On Herb Compatibility Rule of Insomnia Based on Machine Learning Approaches

Recent research in machine learning has led to significant progress in various research fields. Especially, the knowledge discovery using this method in Traditional Chinese Medicine (TCM) has been becoming a hot topic. In this paper, we studied on the herb compatibility rule of insomnia using some machine learning approaches. We have extracted insomnia data set with 807 samples from the real-world Electronic Medical Records (EMRs). After cleaning and selecting the theme data referring to the prescriptions and their herbs, we constructed the herb network analysis model using the theory of complex network. In order to explore the hidden relationships among the herbs, we trained each herb node in network to obtain the herb embeddings using the Skip-Gram model in word embedding theory. After acquiring the vocabulary of herbs with the formation of vectors, we calculated the similarity among any two herb embeddings, and clustered these herb embeddings into seven communities using the Spectral Clustering (SC) algorithm. The experimental results shed light on that the methodologies used in this paper can objectively and effectively discover the relationships among herbs, and reveal the herb compatibility and herb clusters for clinical treatment research of insomnia.