| 27 | 0 | 50 |
| 下载次数 | 被引频次 | 阅读次数 |
数据样本不平衡与空间异质性成为制约大尺度城市住宅用地价格评估精度的关键瓶颈。本文建立“数据—模型”双模块协同评估框架,基于自适应高斯噪声过采样技术(SMOGN)定向合成稀缺价格区间样本缓解数据不平衡问题,采用基于差分进化(DE)算法的模型优选策略提升空间异质性下住宅用地价格评估精度,实现动态调节网络结构以适应大尺度区域“价格特征—区域属性”关系的建模需求,并应用SHAP方法增强模型的可解释性。基于浙江省11个地市4 951宗住宅用地交易数据的实证表明:SMOGN-DE-XGBoost模型比传统XGBoost模型具有更精细的非线性响应机制与更稳定的空间泛化能力,基准地价、公交站数量、最大容积率等是影响地价的主要因素,实证本文框架在处理大尺度复杂区域差异化特征时性能较好。
Abstract：The price of large-scale urban residential land exhibits significant regional disparities.As a result,data imbalance and spatial heterogeneity pose crucial bottlenecks to accurate land price assessment.To address this challenge,this study establishes a "data-model" dual-module collaborative assessment framework.The framework first adopts the synthetic minority over-sampling technique for regression with Gaussian noise(SMOGN) to directionally synthesize samples in scarce price ranges,thereby mitigating the issue of data imbalance.Simultaneously,to improve assessment accuracy under spatial heterogeneity,a model optimization strategy based on the differential evolution(DE) algorithm is employed.This strategy facilitates the dynamic adjustment of the model's network structure,tailoring it to the complex "price characteristic-regional attribute" relationships inherent to large-scale areas.Furthermore,the Shapley additive explanations(SHAP) method is applied to enhance interpretability.Empirical evidence from 4 951 residential land transactions across 11 cities in Zhejiang Province demonstrates the framework's effectiveness,with the SMOGN-DE-XGBoost model emerging as the top performer.Through distinct feature activation paths and assessment surfaces,it exhibits a more refined non-linear response mechanism and a more stable spatial generalization capability compared to the traditional XGBoost model.It accurately identifies factors such as benchmark land price,number of bus stations,and maximum floor area ratio as major influences on land price.This empirical study validates the framework's efficacy in handling the complex and heterogeneous characteristics of large-scale regions.
[1]付钦太.推动城市土地利用方式由粗放型向节约集约型转变[J].中州学刊,2015(7):32-36.
[2]马瑞衢.商品住宅价格影响因素研究综述[J].房地产导刊,2013(19):33-34.
[3]张鹏,胡守庚,杨剩富,等.基于多源数据和集成学习的城市住宅地价分布模拟:以武汉市为例[J].地理科学进展,2021,40(10):1664-1677.
[4]秦梦,祁新洲,宋玉茹,等.基于混频误差修正模型的土地价格预测[J].统计与决策,2022,38(3):178-183.
[5] Im S,Kim K,Lee G,et al.Development of a weighted average ensemble model for predicting officially assessed land prices using grid map data and SHAP[J].IEEE Access,2025,13:96251-96260.
[6]Yang Yuzhe,Zha Kaiwen,Chen Yingcong,et al.Delving into deep imbalanced regression[C]//Proceedings of the 38th International Conference on Machine Learning,New York:PMLR,2021:11842-11851.
[7]Branco P,Torgo L,Ribeiro R P.SMOGN:a pre-processing approach for imbalanced regression[C]//Proceedings of the First International Workshop on Learning with Imbalanced Domains:Theory and Applications,New York:PMLR,2017:36-50.
[8]Kocoglu Y,Gorell S B,Emadi H,et al.Enhancing shale gas EUR predictions with TPE optimized SMOGN:a comparative study of machine learning algorithms in the Marcellus shale with an imbalanced dataset[J].Gas Science and Engineering,2024,131:205475.DOI:10.1016/j.jgsce.2024.205475.
[9]Rad M,Rafiei A,Grunwell J,et al.Tackling the small imbalanced horizontal dataset regressions by stability selection and SMOGN:a case study of ventilation-free days prediction in the pediatric intensive care unit and the importance of PRISM[J].International Journal of Medical Informatics,2025,196:105809.DOI:10.1016/j.ij medinf.2025.105809.
