Aiming at the characteristics of "more frequent and more ordinary" of traditional Chinese medicine materials, this paper proposes an association rule algorithm based on double support degree and compression matrix to improve the execution efficiency of drug search. Specifically, by introducing the principle of compression matrix, adding arrays to improve the matrix storage mode, and compressing the matrix according to the correlation theorem of the association rule, so as to reduce the number of scans of the matrix and the number of candidate sets generated. By setting the maximum and minimum support threshold, the generation of infrequent itemsets and overfrequent itemsets is limited to ensure the validity of the rules, thereby improving the practicability of the obtained rules. Experiments show that the proposed algorithm has obvious advantages in the data set with high support degree and high number of transactions.