摘要
为了明确城市公共交通可达性对房价影响,基于步行可达性、公交可达性和地铁可达性3种城市公共交通指标,应用多种机器学习算法构建房价预测模型,根据2021年西安市公共交通路网数据,探究其交通可达性对房价的影响。首先,采集西安市主城区的房屋属性数据、城市交通数据以及城市空间数据并进行数据预处理;其次,基于空间句法理论分别建立步行可达性、公交可达性和地铁可达性3种交通可达性指标;最后,以构建的特征指标代入随机森林(RF)、梯度提升迭代决策树(GBDT)、轻量梯度提升机(LGBM)以及特征价格模型(HPM)这4种机器学习算法建立房价预测模型,确定出最优房价预测模型并探究城市公共交通可达性对房价的影响。并以西安市地铁三号线为例应用缓冲区分析和RF算法,分析运营前后地铁可达性对房价的影响。结果表明:在房价预测模型搭建方面,RF算法的房价预测精度为89.2%,均方根误差为1766.89,该算法满足实时性要求,符合研究预期,且优于其他模型;在房价影响因素分析方面,基于空间句法计算的交通可达性特征在模型中的重要性百分比为23.8%,说明交通因素对房价具有重要影响。因此,城市公共交通的发展在一定程度上能够提高区域经济活力,加快房地产经济发展,为促进经济社会协调和可持续发展起到了重要的作用。
In order to clarify the impact of urban public transportation accessibility on housing price,three urban public transportation indexes of walking accessibility,bus accessibility and metro accessibility were established,as well as a real estate price prediction model was built by using several machine learning algorithms.The influence of transport accessibility on housing price was explored according to the public transportation network data of Xi’an in 2021.First,the house property data,urban traffic data and urban spatial data of Xi’an were collected and preprocessed.Second,the spatial syntax theory method was applied to establish three indexes of walking accessibility,bus accessibility and metro accessibility.Last,the constructed data was took as the characteristic indexs,the real estate price prediction model of Xi’an was constructed by using RF,LGBM,GBDT and HPM algorithm.Xi’an Metro Line 3 was taken as an example,buffer analysis and RF algorithm were used to analyze the influence of subway accessibility on housing price before and after operation.The results show that the prediction accuracy of housing price by RF algorithm is 89.2%and the root mean square error is 1766.89,which proves that real-time requirements and research expectations can be achieved by using RF algorithms,and it is superior to other models.The importance percentage of public transportation accessibility based on space syntax calculation is 23.8%,which indicates that the urban public transportation accessibility has an important impact on housing prices.Therefore,the improvement of regional economic vitality and the development of real estate economy depend on the development of urban public transportation to some extent,and it plays an important role in promoting economic and social coordination and sustainable development.2 tabs,11 figs,25 refs.
作者
刘青青
薛超
巨永锋
冯红霞
LIU Qing-qing;XUE Chao;JU Yong-feng;FENG Hong-xia(School of Civil Engineering,Chang’an University,Xi'an 710061,Shaanxi,China;School of Electronics and Control Engineering,Chang’an University,Xi'an 710064,Shaanxi,China;College of Architecture,Xi’an University of Architecture and Technology,Xi'an 710055,Shaanxi,China)
出处
《长安大学学报(自然科学版)》
CAS
CSCD
北大核心
2022年第4期87-97,共11页
Journal of Chang’an University(Natural Science Edition)
基金
国家自然科学基金项目(61603057)
陕西省自然科学基础研究计划项目(2015GY033)
陕西省教育厅科研计划项目(20JT036)。
关键词
交通工程
城市公共交通
交通可达性
机器学习算法
房价预测
西安
traffic engineering
urban public transportation
traffic accessibility
machine learning algorithm
housing price prediction
Xi’an