SciFest National Final 2024

Stand 29

Stand 29

Machine Learning Powered Intruder Detection System in the Context of our Medical Appointment Computing System Software (MACSS)

Student Max Grogan
School St Andrew’s College, Booterstown Avenue, Blackrock, Co. Dublin
Teacher Laura Brogan
Venue TU Dublin Grangegorman
O
Abstract

This project aimed to develop the best Intruder Detection System (IDS) for the Medical Appointment Computing System Software (MACSS), designed to address Ireland's 10% of people who lack access to General Practitioners due to inaccessibility, lack of funds and lack of user organization. MACSS aims to increase accessibility to GPs across Ireland and manage their appointments, invoices, medical history, prescriptions, and test results thus improving the quality of the Irish healthcare system.

The project involved creating a demo website, which has received positive feedback from a GP, 50 user testers, and five readability tests.

The IDS was created using the KDD Cup 1999 dataset. The data was then processed by Reading Data, Data Processing, Categorical features, distribution, data correlation, feature mapping and modelling. Seven algorithms were tested: Gaussian Naïve Bayes, Decision Tree, Random Forest, SVM, Logistic Regression, Gradient Boosting, and ANN. The Decision Tree model was found to be the best machine learning model for the IDS and further development was made.

The research involved extensive investigation into the state of the Irish healthcare system, interviews and articles about IDS functions. The effectiveness of the software was proven through user testing, readability testing and GP input. Training time, accuracy and testing confirmed that the Decision Tree model was the best model for the IDS, and data processing results further verified its effectiveness.

Check out the software demo here: www.macss.ie
Poster Click here
Enquire
Contact Us
+353 86 379 6143



SciFest National Final 2024
© 2025 SciFest National Final 2024