Collaborative filtering recommendation algorithm based on high-order connection graph

Abstract

The invention relates to a collaborative filtering recommendation algorithm based on a high-order connection graph, belongs to the technical field of recommendation algorithms and recommendation systems, and aims to alleviate the problems of cold start and data sparseness of a traditional recommendation algorithm in practical application. In order to solve the problem, a high-order connection graph is introduced, a unified heterogeneous graph is established for multiple users, multiple projects and multiple relations, the high-order relations between the users, between the users and the projects and between the projects are mined, and the influence of data sparsity and cold start on a recommendation result is relieved. In the method, the high-order connection graph is a heterogeneous graph, nodes serve as entities, and edges represent relationships among the entities. The items and their attributes may be mapped into a higher order connection graph to understand higher order connectivity between the items. Moreover, the user and the user party information can also be integrated into the graph, so that the relationship between the user and the project and the preference of the user can be captured more accurately.

Description

technical field [0001] The invention belongs to the technical field of recommendation systems, and relates to a collaborative filtering recommendation algorithm based on high-order connection graphs. Background technique [0002] With the rapid development of Internet technology, technologies such as cloud computing, big data, and the Internet of Things have gained unprecedented popularity. Various types of data in the Internet space are emerging one after another, triggering an explosive growth in data scale. Human beings have entered the era of content overload and data noise, and can obtain richer information through various channels. Various social tools such as Weibo and WeChat and self-media have brought users more convenient channels for information acquisition. At the same time, information overload has also become a problem for people to encounter. This has led to one of the characteristics of the era of big data. The data capacity is huge, but the data value densi...

AI Technical Summary

Problems solved by technology

However, all these methods still suffer from the data sparsity problem since most of them cannot extensively exploit high-order structural information in recommender systems

Images