Abstract: Traditional Chinese medicine (TCM) is an interesting research topic in China’s thousands of years of history. With the recent advances in artificial intelligence technology, some researchers have started to focus on learning the TCM prescriptions in a data-driven manner. This involves appropriately recommending a set of herbs based on patients’ symptoms. Most existing herb recommendation models disregard TCM domain knowledge, for example, the interactions between symptoms and herbs and the TCM-informed observations (i.e., TCM formulation of prescriptions). In this paper, we propose a knowledge-guided and TCM-informed approach for herb recommendation. The knowledge used includes path interactions and co-occurrence relationships among symptoms and herbs from a knowledge graph generated from TCM literature and prescriptions. The aforementioned knowledge is used to obtain the discriminative feature vectors of symptoms and herbs via a graph attention network. To increase the ability of herb prediction for the given symptoms, we introduce TCM-informed observations in the prediction layer. We apply our proposed model on a TCM prescription dataset, demonstrating significant improvements over state-of-the-art herb recommendation methods.

一种知识引导的基于中医学信息的药材推荐方法

[1]Atwood J, Towsley D, 2016. Diffusion-convolutional neural networks. Proc 30^th Int Conf on Neural Information Processing Systems, p.2001-2009.

[2]Bordes A, Usunier N, Garcia-Duran A, et al., 2013. Translating embeddings for modeling multi-relational data. Proc 26^th Int Conf on Neural Information Processing Systems, p.2787-2795.

[3]Braga-Neto U, 2020. Fundamentals of Pattern Recognition and Machine Learning. Springer, Cham, Switzerland.

[4]Brody S, Alon U, Yahav E, 2022. How attentive are graph attention networks? Proc 10^th Int Conf on Learning Representations.

[5]Bruna J, Zaremba W, Szlam A, et al., 2014. Spectral networks and locally connected networks on graphs. Proc 2^nd Int Conf on Learning Representations.

[6]Chen XT, Ruan CY, Zhang YC, et al., 2018. Heterogeneous information network based clustering for categorizations of traditional Chinese medicine formula. Proc IEEE Int Conf on Bioinformatics and Biomedicine, p.839-846.

[7]Cheung F, 2011. TCM: made in China. Nature, 480(7378):S82-S83.

[8]Do K, Tran T, Venkatesh S, 2019. Graph transformation policy network for chemical reaction prediction. Proc 25^th ACM SIGKDD Int Conf on Knowledge Discovery & Data Mining, p.750-760.

[9]Gehring J, Auli M, Grangier D, et al., 2017. A convolutional encoder model for neural machine translation. Proc 55^th Annual Meeting of the Association for Computational Linguistics, p.123-135.

[10]Girshick R, Donahue J, Darrell T, et al., 2014. Rich feature hierarchies for accurate object detection and semantic segmentation. Proc IEEE Conf on Computer Vision and Pattern Recognition, p.580-587.

[11]Guo SN, Lin YF, Feng N, et al., 2019. Attention based spatial-temporal graph convolutional networks for traffic flow forecasting. Proc 33^rd AAAI Conf on Artificial Intelligence, p.922-929.

[12]Hamilton WL, Ying R, Leskovec J, 2017. Inductive representation learning on large graphs. Proc 31^st Int Conf on Neural Information Processing Systems, p.1025-1035.

[13]Ji WD, Zhang Y, Wang XL, et al., 2017. Latent semantic diagnosis in traditional Chinese medicine. World Wide Web, 20(5):1071-1087.

[14]Jin YY, Zhang W, He XN, et al., 2020. Syndrome-aware herb recommendation with multi-graph convolution network. Proc IEEE 36^th Int Conf on Data Engineering, p.145-156.

[15]Jin YY, Ji WD, Zhang W, et al., 2022. A KG-enhanced multi-graph neural network for attentive herb recommendation. IEEE/ACM Trans Comput Biol Bioinform, 19(5):2560-2571.

[16]Kingma DP, Ba J, 2015. Adam: a method for stochastic optimization. Proc 3^rd Int Conf on Learning Representations.

[17]Kipf TN, Welling M, 2017. Semi-supervised classification with graph convolutional networks. Proc 5^th Int Conf on Learning Representations.

[18]LeCun Y, Bottou L, Bengio Y, et al., 1998. Gradient-based learning applied to document recognition. Proc IEEE, 86(11):2278-2324.

[19]Li W, Yang Z, 2017. Distributed representation for traditional Chinese medicine herb via deep learning models. https://arxiv.org/abs/1711.01701

[20]Li W, Yang Z, 2019. Exploration on generating traditional Chinese medicine prescriptions from symptoms with an end-to-end approach. Proc 8^th CCF Int Conf on Natural Language Processing and Chinese Computing, p.486-498.

