基于SVM的IDSS研究与应用设计
|
摘要
智能决策支持系统(IDSS)是决策支持系统和人工智能技术相结合的产物,有效地解决了系统中定量与定性相结合以及半结构化、非结构化的问题,扩大了决策支持系统的应用范围,提高了系统求解问题的能力。随着计算机技术以及人工智能技术的不断发展,智能决策支持系统在许多领域得到研究和应用,并发挥着越来越重要的作用。
本文以辅助房地产企业决策人员进行科学决策为出发点,在对房地产投资决策的理论和程序进行深入分析的基础上,以数据仓库、数据挖掘、联机分析处理和支持向量机(SVM)等先进智能决策技术为手段,设计了房地产投资智能决策支持系统(REIIDSS)的原型系统。该系统对减少盲目性带来的风险、提高决策效率等具有重要的理论意义和现实价值。
首先,借助智能决策支持系统的基本理论和方法,提出了房地产投资IDSS的设计思想,设计了体系结构,并对数据仓库、模型库系统和知识库系统进行了具体分析和设计。
其次,针对关系投资成败的关键步骤——房地产投资选址决策问题,采用支持向量机建立了相应的模型。该模型可以降低决策者的自身经验和主观认知对选址决策的影响,增加选址决策的客观性和科学性。
最后,根据房地产投资项目的特点,结合支持向量机和多层模糊综合评价法,建立了房地产投资方案优选模型。该模型将具有模糊性质的经验、意见、观点等定性指标加以解析化和定量化,使方案建立在科学基础上,从而保证了结论的科学性,同时又利用支持向量机的先处理提高了决策分析的效率。
Intelligent Decision Support System (IDSS) is a combination of Decision Support Systems and Artificial Intelligence. IDSS can effectively solve the problem of qualitative analysis combined with quantitative analysis, and those semi-structural and non-structural problems. Its application realms are extended, and its ability of solving problems is improved. Along with the unceasing development of the computer technology as well as the artificial intelligence technology, the intelligence decision support system will be researched and applied further, and will play a more and more vital role in many realms.
Focused on assisting the decision-makers of real estate enterprise to make scientific decision, based on the In-depth analysis of real estate investment decision-making theory and procedures, and by making use of the advanced techniques of Data Warehouse, Data Mining, On-Line Analytical Processing and Support Vector Machine (SVM), this thesis designed a prototype system of Real Estate Investment Intelligent Decision Support System (REIIDSS). The system has important theory meaning and real-life value for reducing the risks of blindness and increasing the efficiency of decision.
Firstly, by means of the basic theories and methods of IDSS, the collectivity structure of REIIDSS were designed in this thesis. And the data warehouse, the model subsystem and the knowledge-based subsystem were specifically analyzed and designed.
Secondly, in allusion to the key steps of the real estate investment location decision-making, Support Vector Machine is used to establish the corresponding model, which can reduce the impacts from the experience and subjective awareness of decision-makers and increase the accuracy of location decisions.
Finally, according to the characteristics of the real estate investment projects, SVM and multi-level fuzzy synthetic judgment are utilized to build up a Real Estate Investment Optimization Model. The model analyzes and quantifies some qualitative and fuzzy features, such as experience, opinion and viewpoint. It causes the scheme being established in the scientific foundation to ensure the science of the conclusion. At the same time, it is favor of SVM to handle those quantified features, and increases the efficiency of decision analysis.
本文以辅助房地产企业决策人员进行科学决策为出发点,在对房地产投资决策的理论和程序进行深入分析的基础上,以数据仓库、数据挖掘、联机分析处理和支持向量机(SVM)等先进智能决策技术为手段,设计了房地产投资智能决策支持系统(REIIDSS)的原型系统。该系统对减少盲目性带来的风险、提高决策效率等具有重要的理论意义和现实价值。
首先,借助智能决策支持系统的基本理论和方法,提出了房地产投资IDSS的设计思想,设计了体系结构,并对数据仓库、模型库系统和知识库系统进行了具体分析和设计。
其次,针对关系投资成败的关键步骤——房地产投资选址决策问题,采用支持向量机建立了相应的模型。该模型可以降低决策者的自身经验和主观认知对选址决策的影响,增加选址决策的客观性和科学性。
最后,根据房地产投资项目的特点,结合支持向量机和多层模糊综合评价法,建立了房地产投资方案优选模型。该模型将具有模糊性质的经验、意见、观点等定性指标加以解析化和定量化,使方案建立在科学基础上,从而保证了结论的科学性,同时又利用支持向量机的先处理提高了决策分析的效率。
Intelligent Decision Support System (IDSS) is a combination of Decision Support Systems and Artificial Intelligence. IDSS can effectively solve the problem of qualitative analysis combined with quantitative analysis, and those semi-structural and non-structural problems. Its application realms are extended, and its ability of solving problems is improved. Along with the unceasing development of the computer technology as well as the artificial intelligence technology, the intelligence decision support system will be researched and applied further, and will play a more and more vital role in many realms.
