建筑与文化Issue(2)：226-228,3.DOI:10.19875/j.cnki.jzywh.2025.02.069
利用可解释性机器学习提取影响杭州市二手住宅价格的因子
Using Interpretable Machine Learning to Extract Factors Affecting the Price of Second-Hand Houses in Hangzhou
曹献之 1陆激2
作者信息
- 1. 浙江大学建筑工程学院||浙江大学平衡建筑研究中心
- 2. 浙江大学平衡建筑研究中心||浙江大学建筑设计研究院有限公司
- 折叠
摘要
Abstract
Geographic artificial intelligence has achieved rapid development in recent years,but its interpretability has been difficult due to the nature of machine learning Black Box Models.This study examines the determinants of second-hand residential property prices in central Hangzhou.Utilizing the XGBoost Model,we regress property prices on the accessibility of surrounding facilities and intrinsic property characteristics.To elucidate the significance of each variable,the SHAP Model was employed,thereby quantifying the impact of each factor on property prices.The analysis revealed that proximity to the nearest metro station and the age of the property had the most significant effect on the price of second-hand houses.Additionally,properties located in the vicinity of West Lake and along the east bank of the Qiantang River demonstrated substantial locational advantages,reflected in higher property prices.关键词
可解释性机器学习/XGBoost/SHAP/二手住宅价格/杭州Key words
interpretable machine learning/XGBoost/SHAP/second-hand residential prices/Hangzhou引用本文复制引用
曹献之,陆激..利用可解释性机器学习提取影响杭州市二手住宅价格的因子[J].建筑与文化,2025,(2):226-228,3.
