Collaborative filtering recommendation method based on parallel auto-encoder

Abstract

The invention discloses a collaborative filtering recommendation method based on a parallel auto-encoderparallel self-encoding machine, which comprises the following steps: 1, constructing a sparse auto-encoderself-encoding machine model to complete an objective function expressed by potential characteristics of a user, and learning high-level abstract characteristics based on the user to obtain areconstruction matrix of a user scoring matrix; 2, constructing a graph regularization automatic coding machine model to complete an objective function of commodity potential feature representation,and learning high-level abstract features based on commodities to obtain a the reconstruction matrix of a commodity scoring matrix; and 3, performing matrix multiplication on the reconstruction matrixbased on the user scoring matrix and the reconstruction matrix based on the commodity scoring matrix to obtain a prediction matrix in which the user is interested in the commodity, and recommending the user according to a result. According to the method, tThe autoencoders of different structures can be utilized in parallel, different feature information of the user and the commodity can be learned at the same time, more accurate high-level abstract features of the user and the commodity are extracted, prediction is conducted through the extracted abstract features, and the purpose of conducting more accurate recommendation for the user is achieved.

Description

technical field [0001] The invention relates to the field of personalized data recommendation research, in particular to a collaborative filtering recommendation method based on a parallel autoencoder. Background technique [0002] In the era of information explosion, recommender systems play an increasingly important role in solving the problem of information overload, and have been widely used in many online services such as e-commerce and social networks. The basic idea of ​​personalized recommendation is to use user-item interaction information to describe users' preferences for items, which we call collaborative filtering. In recent years, collaborative filtering has become one of the most widely used tools in recommender systems and has attracted extensive attention and research from multidisciplinary perspectives. Most traditional collaborative filtering methods use matrix decomposition, which decomposes the user's evaluation matrix of products into a user-based matr...

AI Technical Summary

Problems solved by technology

[0003] Although there have been some methods based on autoencoders that can better learn feature representations for recommendation systems and have achieved good results in personalized recommendations, there are still two major deficiencies that hinder the development of these methods. Further development

The first is the problem of the model structure of the autoencoder. Most of the existing methods rely on the same autoencoder structure, and it is difficult to capture the different characteristics of user-based and item-based data.

The second problem is that the user's rating matrix for the product is often very sparse, which leads to a significant decline in the performance of personalized recommendation. The current method often needs to import auxiliary information, but auxiliary information is sometimes difficult to obtain, and it is also very sparse

Images