House Price Prediction Based on Machine Learning: A Case of King County

Abstract

This paper focuses on formulating a feasible method for house price prediction. A dataset containing features and house price of King County in the US is used. During the data preprocessing, extreme values are winsorized and highly correlated features are removed. Eight models including Catboost, lightGBM and XGBoost serve as candidate models. They are evaluated by several evaluation indicators, including rooted mean square error, R-squared score, adjusted Rsquared score and K-fold cross validation score. The model that has low RMSE, achieves a high R-squared score and adjusted R-squared score, especially in the test set, and acquires a high score in cross validation is considered a better model. This paper finds out that Catboost performs the best among all models and can be used for house price prediction. Location, living space and condition of the house are the most important features influencing house price. After comparison and contrast with other papers, it is attested that findings in this paper conform to real life. This paper formulates a model that fits better than preceding studies for house price prediction and makes necessary supplement to the exploration of features that influence house price from a microscope.

A Hybrid Approach for Mass Valuation of Residential Properties through Geographic Information Systems and Machine Learning Integration

“…There is also a remarkable increase in the use of Machine Learning methods in housing price prediction applications. Ensemble Machine Learning methods like Random Forest (Dimopoulos et al 2018; Yilmazer and Kocaman 2020; Aydinoglu, Bovkir, and Colkesen 2021; Ho, Tang, and Wong 2021), XGBoost (Peng, Huang, and Han 2019; Li 2022), CatBoost (Wang, Wang, and Liu 2021; Wang and Zhao 2022), LightGBM (Truong et al 2020; Liu et al 2021) provide more accurate predictions by aggregating outputs from several learners. Yet, ensembles are less interpretable, can be computationally expensive, and lack considering spatial effects of determinants during the prediction process.…”

Advanced Machine Learning Techniques for Predictive Modeling of Property Prices

“…The study in [11] developed Bayesian deep learning approaches to represent uncertainty in property valuation for specific regions. The study in [12] explored the impact of vegetation on residential property values, highlighting the importance of incorporating localized factors. These studies resonate with the objective of incorporating spatial and regional features to enhance predictive performance in the UK real estate market.…”

“…While many studies have focused on property-specific and spatial features, the integration of macroeconomic and temporal factors has been relatively limited. The studies in [12,15] acknowledged the importance of considering macroeconomic factors and other influential variables, such as interest rates, construction costs, and disposable income, which can significantly impact real estate prices. Addressing this gap by incorporating relevant macroeconomic and temporal features aligns with the research objective of developing comprehensive predictive models for the UK real estate market.…”

“…Comprehensive understanding of real estate price dynamics [5] Lack of research on interpretability of ML models in real estate pricing Challenges in understanding factors driving predictions [12] Need for investigating generalizability of findings across diverse real estate markets…”

“…Gradient Boosting Models (XGBoost, LightGBM, CatBoost): These advanced treebased ensemble methods were included due to their proven ability to achieve high predictive accuracy, effective handling of mixed data types, and automatic feature selection capabilities [12,32,33]. They have been successfully applied in various property valuation studies [10,32,33].…”