Supervisor
Sam Weiss
Programme
MSc in Data Analytics
Subject
Computer Science
Abstract
This research presents a comparative analysis of machine learning and neural network performance in predicting house prices across Ireland’s regional housing markets. It addresses important methodological challenges and offers new empirical insights into how market complexity influences algorithm accuracy and performance. Drawing on 627,294 residential transactions from the Irish Property Price Register (2012–2024), the study applies a dual validation strategy, temporal and stratified sampling, across four regional classifications: Dublin, Other Cities, the Commuter Belt, and Rural areas.
The study makes three main contributions to data analytics theory and practice. First, it identifies and resolves temporal confounding effects in algorithm evaluation. A 39% inflation gap between training and testing data periods was found to distort performance evaluations, while stratified sampling improved results by 59–840% across models. This approach restored Linear Regression from failure to an R² = 0.318 and it enabled neural networks to reach competitive accuracy (R² = 0.33) alongside traditional machine learning methods.
The research reveals a counterintuitive relationship between sample size and predictive accuracy. Despite having the largest regional dataset (39,015 transactions), Dublin consistently produced the weakest results (R² = 0.096), whereas the Commuter Belt achieved the highest accuracy (R² = 0.231) with the smallest sample (19,763 transactions). These results support market complexity theory, showing that urban market factors can outweigh the benefits of data volume.
Third, the study develops a regional algorithm deployment framework theory that accounts for fundamental differences in market structure. Feature importance analysis revealed near-independence across the four regions (ρ = 0.013–0.354), suggesting that distinct pricing factors operate in different geographic areas. These insights contribute to both academic understanding and practical modelling strategies, with direct impact on automated valuation systems and housing policy design.
Date of Award
2025
Full Publication Date
2025
Access Rights
open access
Document Type
Capstone Project
Resource Type
thesis
Recommended Citation
Carroll, D.
(2025) A Comparative Analysis of Machine Learning and Neural Network Performance in House Price Prediction: Dublin Vs. Other Irish Regions CCT College Dublin.
DOI: https://doi.org/10.63227/652.299.79