Enterprise benefit item pushing method based on spectral clustering algorithm

The invention relates to the technical field of intelligent data pushing, in particular to an enterprise benefit item pushing method based on a spectral clustering algorithm. According to the method, the enterprise-item matrix is determined by establishing the enterprise benefit item database and the enterprise database and analyzing the behavior of the enterprise user, and the enterprise-item category matrix is further established in order to avoid the situation that the enterprise-item matrix is too sparse. Vector representation of the enterprises is obtained according to the enterprise-matter category matrix, and the enterprise similarity based on the enterprise-matter category matrix is calculated through a cosine similarity calculation method. And determining an association map between the enterprises through the enterprise similarity based on the enterprise-item category matrix and the enterprise similarity based on the enterprise basic information. And clustering the enterprise nodes in the association