Method and device for predicting incidence relation between user and activity and storage medium

The invention provides a method and device for predicting an association relationship between a user and an activity and a storage medium, and the method comprises the steps: obtaining a relationship matrix between the user and the activity, and determining a similarity matrix of all portrait features between the users according to all portrait features of all historical users; determining a similarity matrix among historical activities according to the relation matrix of the users and the activities; performing fusion processing on the similarity matrixes of the portrait features among the users to obtain a user similarity fusion matrix; according to the similarity matrix between the historical activities, the user similarity fusion matrix and the relation matrix, utilizing a Laplacian regularization least square method to obtain a minimum cost matrix of the activities and a minimum cost matrix of the users; and integrating the minimum cost matrix of the user and the minimum cost matrix of the activity to ob