Classification association rules mining for Zhongjing prescriptions
Yang Xuemei, Lin Duan-yi, Changen Zhou, Xinmei Lai · 2008
Zhongjing prescriptions have experienced approximately two thousand years of Clinical verification. The prescriptions are commended by the later generation as classical prescriptions because of strict compatibility and remarkable treatment result. So Zhongjing prescriptions are selected as data set in this paper. A classification association rules mining algorithm based on information gain feature selection CARM-IG is achieved, which is used to mine compatibility principles of Zhang Zhongjing for the treatment of cold zheng, heat zheng, deficiency zheng and excess zheng etc. Experiments show that the run time and memory cost of algorithm is reduced, at the same time, the concision and precision of classification compatibility principles is guaranteed because of the deletion of irrelevant Chinese medicines with classification.