AI Spam Filter

A Naïve Bayes classifier algorithm for detecting spam emails/messages, written from scratch using Bayes Theorem.

Table of Contents

1. About The Project
   • Built With
2. Contact

About The Project

I built this project for an assigment at univerity, where the aim was to design a Naïve Bayes classifier algorithm without using scikit-learn or similar libraries. I used a large database of labelled examples of spam and ham messages/emails to train, test and validate the functionality and accuracy of the model.

In this repository, you can find the Jupyter notebook, which contains lab notes on the steps I took to build this algorithm.

Beyond the algorithm, I have also implemented the algorithm within my AI tools website: myaitools.pythonanywhere.com

You can try the spam classifier tool on this site. Bootstrapped with Flask, I built the site as a "hub" for a range of AI projects I wanted to try. With this site, I can easily share tools I've made with others.

To protect against academic malpractice/copyright, I have not included the database for this project in this repository. For alternative databases, you could use some of these:

• https://huggingface.co/datasets/TrainingDataPro/email-spam-classification
• https://www.kaggle.com/datasets/cavinlobo/email-classification-dataset

(back to top)

Built With

• Python
• Jupyter Notebook
• Flask
• HTML/CSS
• Javascript

(back to top)

Contact

Affan - affank1020@gmail.com

Project Link: https://github.com/affank1020/ai-spam-filter

(back to top)