CAJ | 학술논문

为进一步提高关联规则挖掘的运行效率，在传统 Apriori 算法的基础上，提出了一种基于分辨矩阵和 Apriori 算法的关联规则挖掘算法。在数据预处理阶段，采用分辨矩阵对原始数据集进行属性约简，达到降维的目的；在关联规则挖掘阶段，采用位图来表示原始数据集，并在每一步运算过程中通过置信度阈值对特征集进行约简。应用实例和仿真实验表明，该方法在时间复杂度和空间复杂度都有了极大的改善，具有一定的应用价值。
위진일보제고관련규칙알굴적운행효솔，재전통 Apriori 산법적기출상，제출료일충기우분변구진화 Apriori 산법적관련규칙알굴산법。재수거예처리계단，채용분변구진대원시수거집진행속성약간，체도강유적목적；재관련규칙알굴계단，채용위도래표시원시수거집，병재매일보운산과정중통과치신도역치대특정집진행약간。응용실례화방진실험표명，해방법재시간복잡도화공간복잡도도유료겁대적개선，구유일정적응용개치。
An algorithm for mining association rules based on discernibility matrix and Apriori algorithm is proposed to further im-prove the efficiency of association rules mining.The discernibility matrix is applied to reduce the attribute of original data set.It represents the original data set as bitmap.It is to feature set reduction based on confidence threshold.The application examples and simulation results show that the method can improve the complexity of time and space.