[Application of data mining on the relationship between and effect].

Objective: To analyze the relationship between and effect by data mining technique with retrieving clinical literature about acupuncture since the founding of the country.

Methods: The modern computerization and data mining technologies were adopted to set up clinical literature about acupuncture database. The relevant clinical literature was collected, screened, extracted, and statistical and correlation analyses were used so as to explore the relationship between and effect.

Results: 82.1% (46/56) of the studies considered that was related to the effect; 17.9% (10/56) of the studies considered that was unrelated to the effect. The support of related to the effect is 100% on dysmenorrhea and facial paralysis. 72% (18/25) of the the articles of pain syndrome considered that was related to the effect; 28% (7/25)of the studies considered that was unrelated to the effect. In the research of the relationship between the features of and effect, 60.7% of the studies suggested that the sense of conduction was positively correlated with the effect. There were 21 studies on the relationship between the intensity of and effect, involving a variety of diseases, which was related to the type of the disease. 58.3% (7/12) of the articles on dysmenorrhea and 63% (34/54) on pain syndrome supported conduction positively correlated to effect, showing the highest frequencies. 50% (3/6) of the papers on facial paralysis supported weak sensation positively correlated to effect, which was the highest frequency.

Conclusions: Most studies considered that can improve clinical efficacy. The relationship between different features of and effect is closely related to the disease. Further study may focus on high quality research on the relationship between and the obvious effect achieved by acupuncture so as to summarize the law of .