Abstract: Aiming to the presented mining algorithms based on spatial transaction inefficiently extracted spatial topology association，an algorithm of spatial topology association rules mining is proposed，which is suitable for searching topology association in spatial association transverse mining.The algorithm turns spatial topology association into integer，and uses the way of value descending to form candidate frequent itemsets.The algorithm uses number operation to gain support of candidate，and uses number attribute to reduce the number of scanned spatial transaction.The algorithm is used to mine spatial topology association rules，and this experiment indicates that the algorithm is fast and efficient.

Key words: topology association, transverse mining, spatial association rules, value descending, number attribute

摘要： 针对现有基于空间事务的挖掘算法不能有效地提取空间拓扑关联，提出一种空间拓扑关联规则挖掘算法，其适合在空间关联横向挖掘中搜索拓扑关联；该算法将空间拓扑关系转换为整数，用整数递减的方法构建候选频繁项；算法用数字运算获取支持数，同时还用数字特性减少被扫描的空间事务数。在挖掘空间拓扑关联规则时，实验证明它是快速而有效的。

关键词: 拓扑关联, 横向挖掘, 空间关联规则, 数值递减, 数字特性