基于空间数据挖掘的超市选址决策研究
摘要
面对激烈的市场竞争和日益复杂的社会经济环境,科学的设施选址决策往往会影响一个商业实体的成功与否,商店的地理位置及其商业性质直接影响着其经营业绩。对于准备投资的业主来说,通过选址分析可以确定最具投资潜力的店址。数据挖掘(data mining)技术的不断向前发展,特别是空间数据挖掘(spatial data mining)技术被广泛的研究与应用,使之在解决空间知识的获取问题上独具优势,这为选址技术与理论提供了一种可行的新思路。
空间数据模型是空间数据挖掘的核心和基础,因此对空间数据模型的研究有着重要的研究意义。本文针对常用空间数据模型在空间数据挖掘中具有较强应用指向性,通用性不强的问题,建立了一种时空特征与相互关系一体化的时空数据模型。形式化定义并详细给出了该模型的元素组成、信息范畴及元素间的相互关系。并在此基础上给出了时空特征对象的特征操作、空间操作、时间操作及三者互操作。
此外,以超市选址具体应用为例,利用时空数据模型并结合空间数据挖掘技术,对青岛市超市选址决策进行了研究,并在研究成果的基础上开发出超市选址决策支持系统,该系统克服了目前选址技术缺乏科学合理的依据以及在面临数量庞大并且可能异构异种的空间数据源时,无法发掘出隐藏在这些数据当中的潜在而有用的知识的问题。通过对该系统进行具体设计和实现,验证了该系统在一定程度上可以对决策者提供决策支持。
Faced with the fierce market competition and an increasingly complex socio-economic environment, scientific site decisions often affects the success of a business entity, the location of store and the commercial nature of its directly impact on the operating results. For the proprietors who prepare to invest, location analysis can identify the most investment potential site. The continued development of data mining technology, especially the wide research and application of spatial data mining technology has been of space has unique advantages to resolve the issue of getting spatial knowledge which provides a feasible new idea for site technology and theory.
Spatial data model is the core and foundation of spatial data mining. For this reason, the research on spatial data model has important significance. In order to solve the problem that common spatial-temporal data models usually point to specific application and cannot be universal, this article uses a spatial-temporal data model which has integrative attributive character and spatial-temporal relation. The elements form, information category and seven interrelations between elements of this model are formalized defined. The attribute operation, spatial operation and temporal operation of spatial-temporal objects and operations among them are also showed in this article.
Furthermore, with a specific application example of supermarket site, this article uses spatial-temporal data model and combine spatial data mining techniques to research on the issue of decision-making site in Qingdao City supermarkets. On the basis of research results, a decision support system for supermarket site is built. The system overcome the problems that some current site technology are lack of scientific reasonable basis and can't discover the hidden potential useful knowledge when facing to a huge number of heterogeneous spatial data sources. Through the design and implementation, the system is proved that it can provide decision makers with supports to a certain extent.
空间数据模型是空间数据挖掘的核心和基础,因此对空间数据模型的研究有着重要的研究意义。本文针对常用空间数据模型在空间数据挖掘中具有较强应用指向性,通用性不强的问题,建立了一种时空特征与相互关系一体化的时空数据模型。形式化定义并详细给出了该模型的元素组成、信息范畴及元素间的相互关系。并在此基础上给出了时空特征对象的特征操作、空间操作、时间操作及三者互操作。
此外,以超市选址具体应用为例,利用时空数据模型并结合空间数据挖掘技术,对青岛市超市选址决策进行了研究,并在研究成果的基础上开发出超市选址决策支持系统,该系统克服了目前选址技术缺乏科学合理的依据以及在面临数量庞大并且可能异构异种的空间数据源时,无法发掘出隐藏在这些数据当中的潜在而有用的知识的问题。通过对该系统进行具体设计和实现,验证了该系统在一定程度上可以对决策者提供决策支持。
Faced with the fierce market competition and an increasingly complex socio-economic environment, scientific site decisions often affects the success of a business entity, the location of store and the commercial nature of its directly impact on the operating results. For the proprietors who prepare to invest, location analysis can identify the most investment potential site. The continued development of data mining technology, especially the wide research and application of spatial data mining technology has been of space has unique advantages to resolve the issue of getting spatial knowledge which provides a feasible new idea for site technology and theory.
