive Bayes text classifier that detects whether a message is Spam or Ham (Not Spam).

---

📌 Features

• Preprocesses text (lowercasing + tokenization with regex)

• Builds a vocabulary from training data

• Uses Multinomial Naive Bayes with Laplace smoothing

• Reports accuracy, precision, recall, and F1-score

• Includes an interactive demo to test custom messages

📂 Dataset

This project uses the SMS Spam Collection dataset:

https://www.kaggle.com/datasets/uciml/sms-spam-collection-dataset?resource=download

• The CSV file has two columns:

  • v1 → Label (ham or spam)

  • v2 → Message text

Example:

• v1
  • SPAM
• v2
  • "Congratulations! You've won a $1000 Walmart gift card. Call now!"

---

• v1
  • HAM
• v2
  • "Can we reschedule our meeting to 3 PM tomorrow?"

---

🚀 Usage

1. Clone this repository

2. Run the script:

• python main.py

---

Example output:

• Accuracy: 0.95
• Precision: 0.94
• Recall: 0.91
• F1-Score: 0.92

---

Try the interactive demo:

--- Spam Classifier Demo ---

Enter a message to classify:

"WIN A FREE iPhone! Click now!"

Prediction: SPAM

---

🔮 Future Improvements

Add stopword removal and stemming/lemmatization

Try different classifiers (Logistic Regression, SVM, Neural Networks)

Deploy as a simple web app with Flask/Streamlit

---