Focused on assisting the decision-makers of real estate enterprise to make scientific decision, based on the In-depth analysis of real estate investment decision-making theory and procedures, and by making use of the advanced techniques of Data Warehouse, Data Mining, On-Line Analytical Processing and Support Vector Machine (SVM), this thesis designed a prototype system of Real Estate Investment Intelligent Decision Support System (REIIDSS). The system has important theory meaning and real-life value for reducing the risks of blindness and increasing the efficiency of decision.
Firstly, by means of the basic theories and methods of IDSS, the collectivity structure of REIIDSS were designed in this thesis. And the data warehouse, the model subsystem and the knowledge-based subsystem were specifically analyzed and designed.
Secondly, in allusion to the key steps of the real estate investment location decision-making, Support Vector Machine is used to establish the corresponding model, which can reduce the impacts from the experience and subjective awareness of decision-makers and increase the accuracy of location decisions.
Finally, according to the characteristics of the real estate investment projects, SVM and multi-level fuzzy synthetic judgment are utilized to build up a Real Estate Investment Optimization Model. The model analyzes and quantifies some qualitative and fuzzy features, such as experience, opinion and viewpoint. It causes the scheme being established in the scientific foundation to ensure the science of the conclusion. At the same time, it is favor of SVM to handle those quantified features, and increases the efficiency of decision analysis.
引文
[1]高洪深.决策支持系统(第二版)[M].北京:清华大学出版社,2005.24-28.
[2]黄梯云.智能决策支持系统[M].北京:电子工业出版社,2001.33-35.
[3]牛振喜,蔡康英.基于回归算法与多层次分布式智能决策支持系统的设计[J].计算机工程与设计,2007,28(16):4004-4006。
[4]杨善林,倪志伟.机器学习与智能决策支持系统[M].北京:科学出版社,2004.34-41.
[5]牟东玮.基于数据仓库技术的决策支持系统研究设计[J].科技信息(学术研究),2007,(12):73-74.
[6]毛海军,唐焕文.智能决策支持系统(IDSS)研究进展[J].小型微型计算机系统,2003,24(05):874-878.
[7]郑颖华,武根友.智能决策支持系统中的模型库及其管理系统[J].科学技术与工程,2006,6(09):1312-1315.
[8]刘军,叶钒.基于数据仓库和数据挖掘的应用研究[J].福建电脑,2007,(03):59-60.
[9]陈京民.数据仓库原理、设计与应用[M].北京:中国水利水电出版社,2004.8-10.
[10]Mehmed Kantardzic.闪四清,陈茵,程雁等译.数据挖掘--概念、模型、方法和算法[M].北京:清华大学出版社,2003.48-50.
[11]Jiawei Han,Micheline Kamber.Data Mining Concepts and Techniques[M].Morgan Kaufmann Publishers,2000.132-137.
[12]张晓伟,杜龙非.XML与Web数据挖掘技术[J].商场现代化,2007,(23):27-28.
[13]V.Vapnik著.张学工译.统计学习理论的本质[M].北京:清华大学出版社,2000.64-67.
[14]CHIH CHUNG CHANG,CHIH JEN LIN.Training V-support Vector Classifiers:Theory and Algorithms[J].Neural Computation,2001,13(09):2119-2147.
[15]Arenas Garcia J,Perez Cruz F.Multi-class Support Vector Machines:A New Approach[C].Proc of the IEEE Int Conf on acoustics,Speech and Signal Processing.New York,2003.81-784.
[16]HSU C W,LIN C J.A simple decomposition method for support vector machines[J].Machine Learning,2002.46(01):291-314.
[17]欧贵文,肖国华.基于支持向量机的文本无关字迹鉴别系统[J].小型微型计算机系统,2004,25(10):1885-1888.
[18]Tefas A.,Kotropoulos C.,Pitas I.Using Support Vector Machine to Enhance the Performance of Elastic Graph Matching for Frontal Face Authentication[J].IEEE TRANSACTION ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE.2001.23(07):735-746.
