Faye Belcher (2115286) Faye Belcher

Spam Identification using Machine Learning and Homomorphic Encryption

Project Abstract

My project is about spam filtering and combining it with encrypted data. �?I took this project to improve my skills in machine learning and learning encryption techniques. This project can benefit the outside world by proving that machine learning techniques can be used with encrypted data.�?�?�?�?�?�?�?�?The main aim of this project was to find the best method to see encrypted data. The project’s design was to have pre-encrypted data and use specific models to determine the most effective.�?�?The primary method employed was clustering, which assigns the data into designated classes to a given sample.�?For my final project, I produced software that employed machine-learning techniques on encrypted data.�?In my result, the best one to use was a random forest with around 93% accuracy. �?�?In conclusion, it is imperative that machine learning leverages encrypted data. With the escalating demand for cybersecurity, the encryption of personal information is no longer a luxury but a necessity. Moreover, this approach should be adopted for category and spam identification, given that most issues stem from the challenge of identifying crucial emails. �?�?�?

Keywords: Machine Learning, Homomorphic Encryption, Classification Model, Spam Identification,, ,

 

 Conference Details

 

Session: Poster Session A at Poster Stand 103

Location: Sir Stanley Clarke Auditorium at Tuesday 7th 13:30 – 17:00

Markers: Hans Ren, C�cilia Pradic

Course: BSc Software Engineering, 3rd Year

Future Plans: I’m looking for work