摘要： 某机载雷达数据库中包含大量连续属性和分类属性，且数据库规模庞大。为此，给出一种改进的模糊关联规则挖掘算法。该算法引入有向无环图和字节向量用以提高频繁项目集的计算效率，解决挖掘时磁盘操作频繁的问题，并定义新的模糊度量提高正规则的识别概率。实验结果表明，该算法比传统算法具有更高的执行效率和准确率。

关键词: 大型机载雷达数据库, QL-算子, 字节向量结构, 模糊度量, 数据挖掘

Abstract: One airborne radar database with a large scale contains massive continuous attributes and categorical attributes. An improved algorithm of fuzzy association rule mining is presented. It improves the computational efficiency of frequent itemsets by introducing the DAG and the byte-vector structure for reducing the I/O overhead generated during the database mining. Meanwhile, the algorithm defines new fuzzy measure to enhance the recognition probability of the positive rules. Experimental results show that the algorithm has a better performance than the traditional algorithm.

Key words: large airborne radar database, QL-operator, byte-vector structure, fuzzy measure, data mining

中图分类号: 

• TP311