[19]郑华,谢莉,张粒子,等.电力市场下系统边际价格混合预测模型的新研究[J].中国电机工程学报,2005,25(17):66-71.
[20]边肇祺等.模式识别(第二版)[M].北京:清华大学出版社,2000.67-69.
[21]van Gestel,T,Suykens,J.A.K,Baesens,B.,Stijn,V,Vanthienen,J,Dedene,G,Bart,D.M,Vandewalle,J.Benchmarking least squares support vector machine classifiers[J].Machine Learning,2004,54(01):5-32.
[22]史忠植.知识发现[M].北京:清华大学出版社,2002.203-207.
[23]Amari S,Wu S.Improving support vector machine classifier by modifying kernel function[J].Neural Networks,1999.12(09):783-789.
[24]肖健华,吴今培,樊可清,等.基于支持向量机的齿轮故障诊断方法[J].中国制造业信息化,2005,25(17):66-71.
[25]陈琳,潘蜀健.房地产项目投资[M].北京:中国建筑工业出版社,2004.78-89.
[26]骈永富,阎俊爱.房地产投资分析与决策[M].北京:中国物价出版社,2003.79-84.
[27]CHUNFULIN,SHENG DE WANG.Fuzzy Support Vector Machines[J].IEEE Transactions on Neural Networks,2002,13(02):464-471.
[28]Chih-Wei Hsu,Chih-Jen Lin.A Comparison of Methods for Multi-class Support Vector Machines[D].Taiwan.Taiwan University,2001.
[29]Kressel U.Pairwise classification and support vector machines[M].Cambridge,Massachusetts:MIT Press,1999.255-268.
[30]J.Weston,C.Watkins.Multi-class Support Vector Machines[J].Technical Report.,2001.(05):212-250.
[31]Dietterich T G,Bakiri G.Solving multiclass learning problems via error-correcting output codes[J].Journal of Artificial Intelligence Research,2003,19(02):263-286.
[32]忻栋.支持向量机算法的研究及在说话人识别上的应用[D]:[硕士学位论文].杭州.浙江大学,2002.
[33]Ancona N,Cicirelli G,Distante A.Complexity reduction and parameter selection in support vector machines[C].In:Proceedings of the 2002 International Joint Conference on Neural Networks.Honolulu,2002.2375-2380.
[34]Jordaan E M,Smits G F.Estimation of the regularization parameter for support vector regression[C].In:Proceedings of the 2002 International Joint Conference on Neural Networks.Honolulu,2002.2192-2197.
[35]Ito K,Nakano R.Optimizing support vector regression hyperparameters based on cross-validation[C].In:Proceedings of the International Joint Conference on Neural Networks.Portland,2003.2077-2082.
[36]全勇.支持向量机算法的若干问题研究[D]:[博士学位论文].上海.上海交通大学,2003.
[37]YANG M H,A HUJA N.A Geometric Approach to Train Support Vector Machines[M].Proceedings of CVPR2000.Hilton Head Island.2000.430-437.
[38]李鸿吉.模糊数学基础及实用算法[M].北京:科学出版社,2005.1-35.
[39]汪培庄.模糊系统理论与模糊计算机[M].北京:科学出版社,2001.49-56.
[40]范如国,房地产投资与管理[M].武汉:武汉大学出版社,2003.153-155.
[2]黄梯云.智能决策支持系统[M].北京:电子工业出版社,2001.33-35.
[3]牛振喜,蔡康英.基于回归算法与多层次分布式智能决策支持系统的设计[J].计算机工程与设计,2007,28(16):4004-4006。
[4]杨善林,倪志伟.机器学习与智能决策支持系统[M].北京:科学出版社,2004.34-41.
[5]牟东玮.基于数据仓库技术的决策支持系统研究设计[J].科技信息(学术研究),2007,(12):73-74.
[6]毛海军,唐焕文.智能决策支持系统(IDSS)研究进展[J].小型微型计算机系统,2003,24(05):874-878.
[7]郑颖华,武根友.智能决策支持系统中的模型库及其管理系统[J].科学技术与工程,2006,6(09):1312-1315.
[8]刘军,叶钒.基于数据仓库和数据挖掘的应用研究[J].福建电脑,2007,(03):59-60.
[9]陈京民.数据仓库原理、设计与应用[M].北京:中国水利水电出版社,2004.8-10.
[10]Mehmed Kantardzic.闪四清,陈茵,程雁等译.数据挖掘--概念、模型、方法和算法[M].北京:清华大学出版社,2003.48-50.
[11]Jiawei Han,Micheline Kamber.Data Mining Concepts and Techniques[M].Morgan Kaufmann Publishers,2000.132-137.
