Predict Condominium Prices in Bangkok Based on Ensemble Learning Algorithm with various factors
Conference proceedings article
Authors/Editors
Strategic Research Themes
Publication Details
Author list: Anchaleechamaikorn T.; Lamjiak T.; Thongpe T.; Thiralertpanit L.; Polvichai J.
Publisher: Institute of Electrical and Electronics Engineers Inc.
Publication year: 2023
ISBN: 979-835032641-3
Languages: English-Great Britain (EN-GB)
Abstract
Real estate valuation is a critical process that determines the monetary value of a property. Three common approaches to property valuation are the sales comparison approach, the income approach, and the cost approach. However, these methods can be time-consuming, costly, and may suffer from human bias. This paper proposes an Artificial Intelligence (AI)-based system to evaluate the price of condominiums faster, more cost-effectively, and without human bias. We present a comprehensive methodology that utilizes web scraping and machine learning techniques to predict condominium prices accurately. The dataset is obtained through Python scripts that extract information on all condominiums listed on a popular Thai property website. The collected data covers appraisal years from 2008 to 2019. The AI models used in this study include linear regression, Random Forest Regressor, Gradient Boosting Regressor, and XGBoost Regressor models. We evaluated the performance of these models using the MAPE metric. Our results show that machine learning models can predict condominium prices accurately, providing valuable insights into the real estate market in Bangkok. This research contributes to the existing literature by highlighting the importance of incorporating various features and using machine learning models to predict condominium prices accurately. ฉ 2023 IEEE.
Keywords
MAPE metric, Predict condominium prices, Real estate valuation, web scraping
Contact Information