To solve the problem of traditional Chinese medicine recommendation, a knowledge graph - based recommendation model is proposed by incorporating more traditional Chinese medicine knowledge and methods into the process. First, a symptom - traditional Chinese medicine knowledge graph was constructed and applied in the pre - training stage to better preserve graph structure information. Then, based on the co - occurrence frequency in prescriptions, a symptom - symptom graph and a traditional Chinese medicine - traditional Chinese medicine graph were built to mine hidden rules among homogeneous nodes and obtain more comprehensive node embeddings. Furthermore, in the prediction layer of the model, a set of multiple symptoms is regarded as a whole, and a multi - layer perception is used to extract features from the symptom set, thereby better representing relationships among symptoms. Experimental results show that the model achieves favorable performance across three evaluation metrics.
RUANC Y, WANGY, ZHANGY C, et al. Exploring regularity in traditional Chinese medicine clinical data using heterogeneous weighted networks embedding[C]//International Conference on Database Systems for Advanced Applications: Springer,2019:310 - 313.
[3]
JINY, ZHANGW, HEX N, et al. Syndrome - aware herb recommendation with multi - graph convolution network[C]//2020 Ieee 36th International Conference on Data Engineering (icde): IEEE,2020:145 - 156.
[4]
JIS X, PANS R, CAMBRIAE, et al. A survey on knowledge graphs: representation, acquisition, and applications.[J]. IEEE transactions on neural networks and learning systems,2022,33(2):494 - 514.
[5]
YANGY, RAOY, YUM, et al. Multi - layer information fusion based on graph convolutional network for knowledge - driven herb recommendation[J]. Neural Networks,2022,146:1 - 10.
[6]
YAOL, ZHANGY, WEIB, et al. A topic modeling approach for traditional Chinese medicine prescriptions[J]. IEEE Transactions on Knowledge and Data Engineering,2018,30(6):1007 - 1021.
[7]
彭怀仁.中医方剂大辞典.第六册[M].北京:人民卫生出版社,1996.
[8]
国家药典委员会.中华人民共和国药典[M].北京: 中国医药科技出版社,2015.
[9]
XUH Y, ZHANGY Q, LIUZ M, et al. ETCM: an encyclopaedia of traditional Chinese medicine[J]. Nucleic Acids Research, 2019, 47(D1):D976 - D982.
[10]
YANGY, HUANGC, XIAL, et al. Knowledge graph contrastive learning for recommendation[C]//Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval.2022:1434 - 1443.
[11]
KIPFT N, WELLINGM. Semi - supervised learning with graph learning - convolutional networks[C]//Proceedings of the Ieee/cvf Conference on Computer Vision and Pattern Recognition,2019:11313 - 11320.
[12]
XUB, WANGN, CHENT, et al. Empirical evaluation of rectified activations in convolutional networks[C]// Proceedings of the 32nd International Conference on Machine Learning. JMLR Workshop and Conference Proceedings, 2015, 37: 1080 - 1088.
[13]
HAMILTONW, YINGZ, LESKOVECJ.Inductive representationlearning on large graphs[J].Advances in Neural Informationprocessing Systems,2017,30:1024 - 1034.
[14]
WANGX, HEX N, CAOY X, et al. Kgat:knowledge graph attention network for recommendation[C]//Proceedings of the 25th Acm Sigkdd International Conference on Knowledge Discovery & Data Mining,2019:950 - 958.
[15]
WANGH, ZHAOM, XIEX,et al. Knowledge graph convolutional networks for recommender systems[C]//The World Wide Web Conference,2019:3307 - 3313.
[16]
DUY, ZHUX, CHENL,et al. HAKG: hierarchy - aware knowledge gated network for recommendation[C]//Proceedings of the 45th International Acm Sigir Conference on Research and Development in Information Retrieval,2022:1390 - 1400.
[17]
HEX, DENGK, WANGX,et al. Lightgcn: simplifying and powering graph convolution network for recommendation[C]//Proceedings of the 43rd International Acm Sigir Conference on Research and Development in Information Retrieval,2020:639 - 648.