An article score prediction method based on matrix decomposition and neural collaborative filtering

Abstract

The invention discloses an article score prediction method based on matrix decomposition and neural collaborative filtering. The method comprises the following steps: (1) constructing a user-; An article scoring matrix; (2) performing matrix decomposition on the training matrix; (3) constructing a neural collaborative filtering network; (4) performing neural collaborative filtering on the trainingmatrix; (5) extracting user and article characteristics of the neural collaborative filtering network embedded layer; (6) constructing a nearest neighbor feature matrix; (7) generating a scoring training set and a scoring test set; (8) training the fully connected neural network; and (9) carrying out score prediction on the score test set. The method has the advantages that the user and article information is fully mined, the article score prediction accuracy is high, and the expandability is high.

Description

technical field [0001] The invention belongs to the field of computer technology, and further relates to an item rating prediction method based on matrix decomposition and neural collaborative filtering in the technical field of item rating prediction. The present invention can use a model-based method to perform training according to the historical rating information of the user on the item, and obtain the rating prediction of the user on the item that has not been evaluated. Background technique [0002] The recommendation system is an information filtering system. By analyzing the user's historical behavior data and the characteristics of each user, it discovers the user's interests and hobbies, and recommends items of interest to the user. Nowadays, there are many recommendation methods that exist. Among them, collaborative filtering is the most widely used and successful recommendation method. Among them, the model-based recommendation algorithm can effectively solve ...

AI Technical Summary

Problems solved by technology

The disadvantage of this method is that this method only uses the global feature information between users and items, which leads to insufficient information mining between users and items, and poor scoring prediction accuracy.

The disadvantage of this method is that this method only uses the user ID and item ID information, and adopts the one-hot encoding method, which leads to sparse network input when the number of users and items is large, poor score prediction accuracy, and poor scalability The problem

Images