Node recommendation method in network graph, server and storage medium

Abstract

The embodiment of the invention discloses a node recommendation method in a network graph, a server and a storage medium, which are used for improving the recommendation efficiency. The embodiment ofthe invention provides a node recommendation method in a network graph, which comprises the following steps that: acquiring a first node from a network graph pre-stored in a recommendation server, wherein the network graph comprises N nodes, adjacent nodes connected through edges exist in the N nodes, the value of N is a positive integer, and the first node belongs to N nodes; from the first node,randomly walking Omega strips in a network diagram at a preset stop probability, with Omega as a positive integer; when the raondomly walking Omega strips are stopped at M nodes, determining personalized webpage ranks PPR according to M nodes corresponding to Omega and randomly walking stopping strips corresponding to M nodes; and determining a second node from the M nodes according to the PPR corresponding to the M nodes, and recommending the second node to the first node.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a node recommendation method in a network graph, a server and a storage medium. Background technique [0002] On a large-scale social network, each user can be represented by a node, and the personalized page rank (Personalized Page Rank, PPR) can be calculated, and the correlation between users can be represented by PPR. For example, relationship chains between people in a social network can be determined. [0003] Existing technologies provide a distributed graph processing framework. For example, Spark GraphX ​​is one of the processing frameworks. It is based on the Spark platform and provides simple, easy-to-use and rich interfaces for graph computing and graph mining, which greatly facilitates the processing of distributed graphs. Requirements for graph processing. [0004] Spark's GraphX ​​has a PPR algorithm package. The PPR in GraphX ​​uses the calculation method of p...

AI Technical Summary

Problems solved by technology

[0008] In the PPR algorithm of GraphX ​​provided by the prior art, multiple matrices whose size is the square of n are required, where n is the number of nodes in the graph. However, the social network The number of nodes is usually more than tens of millions, which consumes a lot of storage space

And because it needs multiple iterative calculations, it has high computational complexity and reduces the recommendation efficiency

Images