Method for predicting heterogeneous network link based on meta-path

Abstract

The invention discloses a heterogeneous network link prediction method based on a meta-path. Rich and complex semantics of the heterogeneous network are processed by utilizing meta-paths; meta-path instances are extracted from the network and network node weights are learnt; the network topology characteristics are integrated to update the network node embedding, the updated network node embeddingis utilized to perform classification, clustering, link prediction and other work, the applicability in the actual life is high, and the accuracy of the classification, clustering, link prediction and other work of the heterogeneous network can be effectively improved.

Description

technical field [0001] The invention relates to the technical field of data mining, in particular to a method for predicting heterogeneous network links based on meta-paths. Background technique [0002] With the development of science and technology and the advent of the era of big data, the recommendation system has emerged as the times require, which aims to help users find the content they are most interested in more conveniently in massive data. Link prediction is a bridge between complex networks and information technology. It can effectively restore missing links and predict upcoming links, and is an important method to implement recommendation systems. The network interlacing in real life is complex, which brings great challenges to the accuracy of link prediction. An effective link prediction method can provide users with accurate prediction results and has important application value. However, existing network link prediction methods often ignore the impact of net...

AI Technical Summary

Problems solved by technology

The complex interlacing of networks in real life brings great challenges to the accuracy of link prediction. An effective link prediction method can provide users with accurate prediction results and has important application value

However, the existing network link prediction methods often ignore the impact of network heterogeneity on link prediction. Most of the existing methods are aimed at homogeneous networks, but most real networks are heterogeneous.

Images