📧 Spam–Ham Classifier

A complete Machine Learning + NLP project to classify messages as Spam or Ham. The project includes data preprocessing, Bag of Word vectorization, training multiple ML models, model comparison, ROC & confusion matrix plots and Streamlit deployment.

---

📘 1. Project Overview

• Classifies SMS/Email messages as spam or ham.
• Preprocessing includes: cleaning text, removing stopwords, tokenization.
• Converts text into numerical features using Bag of Word vectorization.
• Trains multiple machine learning models and evaluates them using cross-validation.
• Selects the best-performing model automatically.
• Provides visualizations: Confusion Matrix, ROC Curve.
• Deployable via Streamlit for interactive prediction.

---

---

🔄 2. Pipeline Workflow

1. Load Dataset → preprocessing/load_data.py
2. Clean Text → preprocessing/clean_text.py
3. Vectorization (BOW) → preprocessing/vectorize.py
4. Train Multiple Models → preprocessing/train.py
5. Evaluate Models → preprocessing/evaluate.py
6. Select Best Model & Save → models/best_spam_model.pkl + models/vectorizer.pkl
7. Visualize Metrics → Confusion Matrix, ROC
8. Deploy with Streamlit → app.py

---

📊 3. Models Used & Comparison

Model	Cross-Validated Accuracy
Multinomial Naive Bayes	0.9758
K-Nearest Neighbors	0.9026
Gradient Boosting	0.9589
AdaBoost	0.9147

🏆 4. Best Model

Multinomial Naive Bayes

Saved as: models/best_spam_model.pkl

---

📈 5. Evaluation Metrics & Graphs

Metrics calculated for test data:

• Accuracy : 0.9749
• Precision : 0.8961
• Recall : 0.9200
• F1 Score : 0.9079

Visualizations Saved in reports/plots/:

1. Confusion Matrix

2. ROC Curve

---

🛠️ 6. Installation

# Clone the repo
git clone https://github.com/roshan-acharya/SpamClassifier
cd SpamClassifier

# Create virtual environment (optional)
python -m venv venv
source venv/bin/activate  # Linux/macOS
venv\Scripts\activate     # Windows

# Install dependencies
pip install -r requirements.txt

🔄 7. Running the Project

• Run Training Pipeline

  python pipeline/pipeline.py

• Run Streamlit App

  streamlit run app.py

🧰 8. Tech Stack

• Python

• Pandas

• NumPy

• Scikit-learn

• Matplotlib, Seaborn

• Streamlit

• Pickle (for saving models)

👨‍💻 9. Author

Roshan Acharya
AI/ML Enthusiast