[21]Lin F, Xiahou J, Xu ZX, 2016. TCM clinic records data mining approaches based on weighted-LDA and multi-relationship LDA model. Multim Tools Appl, 75(22):14203-14232.

[22]Lu YT, Liu HY, Yang Kuo, et al., 2022. A comprehensive update: gastrointestinal microflora, gastric cancer and gastric premalignant condition, and intervention by traditional Chinese medicine. J Zhejiang Univ-Sci B (Biomed & Biotechnol), 23(1):1-18.

[23]Ma JL, Wang ZJ, 2016. Discovering syndrome regularities in traditional Chinese medicine clinical by topic model. Proc 11^th Int Conf on P2P, Parallel, Grid, Cloud and Internet Computing, p.157-162.

[24]Mao YJ, Meng LK, Liu HY, et al., 2022. Therapeutic potential of traditional Chinese medicine for vascular endothelial growth factor. J Zhejiang Univ-Sci B (Biomed & Biotechnol), 23(5):353-364.

[25]Raissi M, Perdikaris P, Karniadakis GE, 2019. Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. J Comput Phys, 378:686-707.

[26]Ruan CY, Ma JG, Wang Y, et al., 2019a. Discovering regularities from traditional Chinese medicine prescriptions via bipartite embedding model. Proc 28^th Int Joint Conf on Artificial Intelligence, p.3346-3352.

[27]Ruan CY, Wang Y, Zhang YC, et al., 2019b. Exploring regularity in traditional Chinese medicine clinical data using heterogeneous weighted networks embedding. Proc Int Conf on Database Systems for Advanced Applications, p.310-313.

[28]Vaswani A, Shazeer N, Parmar N, et al., 2017. Attention is all you need. Proc 31^st Int Conf on Neural Information Processing Systems, p.6000-6010.

[29]Veličković P, Cucurull G, Casanova A, et al., 2018. Graph attention networks. Proc 6^th Int Conf on Learning Representations.

[30]Wang S, Huang EW, Zhang RS, et al., 2016. A conditional probabilistic model for joint analysis of symptoms, diseases, and herbs in traditional Chinese medicine patient records. Proc IEEE Int Conf on Bioinformatics and Biomedicine, p.411-418.

[31]Wang X, Ji HY, Shi C, et al., 2019. Heterogeneous graph attention network. Proc World Wide Web Conf, p.2022-2032.

[32]Wang XY, Zhang Y, Wang XL, et al., 2019. A knowledge graph enhanced topic modeling approach for herb recommendation. Proc 24^th Int Conf on Database Systems for Advanced Applications, p.709-724.

[33]Xu KYL, Hu WH, Leskovec J, et al., 2019. How powerful are graph neural networks? Proc 7^th Int Conf on Learning Representations.

[34]Yang Y, Rao YL, Yu MH, et al., 2022. Multi-layer information fusion based on graph convolutional network for knowledge-driven herb recommendation. Neur Netw, 146:1-10.

[35]Yao L, Zhang Y, Wei BG, et al., 2018. A topic modeling approach for traditional Chinese medicine prescriptions. IEEE Trans Knowl Data Eng, 30(6):1007-1021.

[36]Zhang JN, Shi XJ, Xie JY, et al., 2018. GaAN: gated attention networks for learning on large and spatiotemporal graphs. Proc 34^th Conf on Uncertainty in Artificial Intelligence, p.339-349.

[37]Zhang S, Tong HH, Xu JJ, et al., 2019. Graph convolutional networks: a comprehensive review. Comput Soc Netw, 6(1):11.

[38]Zhao GS, Zhuang XT, Wang XM, et al., 2018. Data-driven traditional Chinese medicine clinical herb modeling and herb pair recommendation. Proc 7^th Int Conf on Digital Home, p.160-166.

[39]Zhao W, Lu WK, Li ZY, et al., 2022. TCM herbal prescription recommendation model based on multi-graph convolutional network. J Ethnopharmacol, 297:115109.

[40]Zheng CP, Fan XL, Wang C, et al., 2020. GMAN: a graph multi-attention network for traffic prediction. Proc 34^th AAAI Conf on Artificial Intelligence, p.1234-1241.

[41]Zhou WA, Yang K, Zeng JY, et al., 2021. FordNet: recommending traditional Chinese medicine formula via deep neural network integrating phenotype and molecule. Pharmacol Res, 173:105752.

[42]Zitnik M, Agrawal M, Leskovec J, 2018. Modeling polypharmacy side effects with graph convolutional networks. Bioinformatics, 34(13):i457-i466.