一种新的正负空间同位规则挖掘算法.pdfVIP

2008 年 1 月第1 期 小型微型计算机系统 Journal of Chinese Computer Systems Vol.29 No.l 2008 一种新的正负空间同位规则挖掘算法 胡彩平，秦小麟 (南京航空航天大学信息科学与技术学院，江苏南京 210016) E咀ail: hucaiping@nuaa. edu. cn 摘 要:空间同位规则是一种新的空间数据挖掘方法.尽管人们对挖掘空间同位规则做丁一些研究，但大多数研究者仅仅对正 交闷闷位规则进行研究，没有考虑负空间向你规则.本义提出丁一种新的成负空间同位规则挖掘算法(Positive and Negative Spatial Co-Location Rules Mining Algorithm ， PNSCLRMA). 为丁减少计算量，界法利用丁旦形邻城去减少迄接运界和定义兴 趣皮去删除不感兴怨的空间同位模式两项优化技术.实验农明，该算法有效可行. 关键词:空间数据挖掘;空间间位模式;ïE负空间问UL规则z 兴趣度 中图分类号: TP30 1. 6 文献标识码:A 文章蝙号:1000-1220(2008)01-0080份05 A Novel Positive and Negative Spatial Co-location Rules Mining Algorithm HUCai喘ping ， QIN Xiao-lin (College o/In/ormati拥， Science αnd Technology. Nanjing University 0/ Aeronautics . Astronautics.Nanjing 210016.China) Abstract: Spatial co-location rules mining is a new method in spatial data mining. Although there have some researching on spa- tial co叩location rules mining ,but mostly researchers discuss only positive spatial co伽location rules ,dont consider negative spa- tial co-location rules. 1n this paper. a novel positive and negative spatial co-location rules mining algorithmCPNSCLRMA) was proposed. 丁o reduce the computational cost , the algorithm uses two optimization techniques of adopting star neighborhoods to reduce join operations and defining the interesting degree to delete uninteresting spatial co斗ocation patterns. Experimental re- sults show that the algorithm is feasible and efficient. Key words:spatÎal data mining;spatial co-location patternslpositive and negative spatial co-location rules;interesting degree 问题、生态问题等. 1 引宫 空间同位规则与空间关联规则既有联系又有区别[12] ，联 信息化的发展使得更多的空间数据被使用，因此，挟取空