A Judgement Model of the Traditional Chinese Medicine's Quality based on Ensemble Algorithms

Abstract

Machine learning algorithms are currently not only widely used in image recognition and natural language processing, but also in other fields such as sports health and traditional Chinese medicine. Taking Polygonatum which is a kind of traditional Chinese medicine as an example, we use soil characteristics to predict the polysaccharide that reflects the quality of Polygonatum. This method can provide new ideas for the quality research on traditional Chinese medicine. With the ensemble model of traditional machine learning and deep learning algorithms, we have achieved a better effect of 15.9% on the MAPE (mean absolute percentage error), comparing with 16.3% on the best single model.