[10]刘奇,梁智昊,訾建潇.SMOGN过采样下导水裂隙带高度的MPSO-BP预测模型[J].煤田地质与勘探,2024,52(11):72-85.
[11]Liu Xiaowei,Long Zhilin,Zhang Wei,et al.A multi-strategy hybrid machine learning model for predicting glass-formation ability of metallic glasses based on imbalanced datasets[J].Journal of Non-Crystalline Solids,2023,621:122645.DOI:10.1016/j.jnoncrysol.2023.122645.
[12]张祚,张跻耀,晏学丽,等.基于Orange集成多机器学习模型的城市周边农地流转价格评估研究:以武汉市为例[J].中国土地科学,2025,39(5):92-106.
[13]Ma Jie,Li Hong,Gao Weifeng,et al.An ensemble learning algorithm based on multimodal differential evolution and extreme learning machine[J].Memetic Computing,2025,17(2):21.DOI:10.1007/s12293-025-00461-7.
[14]Mayet A M,Gorelkina E I,Parayangat M,et al.Enhancing accuracy in X-ray radiation-based multiphase flow meters:integration of grey wolf optimization and MLP neural networks[J].Flow Measurement and Instrumentation,2024,100:102734.DOI:10.1016/j.flowmeasinst.2024.102734.
[15]Abualigah L.Particle swarm optimization:advances,applications,and experimental insights[J].Computers,Materials and Continua,2025,82(2):1539-1592.
[16]Blum C.Ant colony optimization:a bibliometric review[J].Physics of Life Reviews,2024,51:87-95.
[17]Mabdeh A N,Ajin R S,Razavi-termeh S V,et al.Enhancing the performance of machine learning and deep learning-based flood susceptibility models by integrating grey wolf optimizer(GWO)algorithm[J].Remote Sensing,2024,16(14):2595.DOI:10.3390/rs16142595.
[18]Deng Yufan,Tang Zhongan,Liu Baoj u,et al.Renovation and reconstruction of urban land use by a cost-heuristic genetic algorithm:a case in Shenzhen[J].ISPRS International Journal of GeoInformation,2024,13(7):250.DOI:10.3390/ij gi13070250.
[19]Houssein E H,Mahdy M A,Eldin M G,et al.Optimizing quantum cloning circuit parameters based on adaptive guided differential evolution algorithm[J].Journal of Advanced Research,2021,29:147-157.
[20]Meng Zhenyu,Yang Cheng.Two-stage differential evolution with novel parameter control[J].Information Sciences,2022,596:321-342.
[21]Bilal,Pant M,Zaheer H,et al.Differential evolution:a review of more than two decades of research[J].Engineering Applications of Artificial Intelligence,2020,90:103479.DOI:10.1016/j.engappai.2020.103479.
[22]陈菲,陈振杰,李飞雪,等.城市土地价格时空预测StackingGWR模型[J].地理与地理信息科学,2024,40(5):1-10.
[23]伏润得,杨振山.中国城市区域高质量发展时空分异及影响因素[J].地理学报,2024,79(4):819-836.
[24]卢宾宾,田小溪,秦思娴,等.基于地理加权建模技术的武汉市土地资源承载力评价研究[J].武汉大学学报(信息科学版),2025,50(3):430-438.
[25]俞艳,蒙文清,樊建,等.经济空间场与ESDA结合的公共卫生事件应急策略制定[J].武汉大学学报(信息科学版),2021,46(2):159-166.
[26]Fotheringham A S,Sachdeva M.Modelling spatial processes in quantitative human geography[J].Annals of GIS,2022,28(1):5-14.
[27]Wang Jinfeng,Zhang Tonglin,Fu Boj ie.A measure of spatial stratified heterogeneity[J].Ecological Indicators,2016,67:250-256.
[28]李加军.基于孤立森林的多离群点数据检测算法设计[J].现代电子技术,2024,47(5):139-142.
[29]周侯伯,肖桂荣,林炫歆,等.基于特征筛选与差分进化算法优化的滑坡危险性评估方法[J].地球信息科学学报,2022,24(12):2373-2388.
基本信息:
中图分类号:P208;F299.23
引用信息:
[1]俞艳,黄致祥,张宸赫,等.数据不平衡与空间异质联合优化的大尺度城市住宅用地价格评估[J].地理与地理信息科学().
基金信息:
国家自然科学基金项目(42471445)
2026-09-17
2026-09-17
2026-09-17