Abstract: Focusing on the continuous optimization and improvement problems of the autonomous adaptive decision-making capability of the software-defined C4ISR system, a decision tree-based system autonomous adaptive structure is proposed to realize the evolution and optimization of the autonomous adaptive decision-making capability with the continuous growth of the decision tree, on the basis of which a decision tree growth method based on the quantitative association rule mining is proposed to transform the decision tree growth problem into an association rule mining analysis problem. By transforming the decision tree growth problem into an association rule min-ing analysis problem, and adopting the data min-ing algorithm combining the K-means method and the Apriori method to realize the mining and growth of association rules of the system adaptive decision tree under the big data environment, and finally the proposed association rule mining method not only realizes the growth of the decision tree, but also reduces the interference of invalid information compared with the traditional decision tree construction method by the veification of simulation experiments.