Spatial data model is the core and foundation of spatial data mining. For this reason, the research on spatial data model has important significance. In order to solve the problem that common spatial-temporal data models usually point to specific application and cannot be universal, this article uses a spatial-temporal data model which has integrative attributive character and spatial-temporal relation. The elements form, information category and seven interrelations between elements of this model are formalized defined. The attribute operation, spatial operation and temporal operation of spatial-temporal objects and operations among them are also showed in this article.
Furthermore, with a specific application example of supermarket site, this article uses spatial-temporal data model and combine spatial data mining techniques to research on the issue of decision-making site in Qingdao City supermarkets. On the basis of research results, a decision support system for supermarket site is built. The system overcome the problems that some current site technology are lack of scientific reasonable basis and can't discover the hidden potential useful knowledge when facing to a huge number of heterogeneous spatial data sources. Through the design and implementation, the system is proved that it can provide decision makers with supports to a certain extent.
引文
[1]Jiawei,Han.M.Kamber,Data Mining:Concepts and Techniques.Los Altos,CA:Morgan Kaufmann Publishers,2001
[2]邵峰品,于忠清.数据挖掘--原理与算法.中国水利水电出版社.2003
[3]Jiawei Han.CLARANS:a method for clustering objects for spatial data mining,IEEE Transactions on Knowledge and Data Engineering,Sept.-Oct.2002(5):lO03-1016.
[4]李德仁,王树良,史文中等.论空间数据挖掘和知识发现.武汉大学学报,信息科学版,2001,26(6):49卜499
[5]宫鹏.一种快速高分辨率遥感影像分割算法.计算机应用研究.2006.10:154一155
[6]马建文,李利伟,温奇.容差粗糙集卫星遥感数据分类.计算机工程与应用.2007.20:11-13
[7]Ester M.,Kriegel H.P.,Sander J.:Algorithms and Applications for Spatial Data Mining,in:Geographic Data Mining and Knowledge Discovery,Research Monographs in GIS,Taylor and Francis,2001,pp.160-187.
[8]欧阳骞,徐宏炳,于立.移动环境下基于矢量栅格的数字地图查询研究与设计.计算机应用与软件.2007.24(1):92-94
[9]王敏,吕强.基于模型移植的高效数据引擎构造方法.计算机工程与应用.2006.42(35):110-112
[10]刘兵,严和平,段江娇,汪卫,施伯乐.度量空间一种自底向上索引树构造算法.计算机研究与发展.2006.43(9):1651-1657
[11]毛先成,彭华熔.空间数据全关系化存储的研究.测绘信息与工程.2005.30(3):8-9
[12]李杰,徐勇,王云峰,王友.最简关联规则及其挖掘算法.计算机工程.2007.33(13):46-48
[13]吉根林,韦素云.分布式环境下约束性关联规则的快速挖掘.小型微型计算机系统.2007.28(5):882-885
[14]孙华梅,郭茂祖,焦杰,黄梯云.一种新的面向属性归纳中概念层次技术研究.管理科学学报.2004.7(1):65-72
[15]郭达志.空间数据挖掘及其不确定性研究.测绘与空间地理信息.2004.27(5):6-10
[16]朱玉全,孙志挥.基于频繁模式树的关联规则增量式更新算法.计算机学 报.2003.26(1):91-96
[17]刘明亮,唐先明,庄大方,刘纪远.基于融合技术的结构性栅格时空数据平台的构建.地球信息科学.2003.5(4):63-68
[18]张锦.多分辨率空间数据模型理论与实现技术研究.测绘学报.2003.32(1):93-95
[19]陈涛,张玮.一个改进的并行关联规则算法研究.计算机技术与发展.2007.17(1):139-141
[20]甄彤,范饱峰.基于Agent的分布式空间数据挖掘模型及实现.2004.31(10):96-97
[21]李宏,蔡之华.空间数据挖掘查询语言GMQL的设计与使刚.计算机应用.2002.22(11):91-93
[22]樊明辉,涂建东,池天河,陈崇成.基于Web的空间关联规则提取及可视化的实现.计算机应用研究.2006.23(2):157一159
[23]C.D.Klose,S.Loew,R.Giese,G.Borm:Spatial predictions of geological rock mass properties based on in-situ interpretations of multi-dimensional seismic data.Engineering Geology,Volume 93,Issues 3-4,30 August 2007,Pages 99-116
[24]杨风召.高维数据挖掘中若干关键问题的研究[博士学位论文].上海:复旦大学,2003.
