Research on resource allocation methods for traditional Chinese medicine services based on deep reinforcement learning

Yuntao Ma, Xiaolin Fang, Jin Qi, Yanfei Sun

Abstract: Chinese medicine resources are the crystallization of traditional Chinese culture, and more and more people are choosing Chinese medicine services for their health. To address the problems of pluralistic heterogeneity, waste of service resources, and lagging demand response in the resource allocation model for traditional Chinese medicine (TCM) services, a deep reinforcement learning-based resource allocation method for TCM services is proposed. To address the fragmentation of TCM service resources, this paper presents a TCM service resource association method based on improved spectral clustering and establishes a good resource-demand matching model. For the problem of TCM service resource allocation after resource association, we establish a TCM service resource allocation model and collaboratively solve the TCM service resource allocation problem via the deep reinforcement learning method. The results show that the proposed solution can accelerate the demand response of TCM service resources, effectively reduce the cost of TCM services for patients, improve the quality of TCM services, and satisfy the demand for TCM services for patients.