Research on the synergy law of Pinellia tuber and Golden thread by improved association rule algorithm

Abstract

In this paper, the association relationship between data is becoming more and more complex. The Apriori algorithm, needs to scan the database several times to calculate the candidate set, which is inefficient. By designing the experiment, the use of professor Lei Zhang from 2006 to 2017, 11038 cases of the basis data set, to put forward four kinds of improved algorithm of association rules: FP-growth algorithm, Apriori_ParallelCount algorithm, GAN algorithm, Apriori_BOMS algorithm do the performance testing. Experimental results show that the Apriori_ParallelCount algorithm has the best comprehensive performance among the five association rule algorithms. Apriori_ParallelCount algorithm is uses the hash stored to nature, the data collection of transaction and the mapping of frequent item sets is stored in a hash structure, to avoid the algorithm when calculating the candidate set support multiple scan data set, proposed multi-threaded operation, improve the efficiency of the algorithm. Using the most efficient algorithm to contain 1625 cases of basis of Pinellia tuber and Golden thread data association rules analysis, it is concluded that professor Zhang Lei the synergy law of Pinellia tuber- Golden thread, using the Python language for visualization mapping of all analytical results.