[25]P.Compieta,S.DiMartino,M.Bertolotto,F.Ferrucci,T.Kechadi:Exploratory spatio-temporal data mining and visualization..Journal of Visual Languages & Computing,Volume 18,Issue 3,June 2007,Pages 255-279
[26]李德仁,王树良,李德毅,王新洲.论空间数据挖掘知识发现的理论与方法.武汉大学学报信息科学版,2002.06,26(6)
[27]邸凯昌.空间数据挖掘和知识发现的理论与方法.武汉测绘科技大学博士论文,1999.
[28]Josep Sardanyes,Ricard V.Sole:Spatio-temporal dynamics in simple asymmetric hypercycles under weak parasitic coupling.Physica D:Nonlinear Phenomena,Volume 231,Issue 2,15 July 2007,Pages 116-129
[29]Koperski K,Adhihary J & Han J.1996b.Spatial Data Mining:Progress and Challenges.SIGMOD'96 Workshop on Research Issues on Data Mining and Knowledge Discovery(DMKD'96),Montreal,Canada,June,1996
[30]J.Han,M.Kamber,and A.K.H.Tung,Spatial Clustering Methods in Data Mining:A Survey[A],2001
[31]Junko Iwahashi,Richard J.Pike:Automated classifications of topography from DEMs by an unsupervised nested-means algorithm and a three-part geometric signature.Geomorphology,Volume 86,1ssues 3-4,1 May 2007,Pages 409-440
[32]陈军,赵仁亮.GIS空间关系的基本问题与研究进展.测绘学报.1999.28(2):95-102
[33]郭薇,陈军.三位拓扑空间关系的形成化描述及拓扑空间关系最小集.武汉水利电力大学学报.1996.29(4):85-90
[34]周海燕,王家耀,吴升.空间数据挖掘技术及其应用.测绘通报.2002第二期
[35]李德仁,王树良,李德毅,王新洲.论空间数据挖掘知识发现的理论与方法.武汉大学学报信息科学版.2002.06,26(6)
[36]邸凯吕,李德仁,李德毅.空间数据发掘和知识发现的框架.武汉测绘科技大学学报.1997,22(4):328-332
[37]Armstrong,M P.Temporality in Spatial databases[A].San_Antonio,Proceeding:GIS/LIS 88[C],1988,(2):880-889.
[38]Langran G,Chrisman,N.R.A framework for temporal'geographic information[J].Cartographica,1988,25(3):11-14.
[39]张英,邵峰晶,孙仁诚.面向对象的四元组时空数据模型.四川大学学报(工程科学版).2007.05:115-118
[40]T.Oxley,H.M.ApSimon:Space,time and nesting Integrated Assessment Models.Environmental Modelling & Software,Volume 22,Issue 12,December 2007,Pages 1732-1749
[41]Volker Gaede,Oliver Guther.Multidimensional Access Methods[J].Computing Surveys,1998,30(2):170-231.
[42]Orhan Aktas,Sonia Waiczies,Alina Smorodchenko,Jan Dorr,Bibiane,etc:Treatment of Relapsing Paralysis in Experimental Encephalomyelitis bX Targeting Thl Cells through Atorvastatin.The Journal of Experimental Medicine,Volume 197,2003,Number 6,725-733
[43]Seeger B.,Kriegel H.-P.The Buddy Tree:An Efficient and Robust Access Method for Spatial Databases[C].In Proc.Of 16th Int.Conf,On Very Large Data Bases.Brisbane,Australia,1990:590-601.
[44]Six H.-W.,Widmayer P.Spatial Access Structures for Geometric Databases,Data Structures and Efficient Algorithms.In Proc.Of Lecture Notes in Computer Science.1992(594):214-232.
