cytical/cs180project

Background

Spam text messages are a rampant concern in the Philippines

• 1 in 4 Filipinos become a victim of schemes like spam messages (National ICT Household Survey, 2019)
• Alleged correlation with contact tracing forms
• Globe has blocked 1.3 billion scam texts during a 9-month span in 2022

Goal

This project aims to contribute to the solution by creating a machine learning model to accurately detect scam/spam text messages in the Philippines.

Method

The model was trained with the Naive-Bayes classification algorithm using existing datasets with labeled SMS messages for text classification

Results

The accuracy score of the initial model with default parameters is 99.28%
The accuracy score of the optimized model after hyperparameter tuning is 99.50%

Running Our App Locally

cd DeployModel
python manage.py runserver

Your Django application is now available at http://localhost:8000.