计算机科学与探索2025，Vol.19Issue(10)：2803-2814,12.DOI:10.3778/j.issn.1673-9418.2409034
基于异构信息网络的多模态食谱表示学习方法
Multimodal Recipe Representation Learning Method Based on Heterogeneous Information Networks
摘要
Abstract
Current cooking recipe representation learning methods primarily depend on aligning recipe texts with corre-sponding images or using adjacency matrix to capture relationships between cooking recipes and their ingredients for embed-ding learning.However,these methods are relatively rough in information fusion processing,fail to deeply mine the inter-action information between different modalities,and face challenges in effectively and dynamically evaluating the strength of correlations between cooking recipe components,restricting the model's representational capacity.To address these problems,this paper proposes a heterogeneous information network-based multimodal cooking recipe representation learning model(CookRec2vec)that integrates visual,textual,and relational information into cooking recipe embedding and fully mines and quantifies the correlation between the major components of the cooking recipes through adaptive adja-cency relationships.At the same time,an explicit modeling approach based on high-order co-occurrence matrices provides complementary information while preserving original characteristics,which significantly improves the expression ability of cooking recipe features.Experimental results show that the proposed model outperforms the existing mainstream methods in cooking recipe classification performance and has made significant progress in the field of innovative dish embedding prediction.关键词
表示学习/图嵌入/异构信息网络/跨模态融合/对抗攻击/节点分类Key words
representation learning/graph embedding/heterogeneous information network/cross-modal fusion/adversarial attack/node classification分类
信息技术与安全科学引用本文复制引用
张霄雁,江诗琪,孟祥福..基于异构信息网络的多模态食谱表示学习方法[J].计算机科学与探索,2025,19(10):2803-2814,12.基金项目
国家自然科学基金(61772249).This work was supported by the National Natural Science Foundation of China(61772249). （61772249）
