Improved Algorithms Research for Association Rule Based on Matrix
XianWen Luo, Weiqing Wang · 2010
In association rules, although Apriori algorithm uses cut-technology when it generates item sets of candidates, it has to scan the entire database while scanning the transaction database each time. The scanning speed is very slow for its large amount of data. The improved Apriori algorithm based on matrix is improved from the Apriori algorithm and the matrix algorithm. Its basic idea is transforming the event database into matrix database so as to get the matrix item set of maximum item set. When finding the frequent k-item set from the frequent k-item set, only its matrix set is found. So only the corresponding data are calculated to get frequent k item set. Therefore the improved Apriori algorithm's computing time is very fast. Simulation data are used in experiments to compare the speeds of the improved Apriori algorithm and the Apriori algorithm. The results of the experiments prove the efficiency of improved Apriori algorithm.