A TCM Disease Retrieval Method Based on Weighted Word Vector

Liwen Deng,Jiajun Xu,Shaozi Li
DOI: https://doi.org/10.1109/itme56794.2022.00096
2022-01-01
Abstract:TCM intelligent diagnosis algorithm is of great significance to promote the standardization, intelligence and popularization of TCM consultation. Disease diagnosis task is the key link of TCM intelligent diagnosis algorithm. In recent years, word vector models have been applied to disease diagnosis tasks. However, it is difficult to use the word vector model for disease diagnosis alone to meet the requirements of accuracy. In order to improve this situation, this paper transforms the problem of disease diagnosis into a disease retrieval problem, and designs and implements a TCM disease retrieval method based on weighted word vectors. The method retrieves the patient’s symptoms collected by the TCM intelligent diagnosis algorithm in the TCM disease database, and finally retrieves the disease that best matches the patient’s symptoms as the disease diagnosis result. The experimental results show that the method has a good effect on the task of disease diagnosis, and can provide a new idea for the realization of the task of disease diagnosis.
What problem does this paper attempt to address?