Design of Knowledge Graph of Traditional Chinese Medicine Prescription and Knowledge Analysis of Implicit Relationship

ABSTRACT

Purpose-As of March 2020, this paper collected 41 prescriptions for Covid-19 in official reports, and used machine learning and knowledge graph technology to discover the rules of common prescription medication for Covid-19 and the Chinese medicinal herb groups applicable to different groups of people to help the TCM prescriptions for the clinical treatment of Covid-19. Methods-The FP-Growth algorithm was used to analyze the properties, tastes, and meridian tropism of Chinese medicinal herb groups in prescription data, constructing a knowledge graph for Covid-19 description and mining the prescription medication rules. Results-In the 41 pieces of prescription, patients in medical observation period were treated with Chinese medicinal herbs that can relieve exterior syndromes and stop vomiting, such as Radix Saposhnikoviae, Semen Sojae Preparatum, etc. Those with mild common syndromes were treated with heat-clearing and detoxicating Chinese medicinal herbs, such as Rhizoma Coptidis, Radix Scutellariae, etc. Infected patients were treated with Chinese medicinal herbs such as Radix Glycyrrhizae, Herba Asari, etc., which can dissolve phlegm, relieve cough and invigorate the spleen, relieve pain, expel wind syndromes, relieve exterior symptoms and help digesting. Conclusion-Combining machine learning and knowledge graph technology to analyze the data characteristics of common prescriptions used by different populations can help analyze the core medication mechanism of TCM (Traditional Chinese Medicine) treatment of different populations with the same disease. © 2020 ACM.