An Algorithm of Association Rule Based on Cloud Computing

Large-scale data processing is one of the focal points of research in information technology. The traditional algorithm of association rule is in a large overhead, due to the frequent itemsets being computed on the dataset. The rapid development of distributed technology makes cloud computing a reality in the implementation of data processing algorithms. To improve the traditional association rule algorithm, in this paper, an AprioriMR algorithm for mining association rule based on cloud computing is proposed. The AprioriMR algorithm takes HDFS to store data and is well adapted to the Hadoop's Map-Reduce computing model. It divides into two parts, deals with Map-Reduce operation, and combines to produce the frequent patterns. The AprioriMR algorithm inherits the Map-Reduce scalability to huge datasets and to thousands of processing nodes. Experimental results show that it is very efficiently compared with the traditional association rules algorithm and has a good speedup when deals with massive data.