Lightweight socialized recommendation method based on hash learning

Abstract

The invention provides a lightweight socialized recommendation method based on hash learning. The method comprises the following steps: constructing a user-article scoring matrix and a user-user social network, and generating a social corpus by applying truncated random walk and negative sampling to the user-user social network; training a discrete matrix decomposition and continuous network embedding hybrid model according to the user-article scoring matrix and the social corpus to obtain a binarized user feature matrix and an article feature matrix; and according to the user feature matrix and the article feature matrix, estimating preference scores of the user for the unscored articles, and recommending one or more unscored articles with the highest estimated score to the user. The performance of the method is equivalent to that of a current mainstream real-valued recommendation method, but due to the fact that a lightweight model design thought is adopted, the obtained binarized user and article characteristics have lower calculation and storage expenditure.

Description

technical field [0001] The present invention relates to the field of computer application technology, in particular to a lightweight social recommendation method based on hash learning. Background technique [0002] As an effective supplement to information retrieval systems, recommendation systems play an important role in providing personalized information services. Collaborative filtering is the core technology for building a personalized recommendation system; among many collaborative filtering methods, matrix decomposition is one of the most mainstream methods at present. The core idea of ​​matrix decomposition is to map users and items to the same low-dimensional latent space by decomposing a partially observed "user-item" interaction matrix (referred to as UI matrix), and then predict Unobserved user-item correlations. Usually, the observed "user-item" interaction records only account for a small part of the UI matrix, which is the so-called "data sparsity" problem,...

AI Technical Summary

Problems solved by technology

[0006] 1) The DSR model learns the social representation of users through the method of shared variables. This method only considers the direct connection between each user and its first-order neighbors, and ignores the indirect connection between users and their high-order neighbors. The obtained user social features need to be further improved;

[0007] 2) Due to the shared variable design idea, the user social features learned by the DSR model are also binary; however, since the user social representation is only a by-product of the modeling process and does not participate in the final recommendation calculation, the binary representation is compared to A real value means that it carries less information, thus causing unnecessary coding loss

Images