[12]张晓伟,杜龙非.XML与Web数据挖掘技术[J].商场现代化,2007,(23):27-28.
[13]V.Vapnik著.张学工译.统计学习理论的本质[M].北京:清华大学出版社,2000.64-67.
[14]CHIH CHUNG CHANG,CHIH JEN LIN.Training V-support Vector Classifiers:Theory and Algorithms[J].Neural Computation,2001,13(09):2119-2147.
[15]Arenas Garcia J,Perez Cruz F.Multi-class Support Vector Machines:A New Approach[C].Proc of the IEEE Int Conf on acoustics,Speech and Signal Processing.New York,2003.81-784.
[16]HSU C W,LIN C J.A simple decomposition method for support vector machines[J].Machine Learning,2002.46(01):291-314.
[17]欧贵文,肖国华.基于支持向量机的文本无关字迹鉴别系统[J].小型微型计算机系统,2004,25(10):1885-1888.
[18]Tefas A.,Kotropoulos C.,Pitas I.Using Support Vector Machine to Enhance the Performance of Elastic Graph Matching for Frontal Face Authentication[J].IEEE TRANSACTION ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE.2001.23(07):735-746.
[19]郑华,谢莉,张粒子,等.电力市场下系统边际价格混合预测模型的新研究[J].中国电机工程学报,2005,25(17):66-71.
[20]边肇祺等.模式识别(第二版)[M].北京:清华大学出版社,2000.67-69.
[21]van Gestel,T,Suykens,J.A.K,Baesens,B.,Stijn,V,Vanthienen,J,Dedene,G,Bart,D.M,Vandewalle,J.Benchmarking least squares support vector machine classifiers[J].Machine Learning,2004,54(01):5-32.
[22]史忠植.知识发现[M].北京:清华大学出版社,2002.203-207.
[23]Amari S,Wu S.Improving support vector machine classifier by modifying kernel function[J].Neural Networks,1999.12(09):783-789.
[24]肖健华,吴今培,樊可清,等.基于支持向量机的齿轮故障诊断方法[J].中国制造业信息化,2005,25(17):66-71.
[25]陈琳,潘蜀健.房地产项目投资[M].北京:中国建筑工业出版社,2004.78-89.
[26]骈永富,阎俊爱.房地产投资分析与决策[M].北京:中国物价出版社,2003.79-84.
[27]CHUNFULIN,SHENG DE WANG.Fuzzy Support Vector Machines[J].IEEE Transactions on Neural Networks,2002,13(02):464-471.
[28]Chih-Wei Hsu,Chih-Jen Lin.A Comparison of Methods for Multi-class Support Vector Machines[D].Taiwan.Taiwan University,2001.
[29]Kressel U.Pairwise classification and support vector machines[M].Cambridge,Massachusetts:MIT Press,1999.255-268.
[30]J.Weston,C.Watkins.Multi-class Support Vector Machines[J].Technical Report.,2001.(05):212-250.
[31]Dietterich T G,Bakiri G.Solving multiclass learning problems via error-correcting output codes[J].Journal of Artificial Intelligence Research,2003,19(02):263-286.
[32]忻栋.支持向量机算法的研究及在说话人识别上的应用[D]:[硕士学位论文].杭州.浙江大学,2002.
[33]Ancona N,Cicirelli G,Distante A.Complexity reduction and parameter selection in support vector machines[C].In:Proceedings of the 2002 International Joint Conference on Neural Networks.Honolulu,2002.2375-2380.
[34]Jordaan E M,Smits G F.Estimation of the regularization parameter for support vector regression[C].In:Proceedings of the 2002 International Joint Conference on Neural Networks.Honolulu,2002.2192-2197.
[35]Ito K,Nakano R.Optimizing support vector regression hyperparameters based on cross-validation[C].In:Proceedings of the International Joint Conference on Neural Networks.Portland,2003.2077-2082.
[36]全勇.支持向量机算法的若干问题研究[D]:[博士学位论文].上海.上海交通大学,2003.
[37]YANG M H,A HUJA N.A Geometric Approach to Train Support Vector Machines[M].Proceedings of CVPR2000.Hilton Head Island.2000.430-437.
[38]李鸿吉.模糊数学基础及实用算法[M].北京:科学出版社,2005.1-35.
[39]汪培庄.模糊系统理论与模糊计算机[M].北京:科学出版社,2001.49-56.
[40]范如国,房地产投资与管理[M].武汉:武汉大学出版社,2003.153-155.