Social platform user occupational prediction method based on graph convolution network

Abstract

The invention discloses a social platform user occupational prediction method based on a graph convolution network. The method comprises the following steps: firstly crawling social network data of auser to construct a network structure adjacency matrix, representing a user personal introduction by using bagofworks to construct a node attribute feature matrix, and then inputting the node attribute feature matrix into a graph convolutional network to obtain a network feature vector; and then splicing with a main user attribute feature vector constructed by the user basic attributes and the behavior attributes to obtain a main user feature vector, and inputting the main user feature vector into a logistic regression classifier for training to obtain a final occupational classification model. According to the invention, the data left by the user on the social platform is fully utilized, and the graph convolution network model is used to construct the social network, so that the user occupational prediction is more accurate; prediction of social network user occupations is beneficial to construction of user portraits, advertisement recommendation and user recommendation algorithms ofthe platform can be optimized, and then the user stickiness of the platform is effectively enhanced.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a method for predicting the occupation of social platform users based on a graph convolution network. Background technique [0002] Feature prediction of social platform users is a hot topic in natural language processing. For example, Sina Weibo, as a large social platform, has hundreds of millions of users, and users’ activities on the platform generate a large number of user attributes (personal profile, gender, age, region), user behavior (comments, reposts, likes) and User social relations (followers, fans) data, data mining research and applications provide a lot of data support. Predicting certain characteristics of users can provide support for users' personalized recommendations, which is conducive to providing better services for users and creating greater benefits for the platform. At the same time, more and more attention has been paid to network information...

AI Technical Summary

Problems solved by technology

The main problems to be solved by the present invention are: one is the problem of too few feature dimensions when the existing method predicts the user's occupation, that is, how to make full use of the structure of the social network and the attributes of the user to construct a neural network model for learning

Images