广东工业大学学报 ›› 2017, Vol. 34 ›› Issue (05): 29-33.doi: 10.12052/gdutxb.170073

• 综合研究 • 上一篇    下一篇

基于主成分分析与支持向量回归的精明增长建模与预测

蔡念, 李飞洋, 陈文杰, 陈伟建   

  1. 广东工业大学 信息工程学院, 广东 广州 510006
  • 收稿日期:2017-03-22 出版日期:2017-09-09 发布日期:2017-07-10
  • 通信作者: 李飞洋(1995-),男,硕士研究生,主要研究方向为机器学习.E-mail:651238355@qq.com E-mail:651238355@qq.com
  • 作者简介:蔡念(1976-),男,教授,博士,主要研究方向为机器视觉、机器学习和数字信号处理等.
  • 基金资助:
    广州市产学研协同创新重大专项项目（201508010001，201604016022，201604016064）

Smart Growth Modeling and Prediction Based on Principle Component Analysis and Support Vector Regression

Cai Nian, Li Fei-yang, Chen Wen-jie, Chen Wei-jian   

  1. School of Information Engineering, Guangdong University of Technology, Guangzhou 510006, China
  • Received:2017-03-22 Online:2017-09-09 Published:2017-07-10

HTML

PDF (PC)

2932

摘要： 随着城市化的迅速蔓延，如何使城市可持续化发展成为当前政府决策者的重要议题.为了有效地制定精明增长的策略，本文提出一种基于主成分分析的评价模型量化精明增长的程度；建立支持向量回归模型预测影响精明增长的各个指标的年际变化趋势，计算未来精明增长的预计得分；通过预计得分值选择最佳的精明增长计划方案.实验表明，该模型能准确地衡量精明增长的程度，并且能对未来的精明增长做出预测，从而为城市的合理健康发展提供决策指导.

关键词: 精明增长, 主成分分析, 支持向量回归

Abstract: With the urbanization extending at a high speed, the sustainable development of cities becomes a significant agenda for government policy makers. In order to effectively develop the strategy of smart growth, an evaluation model is proposed. First, principle component analysis (PCA) is applied to quantify the level of smart growth. Then, support vector regression (SVR) is employed to predict annual variation tendency of each indicator of smart growth. Finally, the total scores of smart growth are calculated for selecting an optimal solution to smart growth. The experiment results show that the proposed evaluation model can accurately measure the level of smart growth and predict the situation of smart growth in the future, which provides a comprehensive decision guidance for rational and healthy development of cities.

Key words: smart growth, principle component analysis, support vector regression

中图分类号: 

  • TP391.9
[1] Department of Economic and Social Affairs, United Nations. 2014 revision of the World Urbanization Prospects[R]. New York:United Nations, 2014:1-32. [2] 关静. 关于精明增长的研究述评[J]. 财经问题研究, 2013,(2):26-31. 2013,(2):26-31. [3] 王丹, 王士君. 美国"新城市主义"与"精明增长"发展观解读[J]. 国际城市规划, 2007, 22(2):61-66.WANG D, WANG S J. Understandings on development view of new urbanism and smart growth of the USA[J]. Urban Planning International, 2007, 22(2):61-66. [4] ANDERSON G. Why smart growth:A primer[M]. Washington DC:ICMA, 1998. [5] U. S. Environmental Protection Agency. Smart growth:A guide to developing and implementing greenhouse gas reductions programs[R]. Washington D C:Environmental Protection Agency, 2011:1-48. [6] BOEING G, CHURCH D, HUBBARD H, et al. LEED-ND and livability revisited[J]. Social Science Electronic Publishing, 2014, 27(1):31-55. [7] 程岚. 基于自然预期的美国房价动态研究[D]. 上海:复旦大学经济学院, 2012. [8] THOMAS V, WANG Y, FAN X. Measuring education inequality:Gini coefficients of education[J]. Social Science Electronic Publishing, 2001, 100(1):43-50. [9] NJEDL. Air quality index:a guide to air quality and your health[R].[S.l.]:United States EPA, 2003. [10] 何晓群. 多元统计分析[M]. 北京:中国人民大学出版社, 2008:164-105. [11] 叶宗裕. 关于多指标综合评价中指标正向化和无量纲化方法的选择[J]. 浙江统计, 2003,(4):24-25. 2003,(4):24-25. [12] SMOLA A J, LKOPF B. A tutorial on support vector regression[J]. Statistics and Computing, 2004, 14(3):199-222. [13] 李航. 统计学习方法[M]. 北京:清华大学出版社, 2012:108-109. [14] 陈龙, NEIL S, WILLIAMS A M. 美国内布拉斯加州林肯市犯罪行为的聚类及热点分布分析[J]. 测绘与空间地理信息, 2015(3):189-192.CHEN L, NEIL S, WILLIAMS A M. Cluster and hot spot analysis in Lincoln, Nebraska, USA[J]. Geometrics & Spatial Information Technology, 2015(3):189-192.
[1] 李晋芳, 韦光扬, 何汉武, 蔡嘉鸿, 陈基荣. 一种基于质点弹簧模型的牙龈软组织形变仿真算法[J]. 广东工业大学学报, 2020, 37(03): 49-54.
[2] 徐林浩, 钱斌, 胡蓉, 于乃康. 绿色车辆路径问题的改进拉格朗日松弛算法[J]. 广东工业大学学报, 2022, 39(05): 61-67.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 2017 《广东工业大学学报》编辑部
地址：广州市东风东路729号广东工业大学 邮编：510090
电话：020-37626653,020-37626139,020-37626396
邮箱：xbzrb@gdut.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发