Item recommendation method and system based on collaborative filtering algorithm

Abstract

The invention relates to the technical field of automatic recommendation, and in particular relates to an item recommendation method and system based on a collaborative filtering algorithm. The method comprises the following steps: performing an cluster analysis on item scoring behaviors of m basic users offline to obtain k user clusters; determining a cluster center matrix Ckxn and a class subordination degree matrix Vmxk offline; carrying out an item recommendation on a target user by utilizing the cluster center matrix Ckxn and the class subordination degree matrix Vmxk online. By obtaining the k user clusters offline in advance, the cluster center matrix Ckxn and the class subordination degree matrix Vmxk are further obtained, and the working amount of online recommendation is reduced, so that the aim of guaranteeing a recommendation generation speed is realized, and a speed bottleneck problem caused by a traditional collaborative filtering algorithm in a recommendation generation speed is solved.

Description

technical field [0001] The invention relates to the technical field of automatic recommendation, in particular to an item recommendation method and system based on a collaborative filtering algorithm. Background technique [0002] Today, with the development of the Internet and e-commerce, the recommendation system has gradually become an important research content of e-commerce, and has attracted more and more researchers' attention. At present, almost all large-scale e-commerce systems, such as Amazon, CDNOW, eBay, etc., use various forms of automatic recommendation systems to varying degrees. [0003] In the existing item recommendation process based on collaborative filtering algorithm, firstly, the rating behavior of basic users is clustered through the ratings of basic users on items (such as commodities in e-commerce, movies, music, web pages and documents, etc.) Analysis, the basic users are divided into several categories, combined with the collaborative filtering ...

AI Technical Summary

Problems solved by technology

[0004] The existing above-mentioned entire recommendation process is completed online. However, with the expansion of the scale of the e-commerce system, the amount of data used to generate recommended item data is also increasing. If the above-mentioned entire recommendation process is still completed online, the workload It takes a long time, which will seriously affect the speed of recommendation generation

Images