南京大学学报（自然科学版）2018，Vol.54Issue(2)：436-451,16.DOI:10.13232/j.cnki.jnju.2018.02.016
从动态空间数据库中挖掘共生关系和竞争关系
Mining the symbiotic and competitive relationships from dynamic spatial databases
摘要
Abstract
Spatial co-location pattern mining is an important direction in spatial data mining.Spatial co-location patterns represent the subsets of spatial features which are frequently located together in a geographic space.Spatial co-location patterns also show the co-located relationship among features.However,the features have further relationships such as symbiotic relationship and competitive relationship,besides co-located relationship.This paper mines symbiotic relationship and competitive relationship from the dynamic spatial databases.The objectives are strong symbiotic patterns and competitive pairs,respectively.The data between two adjacent time slots will change, and the changed data will incur changed neighbor relationships.The changed neighbor relationships reflect the symbiotic/competitive relationships among the features.According to this idea,the definitions of strong symbiotic patterns and competitive pairs are given,and the novel methods and pruning strategies for mining strong symbiotic patterns and competitive pairs are proposed.The experiments on synthetic databases evaluate the efficiency and scalability of the algorithms.Statistical comparison and partial results show of prevalent co-location patterns,strong symbiotic patterns and competitive pairs are conducted on real databases.关键词
空间数据挖掘/空间co-location模式/共生关系/竞争关系Key words
spatial data mining/spatial co-location patterns/symbiotic relationships/competitive relationships分类
信息技术与安全科学引用本文复制引用
芦俊丽,王丽珍,赵家松,肖清..从动态空间数据库中挖掘共生关系和竞争关系[J].南京大学学报（自然科学版）,2018,54(2):436-451,16.基金项目
国家自然科学基金(61472346,61662086),云南省自然科学基金(2015FB149,2016FA026) （61472346,61662086）
