| abstract |
The invention provides an implicit group discovery method based on the hidden characteristics of online social users, and relates to the field of network technology. It includes the following steps: constructing a user social relationship matrix for the social network; using a sparse autoencoder to learn the latent features of the user's social relationship; using the hidden layer output encoding matrix for robust continuous clustering; using the feature matrix to build a graph structure, from the graph structure The implicit groups are determined in the connected branch of . By discovering the internal mechanism of user connection, the present invention considers the connection relationship between users from the perspective of implicit features, which is more in line with the actual implicit group aggregation situation, and realizes more accurate discovery of implicit user groups. |