Research and application of improved K-means based on MapReduce

Abstract

With the development of big data, the traditional data mining clustering algorithm K-Means is inefficient and has poor scalability in dealing with massive data. MapReduce on the Hadoop platform was used to realize the parallel processing of the K-Means algorithm, the performance of the algorithm was tested by experiments. The results show that the improved K-Means algorithm has good parallel expansion capability, high efficiency, and great potential when processing big data mining. The algorithm is applied to the big data processing of customer consumption in a restaurant chain, and the effectiveness of the algorithm is verified, which can better serve the decision of restaurant.