中医医案方剂具有多维度的特性，单独采用一种相似度算法难以展现方剂数据在各维度的相似情况，具有一定的局限性。本研究基于中医医案方剂的多维数据特点，结合中医临床遣方用药的思维过程，设计了中医医案方剂的多维相似度算法，并以七则国医大师医案为示例，详细展示了该算法的计算过程，同时将多维相似度算法计算结果与传统算法计算结果进行了对比分析。通过对比可以看出本算法能够相对全面地反映不同方剂间的相似度，且结果更为合理也更贴合中医临床实际。

Prescriptions of traditional Chinese medicine (TCM) case records are multidimensional in nature, and it is difficult to show the similarity of prescription data in each dimension by using one similarity algorithm alone, which has certain limitations. Based on the multidimensional data of prescriptions of TCM case records, this paper designs a multidimensional similarity algorithm for TCM prescriptions based on the thinking process of TCM clinical prescriptions and uses 7 case records of national TCM masters as examples to demonstrate the calculation process of the algorithm in detail. The results of the multidimensional similarity algorithm are compared with those of the traditional algorithm. The comparison shows that the algorithm can reflect the similarity between different prescriptions in a relatively comprehensive way, and the results are more reasonable and more suitable for the clinical practice of TCM.

R249