Link prediction method in dynamic social network

Abstract

The invention relates to a link prediction method in a dynamic social network, which comprises the following steps of: mapping nodes in the network at the moment t into a low-dimensional embedded space, and writing the low-dimensional embedded space into a low-dimensional representation vector of each node; then respectively calculating local features and second-order similarity of nodes in the network at the moment t and a loss function corresponding to keeping network evolution smoothness, and finally obtaining an optimal low-dimensional representation vector of the nodes according to a minimized total loss function; obtaining low-dimensional representation vectors of all nodes in the test set by using an optimal low-dimensional representation vector method, and sequentially inputting the low-dimensional representation vector of each node pair into a logistic regression classifier for training to obtain a trained logistic regression classifier; and inputting the low-dimensional representation vector of each node pair in the network at the moment T into the trained logistic regression classifier to obtain network information at the moment T + 1. According to the link prediction method, the data storage space in the network is reduced, and the link prediction accuracy is higher.

Description

technical field [0001] The invention relates to the field of link prediction, in particular to a link prediction method in a dynamic social network. Background technique [0002] With the continuous accumulation of massive data in social, communication, biological and other networks, this kind of network structured data very effectively simulates various types of linked data in the real world. Among them, nodes represent entities, and edges represent links between entities. Mining network information, especially link information, has become a new research direction. Link prediction is based on the existing structure of the social network to predict the hidden links or the links that may be generated in the future. In addition to its high academic research value, link prediction also has many important commercial applications. For example, social networking sites such as Facebook recommend friends; e-commerce sites such as Taobao recommend products of interest to users; me...

AI Technical Summary

Problems solved by technology

Most of the current research ignores the time information of the network. The obtained node vector representation only captures the structural information of the network at the current moment, and cannot capture the evolution trend information hidden in the network structure, which greatly reduces the accuracy of link prediction.

Images