Publicação
Predictive Modeling of Real Estate Prices in Portugal: Integrating Property Features for Accurate Valuation
| Resumo: | This study is a comprehensive exploration of the application of machine learning algorithms in predicting real estate prices in Portugal. It tackles the challenges posed by the complex and dynamic nature of the real estate sector, leveraging advanced analytical tools to enhance prediction accuracy. The study develops and compares five distinct machine learning models—Multilinear Regression, Regression Trees, Random Forest, Extreme Gradient Boosting (XGBoost), and Neural Networks—to identify the most effective real estate price prediction algorithm. The methodology involves data preparation, feature engineering, and model evaluation using three widely accepted evaluation metrics: Root Mean Squared Error (RMSE), R-squared (R2), and Mean Absolute Error (MAE). The findings indicate that XGBoost outperforms other models, providing superior accuracy and generalization with a mean absolute error of approximately 47 943 euros and an R2 value of 0.76. The study demonstrates how machine learning can address the intricacies of the real estate market but also offers practical, data-driven insights to support informed decision-making, thereby contributing to the academic understanding of market dynamics. |
|---|---|
| Autores principais: | Costa, Raimundo Mujica |
| Assunto: | Machine Learning Real Estate Property Prices Predictive Modeling Housing Prices SDG 1 - No poverty SDG 9 - Industry, innovation and infrastructure SDG 10 - Reduced inequalities |
| Ano: | 2024 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade Nova de Lisboa |
| Idioma: | inglês |
| Origem: | Repositório Institucional da UNL |
| Resumo: | This study is a comprehensive exploration of the application of machine learning algorithms in predicting real estate prices in Portugal. It tackles the challenges posed by the complex and dynamic nature of the real estate sector, leveraging advanced analytical tools to enhance prediction accuracy. The study develops and compares five distinct machine learning models—Multilinear Regression, Regression Trees, Random Forest, Extreme Gradient Boosting (XGBoost), and Neural Networks—to identify the most effective real estate price prediction algorithm. The methodology involves data preparation, feature engineering, and model evaluation using three widely accepted evaluation metrics: Root Mean Squared Error (RMSE), R-squared (R2), and Mean Absolute Error (MAE). The findings indicate that XGBoost outperforms other models, providing superior accuracy and generalization with a mean absolute error of approximately 47 943 euros and an R2 value of 0.76. The study demonstrates how machine learning can address the intricacies of the real estate market but also offers practical, data-driven insights to support informed decision-making, thereby contributing to the academic understanding of market dynamics. |
|---|
Atividades financiadas
Não foram encontrados projetos financiados.