Remote interest point recommendation method based on user preference and personalized preference transfer

The invention discloses a remote interest point recommendation method based on user preference and personalized preference transfer, and belongs to the field of terminal positioning recommendation, and the method comprises the following steps: constructing a heterogeneous hypergraph for five different types of nodes, and training the hypergraph to obtain representation of user preference; constructing a POI (Point of Interest)-category graph, and realizing representation learning of the POI through a continuous character skipping model; constructing an attention network, and taking the representation of the POI as input to obtain transferable features of the user; a parameter learning network is established through a multi-layer perceptron, transferable features of a user are used as input, a transfer network is established, preference characterization of the user is used as input, output of the parameter learning network is used as parameters, and therefore personalized preference transfer of the user is ach