House Price Prediction with Optimistic Machine Learning Methods Using Bayesian Optimization

Haolan Jiang

2024

Abstract

Recently, housing has been a fundamental necessity for human survival. However, the challenge lies in the often-inflated housing prices, particularly in high-GDP cities. House price prediction is crucial for citizens as it aids in effective financial planning and contributes to social stability. This study delves into the factors influencing house prices, employing and evaluating four regression models: multiple linear regression, regression decision tree, Random Forest, and XGBoost. The focus is on optimizing the performance of the two most promising models. The study finds that there is no substantial positive or negative association between the prediction label, which is the average house price of a region, and any individual attribute in the dataset. Through model comparisons, it is observed that the decision tree model outperforms the regression model significantly, with the integrated models, specifically Random Forest and XGBoost, outshining the regular regression tree model. In 5-fold cross-validation, the Bayesian optimized XGBoost model yields the best results in this study. The post-optimization R2 value of XGBoost is 0.846, showcasing an improvement of 0.024 compared to the pre-optimization phase. The hybrid model introduced in this study holds significant research potential in the realm of house price prediction. Additionally, it provides valuable insights for individuals, enabling them to make well-informed financial plans, particularly in terms of home purchase decisions. This, in turn, contributes to addressing potential social issues and fostering greater social harmony and stability.

Paper Citation

in Harvard Style

Jiang H. (2024). House Price Prediction with Optimistic Machine Learning Methods Using Bayesian Optimization. In Proceedings of the 1st International Conference on Data Science and Engineering - Volume 1: ICDSE; ISBN 978-989-758-690-3, SciTePress, pages 488-496. DOI: 10.5220/0012825400004547

in Bibtex Style

@conference{icdse24,
author={Haolan Jiang},
title={House Price Prediction with Optimistic Machine Learning Methods Using Bayesian Optimization},
booktitle={Proceedings of the 1st International Conference on Data Science and Engineering - Volume 1: ICDSE},
year={2024},
pages={488-496},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012825400004547},
isbn={978-989-758-690-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Data Science and Engineering - Volume 1: ICDSE
TI - House Price Prediction with Optimistic Machine Learning Methods Using Bayesian Optimization
SN - 978-989-758-690-3
AU - Jiang H.
PY - 2024
SP - 488
EP - 496
DO - 10.5220/0012825400004547
PB - SciTePress