Project Abstract

Arguably the most prominent areas of technological advancement in the modern world are Artificial Intelligence and Cyber Security. Malware threats are an ever-increasing issue that cost people, corporations and governments vast amounts of money each year. The main aim of this project was to create an Artificially Intelligent tool that can receive executable files from a user and analyse them for malware, unique in the sense that it solely uses AI. Initially, a large dataset of malicious and legitimate executable samples was curated, this dataset included approximately 140,000 unique file entries and was formatted using the PE (Portable Executable) headers of each file. The data was then preprocessed, allowing a Feed-Forward Neural Network to be trained upon the dataset. Once trained, the Neural Network was vigorously tested, collecting performance statistics to gain an understanding of how well the model can perform. A User Interface was then developed allowing the user to upload an executable file, which would be stripped of its PE data, converted to the same format as the training data, and then parsed to the trained Neural Network which gives the user a verdict on whether the file is malicious or legitimate (of course, a prediction). Ultimately, a Neural Network model was created that can analyse an executable file and assess whether the file is malicious or legitimate, it was then delivered to the user through a web application. A Feed-Forward Neural Network with multiple hidden layers paired with an extensive, thorough dataset produces a detection model that can detect malware to a relatively high standard. Before the beginning of this project, there was no definitive answer to whether a solely AI tool could analyse and detect malware to a high accuracy, now, after the projects completion, there is an insight into what technologies achieve this feat.

Keywords: Artificial Intelligence, Cyber Security, Neural Network

 Conference Details

Session: Poster Session A at Poster Stand 50

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

Markers: Bertie Muller, Liam O’Reilly

Course: BSc Computer Science, 3rd Year

Future Plans: I have a job lined-up