Title :
On classification compatibility laws of Zhongjing prescriptions
Author :
Yang, Xuemei ; Lai, Xinmei ; Li, Candong
Author_Institution :
Fujian Univ. of Traditional Chinese Med., Fuzhou, China
Abstract :
Zhongjing prescriptions are selected as data set in this paper. First, a classification association rules mining algorithm, which selected features based on information gain, short for CARM-IG, is achieved. The algorithm is used to mine compatibility laws of Zhang Zhongjing from cold-heat, deficiency-excess in eight principals of TCM (Traditional Chinese Medicine). Second, 21 rules which are applied to the treatment of cold, heat or cold-heat mixing syndromes are mined. And 29 rules which are applied to the treatment of deficiency, excess or deficiency-excess mixing syndromes are discovered too. The reduction of run time and memory cost of algorithm CARM-IG had been proved, at the same time, the concision and precision of classification compatibility laws are guaranteed because of the deletion of irrelevant Chinese medicines with classification. Finally, the compatibility laws of Zhang Zhongjing are listed in this paper, which would be used to direct clinical practice sometimes.
Keywords :
bioinformatics; data mining; medicine; patient treatment; pattern classification; CARM-IG; Traditional Chinese Medicine; Zhang Zhongjing; Zhongjing prescriptions; classification association rules mining algorithm; classification compatibility laws; cold-heat mixing syndromes; deficiency-excess mixing syndromes; information gain; mine compatibility laws; classification association rules; features selection; prescription compatibility laws of Zhang Zhongjing;
Conference_Titel :
Bioinformatics and Biomedicine Workshops (BIBMW), 2010 IEEE International Conference on
Conference_Location :
Hong, Kong
Print_ISBN :
978-1-4244-8303-7
Electronic_ISBN :
978-1-4244-8304-4
DOI :
10.1109/BIBMW.2010.5703895