[45]M.Ester,H.Kriegel,And X.Xu.Knowledge Discovery in Large Spatial Databases:Focusing Techniques for Efficient Class Identification.In Proc.Of 46 'Int'LSymp,On Large Spatial Databases(Ssd'95).Portland,1995(8):67-82.
[46]普雷帕拉塔等著,庄心谷译.计算几何导论[M].北京:科学出版社,1990
[47]周培德.计算几何-算法分析与设计[M].北京:清华大学出版社,1999
[48]Kwon Ho Lee,Young Joon Kim,Wolfgang von Hoyningen-Huene,John P.Burrow:Spatio-temporal variability of satellite-derived aerosol optical thickness over Northeast Asia in 2004.Atmospheric Environment,Volume 41,Issue 19,June 2007,Pages 3959-3973
[49JEster Martin,Bans-Peter Kriegel,el.al.A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise.Proc.Of the and International Conference on Knowledge Discovery and Data Mining.Portland,Oregon,August 1996.
[50]Agrawal Rakesh,Johannes Gehrke,Dimitrios Gunopulos,el.al.Automatic Subspace Clustering of High Dimensional Data for Data Mining Applications.Proc.of the 1998ACM-SIGMOD International Conference on Management of Data,USA,1998.
[51]Goyal R.K.,Similarity Assessment for Cardinal Directions Between Extended Spatial Objects,PHD Thesis,The University of Maine,2000.
[52]Martin Ester,Hans-Peter Kriege,Matthias Schubert:Accurate and Efficient Crawling for Relevant Websites.Proceedings 2004 VLDB Conference,2004,Pages 396-407
[53]jennifer L.Gardy,Cory Spencer,Ke Wang,Martin Ester,etc:improving protein subcellular localization prediction for Gram-negative bacteria.Nucleic Acids Research,2003,Vol.31,No.13%13-3617
[54]Langran G.A review of temporal database research and its use in GIS applications[J].International journal of Geographical Information Systems,1989,3(3):215-232.
[55]Zlatanoca S,Rahman A,Shi W Z.Topological models and frameworks for 3D spatial objects[J].Computers and Geosciences,2004(30):419-428.
[56]J.Allen.Maintaining knowledge about temporal intervals[J].Communications of the ACM,1983,26(11):823-843.
[57]陈颖.基于GIS的选址空间决策支持系统[J]中国水运(学术版).2007(04).
[58]黄玲,柳宗伟.选址空间决策支持系统的研究与实践[J].山东师范大学学报(自然科学版),2004(03)
[59]陈姚,朱韬,李石华.基于GIS的零售商业网点选址模型研究[J].云南地理环境研究,2005(06)
[2]邵峰品,于忠清.数据挖掘--原理与算法.中国水利水电出版社.2003
[3]Jiawei Han.CLARANS:a method for clustering objects for spatial data mining,IEEE Transactions on Knowledge and Data Engineering,Sept.-Oct.2002(5):lO03-1016.
[4]李德仁,王树良,史文中等.论空间数据挖掘和知识发现.武汉大学学报,信息科学版,2001,26(6):49卜499
[5]宫鹏.一种快速高分辨率遥感影像分割算法.计算机应用研究.2006.10:154一155
[6]马建文,李利伟,温奇.容差粗糙集卫星遥感数据分类.计算机工程与应用.2007.20:11-13
[7]Ester M.,Kriegel H.P.,Sander J.:Algorithms and Applications for Spatial Data Mining,in:Geographic Data Mining and Knowledge Discovery,Research Monographs in GIS,Taylor and Francis,2001,pp.160-187.
