Utilizing semantically enhanced self-supervised graph convolution and multi-head attention fusion for herb recommendation by Xianlun Tang ISBN 101016/JARTMED2025103112 instant download

Utilizing semantically enhanced self-supervised graph convolution and multi-head attention fusion for herb recommendation by Xianlun Tang ISBN 101016/JARTMED2025103112 instant download

ABSTRACT Dataset link: Chinese herbal medicine has long been recognized as an effective natural therapy. Recently,SBC/BSGAMthe development of recommendation systems for herbs has garnered widespread academic attention, asKeywords:these systems significantly impact the application of traditional Chinese medicine. However, existing herbrecommendation systems are limited by data sparsity, insufficient correlation between prescriptions, andHerb recommendationinadequate representation of symptoms and herb characteristics. To address these issues, this paper introducesSemantic enhancementSelf-supervised graph convolutionan approach to herb recommendation based on semantically enhanced self-supervised graph convolution andMulti-head attentionmulti-head attention fusion (BSGAM). This method involves efficient embedding of entities following fineRepresentation learningtuning of BERT; leveraging the attributes of herbs to optimize feature representation through a residual graphconvolution network and self-supervised learning; and ultimately employing a multi-head attention mechanismfor feature integration and recommendation. Experiments conducted on a publicly available traditional Chinesemedicine prescription dataset demonstrate that our method achieves improvements of 6.80%, 7.46%, and6.60% in F1-Score@5, F1-Score@10, and F1-Score@20, respectively, compared to baseline methods. Theseresults confirm the effectiveness of our approach in enhancing the accuracy of herb recommendations.