mikoweb/php-spam-classification

Spam classification application written in PHP and the Rubix ML library. The Random Forest classifier was used for classification.

php-spam-classification

The aim of the project is to create an application for classifying spam in emails. The application will analyze the content of emails and assign them a label indicating whether they are spam or not.

The application was written using, among others: Rubix ML library and Random Forest classifier.

The REST API was made in Symfony and the frontend in Angular.

Most important classes

• LearnerFactory
• SpamModelTrainer
• SpamModelReport
• SpamModelTester
• Constant
• SpamDatasetReader

Example of using prediction

For a complete usage example, see the class:

• AskForPredictionHandler

Commands

Available commands for the "app" namespace:
  app:ml:train   Command to training a spam classification model.
  app:ml:report  Command to generate a report about the model.
  app:ml:test    Command to testing a spam classification model.

Demo

The application demo is available at:

https://spam-classification.mikoweb.usermd.net/

Copyrights

Copyright © Rafał Mikołajun 2024.