基于改进关联规则算法的中药药对药味间性味归经功效属性关系的发现研究

Discovery of Association Rules between TCM Properties in Crude Drug Pairs by Mining between Datasets and Probability Test

中药药对是联系方剂和中药的重要环节。虽然传统中医理论对药对中药味的组成方式有多角度的论述,但相比论著丰富的方剂理论还不够系统。本文分别采用标准关联规则发现Apriori算法以及改进多数据库计算方法,对从历代药对文献中收集整理得到的625个药对347味药中包括性味、归经、功效等共49个属性形成的数据库进行挖掘研究,并对两种方法得到的结果进行比较。结果表明,本文提出的新方法更适用于分析药物间的关联规则,集中度更好,更易于发现有价值的关联规则。

The TCM Crude Drug Pair is an important link node between TCM drugs and formulae. In the traditional TCM theory, there are many descriptions in multi-aspects. However, they seem still scattered compared with formulae theories. In this work, the association rule mining method was applied to reveal the associations between the structures of the crude drug pairs. The standard and enhanced Apriori algorithms were applied and run on the crude drug pair database including 625 drug pairs, 347 drugs and 49 properties. The results from the two algorithms were compared and analyzed. The new enhanced Apriori algorithm was found to be more suitable for the association rule discovery between drugs in the drug pair or formulae.