A study of Chinese herbal properties based on machine learning

A concept of Four Properties (SiQi) of Chinese herbs is the important part of traditional Chinese medicine theory. The Chinese clinical medicine is a process of dialectical theory of governance of Chinese medicine prescriptions based these four properties. The Chinese medicine prescription uses a "Cold" and "Hot" model to judge the properties of Chinese herbs, and also judge the properties of these composed recipes. It is both important and difficult in the Chinese medical practice to use the model, and therefore, it has become one of the hot issues to be addressed in the research of the modern Chinese clinical medicine. This paper studies the entire prescription as a whole. According to the property of every composing individual herb in the recipe and overall effects and by using several machine-learning methods a new system has demonstrated an optimization for the entire prescriptions. This system is designed and implemented as an online forecasting system to determine the overall properties for a modified traditional well-known prescription. It may help the doctor to decide the dose of each composing herb to automatically generate a well-tuned recipe as the final prescription to fully balance the four properties for his patients.