CAJ | 학술논문

分析了Apriori算法存在之不足，在此基础上提出了一种基于分区思想的IABP算法．该算法首先将待挖掘的数据集分成若干块，然后分别对各块进行挖掘．在挖掘过程中，只存储满足最小支持度的频繁项，并删除不满足最小支持度的非频繁项．测试结果表明，该挖掘方法降低了挖掘算法的时闻复杂度，提高了挖掘算法的效率．
분석료Apriori산법존재지불족，재차기출상제출료일충기우분구사상적IABP산법．해산법수선장대알굴적수거집분성약간괴，연후분별대각괴진행알굴．재알굴과정중，지존저만족최소지지도적빈번항，병산제불만족최소지지도적비빈번항．측시결과표명，해알굴방법강저료알굴산법적시문복잡도，제고료알굴산법적효솔．
Basing on the disadvantages of the Apriori algorithm being analyzed, an improved apriori algorithm based on partition is proposed in this paper. In this algorithm, the dataset which is waiting for mining is first parted into some partitions, and then each parttition is mined by using parallel operation. During the procedure of mining, only the frequently items which is satisfied the condition of minimal-support are stored in the memory, and the non-frequently items which don' t satisfy the condition of minimal-support will be deleted at the same time. The test result shows that this algorithm can decrease the time complexity of mining and increase the mining efficiency.