Abstract:The traditional data mining algorithm for heterogeneous network is based on the correlation between data to make clustering. When a large number of redundant data occur, the correlation between data is weakened and it makes the accuracy of data mining decrease. To solve this problem, a new mining algorithm based on weakly correlation redundant environment is proposed in this paper. Firstly, in this algorithm, the original cluster center of the big data set is determined through the data clustering method, and the cluster center is updated to ensure that it is close to the real center, so as to realize the data clustering of big data set. Then, the weak association rule of big data set is mined to calculate the association between the data in the weak association rule. Finally, the weak association rule is used to mine the data in the weak association redundant environment. The experimental results show that the proposed mining algorithm has higher mining efficiency and accuracy, as well as lower complexity.