Deep mining method for high-dimensional big data based on association rule

摘要

Existing high-dimensional deep data mining methods have the problems of low precision and high energy consumption. Therefore, a deep mining method of high-dimensional big data based on association rules is proposed. Ealat algorithm is used to change the format of high-dimensional large data set. On this basis, MapReduce computing model is introduced to divide parallel tasks into map and reduce phases to realise the construction of operation platform. Hadoop's distributed file system is used to store distributed data. The input and output of the algorithm are converted into the form required by the MapReduce computing model to realise the deep mining of high-dimensional big data. Experimental results show that this method has higher mining accuracy and lower energy consumption. The result of practical application is good.