Microblog friend recommendation method based on heterogeneous information network

Abstract

The invention discloses a microblog friend recommendation method based on a heterogeneous information network. The microblog friend recommendation method comprises the following steps: S1, constructing heterogeneous information network data of a microblog user; S2, extracting meta-paths in the heterogeneous information network, and converting the heterogeneous information network into a homogeneous information network; S3, obtaining vector representation of the user by using a graph convolution network; and S4, calculating the similarity between different users and carrying out similarity fusion. By using the heterogeneous information network and the graph convolution network, rich user behavior information is captured, and high-performance user vector representation is obtained, so that the recommendation accuracy is effectively improved.

Description

technical field [0001] The invention belongs to the technical field of the Internet, and in particular relates to a microblog friend recommendation method based on a heterogeneous information network. Background technique [0002] Recommender systems are playing an increasingly important role in different online services. Taking Weibo as an example, in the process of using Weibo, it can recommend other users who may know or have similar hobbies for users, thereby improving user Usage rate and satisfaction of Weibo. [0003] Microblog friend recommendation involves research on social networks. At this stage, network research usually adopts a homogeneous information network with a single structure, that is, all nodes in the network are of the same type, focusing on the internal structure and characteristics of the network; and homogeneous It is difficult for the information network to integrate different data information, thus ignoring the diversity of behavior among users. ...

AI Technical Summary

Problems solved by technology

[0003] Microblog friend recommendation involves research on social networks. At this stage, network research usually adopts a homogeneous information network with a single structure, that is, all nodes in the network are of the same type, focusing on the internal structure and characteristics of the network; and homogeneous It is difficult for the information network to integrate different data information, thus ignoring the diversity of behavior among users

In addition, most application nodes in the real world also contain multiple link relationships, and their network structure is more complex and information is richer. The homogeneous information network cannot express multiple complex entities and the connection relationship between entities.

Images