期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2016
卷号：87
期号：3
出版社：Journal of Theoretical and Applied
摘要：With the increase in spatial data analysis, the co-location patterns and its dependencies are used to discover the complex patterns on spatial databases. Most of the traditional spatial data mining techniques have been implemented based on the assumption that the data is meaningful and clean. It is essential to study the data integration issues along with spatial co-locating patterns. Generally, spatial co-location mining algorithms are used to discover the spatial objects and its dependencies among them. As the data size increases, the co-location objects and its patterns are difficult to process on complex spatial objects. In this paper, an optimized spatial co-locating pattern mining framework was developed to discover the highly ranked correlated patterns using the hadoop framework. This MapReduce model was used to minimize computational time and space on complex spatial databases. Finally, the experimental results on the complex spatial data are evaluated using the proposed framework and the traditional hadoop based pattern mining models.
关键词：Spatial Dataset; Co-Location Models; Association Rules; Prevalence Threshold; Mapreduce