[8]欧阳骞,徐宏炳,于立.移动环境下基于矢量栅格的数字地图查询研究与设计.计算机应用与软件.2007.24(1):92-94
[9]王敏,吕强.基于模型移植的高效数据引擎构造方法.计算机工程与应用.2006.42(35):110-112
[10]刘兵,严和平,段江娇,汪卫,施伯乐.度量空间一种自底向上索引树构造算法.计算机研究与发展.2006.43(9):1651-1657
[11]毛先成,彭华熔.空间数据全关系化存储的研究.测绘信息与工程.2005.30(3):8-9
[12]李杰,徐勇,王云峰,王友.最简关联规则及其挖掘算法.计算机工程.2007.33(13):46-48
[13]吉根林,韦素云.分布式环境下约束性关联规则的快速挖掘.小型微型计算机系统.2007.28(5):882-885
[14]孙华梅,郭茂祖,焦杰,黄梯云.一种新的面向属性归纳中概念层次技术研究.管理科学学报.2004.7(1):65-72
[15]郭达志.空间数据挖掘及其不确定性研究.测绘与空间地理信息.2004.27(5):6-10
[16]朱玉全,孙志挥.基于频繁模式树的关联规则增量式更新算法.计算机学 报.2003.26(1):91-96
[17]刘明亮,唐先明,庄大方,刘纪远.基于融合技术的结构性栅格时空数据平台的构建.地球信息科学.2003.5(4):63-68
[18]张锦.多分辨率空间数据模型理论与实现技术研究.测绘学报.2003.32(1):93-95
[19]陈涛,张玮.一个改进的并行关联规则算法研究.计算机技术与发展.2007.17(1):139-141
[20]甄彤,范饱峰.基于Agent的分布式空间数据挖掘模型及实现.2004.31(10):96-97
[21]李宏,蔡之华.空间数据挖掘查询语言GMQL的设计与使刚.计算机应用.2002.22(11):91-93
[22]樊明辉,涂建东,池天河,陈崇成.基于Web的空间关联规则提取及可视化的实现.计算机应用研究.2006.23(2):157一159
[23]C.D.Klose,S.Loew,R.Giese,G.Borm:Spatial predictions of geological rock mass properties based on in-situ interpretations of multi-dimensional seismic data.Engineering Geology,Volume 93,Issues 3-4,30 August 2007,Pages 99-116
[24]杨风召.高维数据挖掘中若干关键问题的研究[博士学位论文].上海:复旦大学,2003.
[25]P.Compieta,S.DiMartino,M.Bertolotto,F.Ferrucci,T.Kechadi:Exploratory spatio-temporal data mining and visualization..Journal of Visual Languages & Computing,Volume 18,Issue 3,June 2007,Pages 255-279
[26]李德仁,王树良,李德毅,王新洲.论空间数据挖掘知识发现的理论与方法.武汉大学学报信息科学版,2002.06,26(6)
[27]邸凯昌.空间数据挖掘和知识发现的理论与方法.武汉测绘科技大学博士论文,1999.
[28]Josep Sardanyes,Ricard V.Sole:Spatio-temporal dynamics in simple asymmetric hypercycles under weak parasitic coupling.Physica D:Nonlinear Phenomena,Volume 231,Issue 2,15 July 2007,Pages 116-129
[29]Koperski K,Adhihary J & Han J.1996b.Spatial Data Mining:Progress and Challenges.SIGMOD'96 Workshop on Research Issues on Data Mining and Knowledge Discovery(DMKD'96),Montreal,Canada,June,1996
[30]J.Han,M.Kamber,and A.K.H.Tung,Spatial Clustering Methods in Data Mining:A Survey[A],2001
[31]Junko Iwahashi,Richard J.Pike:Automated classifications of topography from DEMs by an unsupervised nested-means algorithm and a three-part geometric signature.Geomorphology,Volume 86,1ssues 3-4,1 May 2007,Pages 409-440
[32]陈军,赵仁亮.GIS空间关系的基本问题与研究进展.测绘学报.1999.28(2):95-102
[33]郭薇,陈军.三位拓扑空间关系的形成化描述及拓扑空间关系最小集.武汉水利电力大学学报.1996.29(4):85-90
[34]周海燕,王家耀,吴升.空间数据挖掘技术及其应用.测绘通报.2002第二期
[35]李德仁,王树良,李德毅,王新洲.论空间数据挖掘知识发现的理论与方法.武汉大学学报信息科学版.2002.06,26(6)
[36]邸凯吕,李德仁,李德毅.空间数据发掘和知识发现的框架.武汉测绘科技大学学报.1997,22(4):328-332
[37]Armstrong,M P.Temporality in Spatial databases[A].San_Antonio,Proceeding:GIS/LIS 88[C],1988,(2):880-889.
