Social network alignment method

The invention belongs to the technical field of Internet, and particularly relates to a social network alignment method, which comprises the following steps of: acquiring network user data, forming a first social network graph and a second social network graph according to a user relationship, forming user pairs by taking a part of users to form a third social network graph, and calculating an adjacent matrix; inputting the adjacent matrix into an encoder to obtain a first user node representation matrix; inputting the first user node representation matrix and the adjacent matrix into a denoising model to obtain a covering matrix; obtaining a covering adjacent matrix according to the covering matrix and the adjacent matrix, and inputting the covering adjacent matrix into the encoder to obtain a second user node representation matrix; according to the second user node representation matrix, obtaining a difference value between the first similarity and the second similarity, and updating learning parameters of