Spam_Detection

📧 Spam Classifier Web App

A machine learning-powered web application built with Flask that classifies text messages as Spam or Not Spam in real-time. This project demonstrates end-to-end NLP pipeline integration — from text preprocessing to model deployment using Flask.

Features

🧠 Trained ML Model (using scikit-learn)

🔤 TF-IDF Vectorization for text feature extraction

🧹 Text Preprocessing: tokenization, stopword removal, stemming

🌐 Interactive Web Interface built with Flask

⚡ Fast, lightweight, and easy to deploy

Project Structure Spam-Classifier/ │ ├── app.py # Flask backend application ├── model.pkl # Trained ML model ├── vectorizer.pkl # TF-IDF vectorizer ├── spam.csv # Dataset used for training ├── spam_classifier.ipynb # Jupyter notebook (EDA + model training) ├── requirements.txt # Python dependencies └── templates/ └── index.html # Frontend HTML page

🔧 Installation & Setup 1️⃣ Clone the Repository git clone https://github.com/yourusername/spam-classifier.git cd spam-classifier

2️⃣ Create a Virtual Environment python -m venv venv source venv/bin/activate # For Linux/Mac venv\Scripts\activate # For Windows

3️⃣ Install Dependencies pip install -r requirements.txt

4️⃣ Download NLTK Data

Open a Python shell and run:

import nltk nltk.download('punkt') nltk.download('stopwords')

5️⃣ Run the Flask App python app.py

Then open your browser and go to:

http://127.0.0.1:5000/

How It Works

Input Text → User enters a message in the input box.

Text Transformation →

Converts text to lowercase

Tokenizes using NLTK

Removes punctuation and stopwords

Applies stemming

Vectorization → Converts processed text into numerical form using TF-IDF.

Prediction → Model predicts Spam (1) or Not Spam (0).

Output → Result is displayed on the webpage.

Model Details

Algorithm: Multinomial Naive Bayes

Vectorizer: TF-IDF (Term Frequency–Inverse Document Frequency)

Training Dataset: spam.csv (SMS Spam Collection Dataset)

Accuracy: ~97% (depending on preprocessing and split)

Requirements

All dependencies are listed in requirements.txt:

nltk scikit-learn Flask Flask-Cors pandas numpy

Install them via:

pip install -r requirements.txt

UI Preview (Example)

| 📧 Enter your message below: | | [ Hey, you won $1000! Click here ] | | | | [ Predict ] |

Result: 🚨 Spam

🛠️ Future Enhancements

Add email spam classification support

Integrate REST API endpoints for external access

Use deep learning models (LSTM/BERT) for better accuracy

Deploy on Render / Heroku / AWS / Railway

References

NLTK Documentation

Scikit-learn Documentation

Flask Documentation

Author

Ankan Pal 📍 IIT Patna | Backend Developer | AI & ML Enthusiast