Social network user multi-label classification method based on dynamic multi-view learning model

Abstract

The invention provides a social network user multi-label classification method based on a dynamic multi-view learning model, and the method comprises the following steps: building multi-view representation of a user for a specific social network data set; based on the user representation, constructing a deep fusion representation model among the multi-view data; using a dynamic routing model, updating parameters, and optimizing multi-view characteristics; and introducing a shared representation model, and constructing an objective function for the features in the step 3; and through model optimization, obtaining an optimal shared representation matrix, and finally, realizing multi-label classification of any user by utilizing the shared matrix. According to the method, multi-label efficient classification of network users is realized. The problems of model learning performance reduction, limited view fusion quantity, incapability of meeting multi-classification task requirements of the model and the like caused by data loss are solved, and the method can be widely applied to scenes such as user accurate analysis, abnormal user detection, user relationship mining and unknown user identification in the network.

Description

technical field [0001] The invention discloses a multi-label classification method for social network users based on a dynamic multi-view learning model, and belongs to the field of information technology services. Background technique [0002] At present, the Internet has become an indispensable part of people's lives. Users generate a large amount of data such as text, images, and user relationships in different social networks (such as Weibo, Twitter, and WeChat), which are called multi-view data. These multi-view data imply a wealth of information, and the update speed is extremely fast. Therefore, the fusion of multi-view data has become a key technical issue in the field of data mining, and its research results can be applied to different fields. Network user analysis is an important application scenario. Users may have multiple category labels at the same time. In recent years, multi-label classification of the same user on different networks has attracted extensive a...

AI Technical Summary

Problems solved by technology

[0003] According to research, due to the constraints of multi-source and multi-model points of network user data, the existing multi-label classification methods still have the following deficiencies: (1) The performance of multi-label prediction depends too much on user data. (2) Limited by the model, most methods only realize the fusion learning of two views; (3) Some new methods can achieve multi-view fusion and achieve better performance, but What is done is the clustering task, it is difficult to achieve the optimal multi-label classification model

Images