Recommendation scoring method and device based on network structure and review text

Abstract

The embodiment of the invention provides a recommendation scoring method and device based on network structures and comment texts. The method comprises the steps that target users in multiple sample users are determined, and target commodities in multiple sample commodities are determined; a first type of characteristic matrixes for the target users are acquired, and a second type of characteristic matrixes for the target commodities are acquired; a third type of characteristic matrixes for the target user are acquired, and a fourth type of characteristic matrixes for the target commodities are acquired; the first type of characteristic matrixes and the third type of characteristic matrixes for the target users and the second type of characteristic matrixes and the fourth type of characteristic matrixes for the target commodities are input into recommendation network models to obtain predicted score values of the target users to the target commodities. Accordingly, interaction information of the users and commodities is fully considered, including comment text information and score information, and the commodity purchase expectations of the users can be more accurately predicted.

Description

technical field [0001] The present invention relates to the technical field of machine learning, in particular to a recommendation scoring method and device based on network structure and review text. Background technique [0002] With the development of e-commerce, Internet companies can provide a large number of products for users to choose from, and it is difficult for users to make choices in the face of a large number of products. Currently, recommender systems are mainly used to help users make choices. Among them, the recommendation system is a system obtained by using the method of deep learning to train the neural network model by using the user's purchase information of the product. [0003] When the recommendation system is used to recommend products for the user, for each candidate product, the user’s purchase information of different products and the purchase information of the candidate product are input into the recommendation system, and then the user’s rati...

AI Technical Summary

Problems solved by technology

[0005] However, the current recommendation system is only trained by using the user's purchase information of the product.

The user's purchase information of the product cannot fully represent the interaction information between the user and the product. Only the user's purchase information of the product is used to train the neural network model, and the obtained recommendation system does not accurately score the candidate products, which leads to ineffective Help users make choices about products

Images