[38]Langran G,Chrisman,N.R.A framework for temporal'geographic information[J].Cartographica,1988,25(3):11-14.
[39]张英,邵峰晶,孙仁诚.面向对象的四元组时空数据模型.四川大学学报(工程科学版).2007.05:115-118
[40]T.Oxley,H.M.ApSimon:Space,time and nesting Integrated Assessment Models.Environmental Modelling & Software,Volume 22,Issue 12,December 2007,Pages 1732-1749
[41]Volker Gaede,Oliver Guther.Multidimensional Access Methods[J].Computing Surveys,1998,30(2):170-231.
[42]Orhan Aktas,Sonia Waiczies,Alina Smorodchenko,Jan Dorr,Bibiane,etc:Treatment of Relapsing Paralysis in Experimental Encephalomyelitis bX Targeting Thl Cells through Atorvastatin.The Journal of Experimental Medicine,Volume 197,2003,Number 6,725-733
[43]Seeger B.,Kriegel H.-P.The Buddy Tree:An Efficient and Robust Access Method for Spatial Databases[C].In Proc.Of 16th Int.Conf,On Very Large Data Bases.Brisbane,Australia,1990:590-601.
[44]Six H.-W.,Widmayer P.Spatial Access Structures for Geometric Databases,Data Structures and Efficient Algorithms.In Proc.Of Lecture Notes in Computer Science.1992(594):214-232.
[45]M.Ester,H.Kriegel,And X.Xu.Knowledge Discovery in Large Spatial Databases:Focusing Techniques for Efficient Class Identification.In Proc.Of 46 'Int'LSymp,On Large Spatial Databases(Ssd'95).Portland,1995(8):67-82.
[46]普雷帕拉塔等著,庄心谷译.计算几何导论[M].北京:科学出版社,1990
[47]周培德.计算几何-算法分析与设计[M].北京:清华大学出版社,1999
[48]Kwon Ho Lee,Young Joon Kim,Wolfgang von Hoyningen-Huene,John P.Burrow:Spatio-temporal variability of satellite-derived aerosol optical thickness over Northeast Asia in 2004.Atmospheric Environment,Volume 41,Issue 19,June 2007,Pages 3959-3973
[49JEster Martin,Bans-Peter Kriegel,el.al.A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise.Proc.Of the and International Conference on Knowledge Discovery and Data Mining.Portland,Oregon,August 1996.
[50]Agrawal Rakesh,Johannes Gehrke,Dimitrios Gunopulos,el.al.Automatic Subspace Clustering of High Dimensional Data for Data Mining Applications.Proc.of the 1998ACM-SIGMOD International Conference on Management of Data,USA,1998.
[51]Goyal R.K.,Similarity Assessment for Cardinal Directions Between Extended Spatial Objects,PHD Thesis,The University of Maine,2000.
[52]Martin Ester,Hans-Peter Kriege,Matthias Schubert:Accurate and Efficient Crawling for Relevant Websites.Proceedings 2004 VLDB Conference,2004,Pages 396-407
[53]jennifer L.Gardy,Cory Spencer,Ke Wang,Martin Ester,etc:improving protein subcellular localization prediction for Gram-negative bacteria.Nucleic Acids Research,2003,Vol.31,No.13%13-3617
[54]Langran G.A review of temporal database research and its use in GIS applications[J].International journal of Geographical Information Systems,1989,3(3):215-232.
[55]Zlatanoca S,Rahman A,Shi W Z.Topological models and frameworks for 3D spatial objects[J].Computers and Geosciences,2004(30):419-428.
[56]J.Allen.Maintaining knowledge about temporal intervals[J].Communications of the ACM,1983,26(11):823-843.
[57]陈颖.基于GIS的选址空间决策支持系统[J]中国水运(学术版).2007(04).
[58]黄玲,柳宗伟.选址空间决策支持系统的研究与实践[J].山东师范大学学报(自然科学版),2004(03)
[59]陈姚,朱韬,李石华.基于GIS的零售商业网点选址模型研究[J].云南地理环境研究,2005(06)