Spam-SMS-Detector

A Flask-based spam detection application that classifies messages as spam or not spam. The model is trained on the Kaggle SMS Spam dataset and provides a simple web interface for predictions.

Features

• Train a machine learning model using Kaggle's spam dataset
• Web interface to enter messages and check if they are spam
• Flask backend for handling predictions
• Preprocessing of text data for better accuracy

Installation

1. Clone this repository:

   git clone https://github.com/Maithilya/Spam-SMS-Detector.git
   cd Spam-SMS-Detector

2. Install dependencies:

   pip install -r requirements.txt

3. Download Dataset :

   https://www.kaggle.com/datasets/uciml/sms-spam-collection-dataset

Steps to Run the Project

1 Ensure you have Python installed (preferably Python 3.12 or later).

2 Navigate to the project directory:

cd Spam-SMS-Detector

3 Run the Flask app:

python Spam-SMS-Detector.py

4 Open http://127.0.0.1:5000 in your browser. 5 Enter a message in the provided text box and check for spam classification.

Importance of index.html

The index.html file is crucial as it serves as the front-end interface for users to enter messages and view classification results. Flask uses Jinja2 templating, which requires index.html to be inside a templates/ directory. If the file is missing or placed incorrectly, the application will throw a TemplateNotFound error.

Common Errors & Fixes

1. TemplateNotFound: index.html

Fix: Ensure you have an index.html file inside a templates/ folder:

mkdir templates

Then move your index.html file inside the templates/ folder.

2. SystemExit: 1 during Flask run

Fix: Check if another instance of Flask is already running. Stop it using:

taskkill /F /IM python.exe

Then restart the app.

3. Flask Debug Mode Restarting Issues

Fix: If Flask keeps restarting or crashing, run it without debug mode:

python spam_detector.py --no-debug

(NOTE: everything should be in same folder)

Future Improvements

• Deploy on a cloud platform (Heroku, Render, or AWS)
• Improve model accuracy with advanced NLP techniques
• Add user authentication for secured access

Contributing

Feel free to fork this repo and improve the project! If you face any issues, open an issue on GitHub.

---

Author: Maithilya Patle 🚀