Prediction of network public opinion features in urban planning based on geographical case-based reasoning

作     者：Rui Li Jingqi Wang Shunli Wang Huayi Wu 

作者机构：State Key Laboratory of Information Engineering in SurveyingMapping and Remote SensingWuhan UniversityWuhanPeople’s Republic of China Hubei Luojia LaboratoryWuhanPeople’s Republic of China Collaborative Innovation Center of Geospatial TechnologyWuhanPeople’s Republic of China 

出 版 物：《International Journal of Digital Earth》 (国际数字地球学报（英文）)

年 卷 期：2022年第15卷第1期

页      面：890-910页

核心收录：

学科分类：0303[法学-社会学] 08[工学] 0813[工学-建筑学] 0814[工学-土木工程] 0833[工学-城乡规划学] 0823[工学-交通运输工程] 

基　　金：supported by the National Natural Science Foundation of China [grant number U20A2091 41930107] 

主　　题：Geographic case-based reasoning urban planning case similarity weight prediction of public opinion features 

摘      要：As a significant part of sustainable urban development proposed by the United Nations,urban planning is related to the ecological environment and transportation,especially affecting quality of life and social well-being. In the process of urban planning,the public express their opinions on open network platforms,resulting in large quantities of network public opinion data,which has important implications for evaluating urban planning. Based on the idea of geographical case-based reasoning (CBR),this paper constructs an expression framework for urban planning cases in the form of a ‘case problem–case attribute–case result’ triad. On this basis,this paper proposes a similarity calculation method of urban planning cases that integrates case attribute. Finally,based on an improvement to the traditional k-nearest neighbors method,the proposed public opinion feature calculation model considers similarity weights,which allow us to predict network public opinion features,including viewpoint-level emotional tendency and concerned groups of urban planning cases. The experimental result shows similarity weights (SWs) model could effectively improve the prediction accuracy,where the average MIC-F1 score reached more than 74%. Based on CBR,the proposed method can predict the development trends of public opinion in future planning cases,and provide scientific and reasonable decision support for urban planning.