本体辅助的约束空间关联规则挖掘方法

Mining Method of Ontology Accessorial Constrained Geospatial Association Rule

提出利用本体辅助进行基于约束的空间关联规则挖掘,引入本体所表达的层次结构和概念属性合理地解释用户定义的约束,包括对实验项的修剪约束和根据层次体系得到的层次约束。通过优化挖掘的实验数据提高挖掘的效率和优化挖掘结果,为解决数据巨大而知识不足的困境提供了可供借鉴的方法。以河南省的统计年鉴和交通数据为基础数据,分析和挖掘经济与交通之间的空间关联关系,有效地验证了方法的性能。

A new way of how the ontology theory being used in assisting constrained geospatial association rules mining was proposed. Constraints defined by user include two types, prune constraints and delaminate （ab- stract） constraints, which are interpreted logically by using hiberarchy about the domain and property of con ception represented by ontology. With the optimization of experiment dataset, the mining efficiency is enhanced and the mined results is optimized, which provide a viable means to figure out the problem of having huge data but lacking useful knowledge. The statistic yearbook and traffic data of Henan province was taken as real data and the performance was validated by analyzing and mining the association rules between economy and traffic.