Menu Knowledge Based Propagation Graph Attention Network Using Consumer Interaction and Behavior

In the era of large volumes of data (big data), suggestions on various methods are crucial, as they utilize vast amounts of data and consumer behavior information to suggest menus based on eating preferences. A novel approach to personalized menu suggestion is proposed, based on a project where the attention of a graph is diverted toward the Neural Network. The network captures the structural message of parts in the menu knowledge graph by fine-grained representation of consumers and their recipes. It uses a bidirectional dissemination approach to various sets of the consumer and system, learns topological adjacent structure features, and designs an attention aggregator to produce a refined embedding. The core product of the things predicts the probability of consumer-menu interactions. Experimental results show that CGRS outperforms baseline advanced suggestion models.