文本挖掘探索再生障碍性贫血证药相应规律

Exploring the correspondence law of formula and syndrome of aregenerative anemia by text mining

目的:探索再生障碍性贫血证药相应规律。方法:在中国生物医学文献数据库中下载相关再生障碍性贫血数据集,采用基于敏感关键词频数统计的数据分层算法,挖掘中医证候、症状及中药规律。结果:再生障碍性贫血中医证候以虚证为核心,核心症状为发热、乏力、头晕、瘀斑等,中药以补虚药物为核心。结论:文本挖掘技术,结合文献回溯,人工阅读降噪,能够比较客观地总结中医病、证、药的规律。

Objective： Explore the rules of Chinese herbal medicine under the framework of syndrome in traditional Chinese medicine （TCM）. Methods： Download the data set on aplastic anemia from Chinese BioMedical literature database. Then, the rules of TCM syndrome, symptoms, and Chinese herbal medicines were mined out. The results are shown in frequency tables and 2-dimention networks. Results： For TCM pattern, aplastic anemia（AA） is mainly focused on deficiency syndrome. For TCM internal organs, heart, spleen and liver are involved. For syndromes, fever, lack of power, and dizziness are the most significant. Whatg more, petechial, nose bleeding, pale face, joint pain and edema are also involved, yet less frequently. For Chinese herbal medicines, medicines on tonification are involved. Conclusions： Text mining, together with artificial reading for anti-noising, is an important approach in exploring the rules of TCM syndrome, symptom, and their association with Chinese herbal medicines.