MichaelAquilina / SpamFilter

Classification of emails using machine learning and natural language processing techniques in Java

Spam Filter

Authors:

• Michael Aquilina
• Uwe L. Korn

Project for "Introduction to Machine Learning" course.

Requirements can be found in the following pages:

• Spam Filter Implementation
• Training and Classification

Results

• Current terms selected from feature selection: https://gist.github.com/KillaW0lf04/d430834e07b4e7aa3901
• Json format for saving EmailClassifier: https://gist.github.com/KillaW0lf04/2ec98c3d2ad29085fcd1\
• Most representative features detected by the Naive Bayes classifier: https://gist.github.com/KillaW0lf04/ddb0871769b75d37b49e

Design

A design wiki has been set up to establish a common understanding of components and milestones needed to complete the project. This should also help us when it comes to writing the report for hand in.

Current Machine Learning Models

• Perceptron
• Naive Bayes
• Decision Tree

Third Party Code

• Porter Stemmer Algorithm in java
• AbstractAdapter is based on this stackoverflow post