Knowledge perception recommendation method and system based on potential interest of user

The invention relates to a knowledge perception recommendation method and system based on potential interests of a user, and the method comprises the steps: constructing a collaborative knowledge graph, generating the embedded representation of entities and relationships in the collaborative knowledge graph, and transmitting associated knowledge and user preferences at the same time in a knowledge perception transmission mode, the importance of adjacent entities is distinguished through a knowledge perception attention mechanism, the embedded representation of each propagation layer of a user and a project is obtained, the high-order potential interest of the user is activated and learned on a collaborative knowledge graph, and a project correlation weight is generated by comparing with the interacted project of the user, so that the highest-order embedded representation of the user and the project is enhanced, and the user experience is improved. Information attenuation factors are adopted to distinguish inf