Abstract: Aiming to presented algorithms doesn’t fast extract spatial topology association.This paper proposes a double mining algorithm of spatial topology association，which is suitable for mining spatial topology association rules in large data.The algorithm uses binary system to express topology element，and uses bit arithmetic to generate candidate frequent itemsets by two methods of top-down and bottom-up，in order to double search spatial topology association rules.The algorithm also uses character of digit spatial transaction to reduce the number of scanned transaction when computing support of itemsets，mining efficiency is also improved by the way.The result of experiment indicates that the algorithm is faster and more efficient than presented algorithms when mining spatial topology association rules in spatial data.

Key words: spatial data mining, topology association rules, bit arithmetic, topology element, double mining

摘要： 针对现有挖掘算法不能快速地提取空间拓扑关联，提出一种空间拓扑关联的双向挖掘算法，其适合在海量空间数据中挖掘空间拓扑关联规则；该算法用二进制数表示空间拓扑元，并用其位运算，按自顶向下和自底向上两种方式计算产生候选频繁项，实现双向搜索空间拓扑关联规则；算法在计算支持数时还用数字式空间事务的特性减少被扫描的事务数，达到提高挖掘效率的目的。实验结果表明在空间数据中挖掘空间拓扑关联规则时，该算法比现有算法更快速更有效。

关键词: 空间数据挖掘, 拓扑关联规则, 位运算, 拓扑元, 双向挖掘