Cold start user-oriented recommended meta-learning method

Abstract

The invention discloses a cold start user-oriented recommended meta-learning method, and the method comprises the steps: carrying out the sampling of a data set in a mode of dynamic subgraph sampling,and enabling the data obtained through sampling to serve as training data, wherein the data set comprises interaction records between a plurality of users and different articles; training a collaborative filtering model by using the training data set, the training comprising an inner cycle and an outer cycle: in the inner cycle, performing recommendation prediction on each user, and updating model parameters of the users based on a prediction result; and in the outer loop, updating the overall model parameters by utilizing the model parameters of all the users. The method can be suitable forany differentiable collaborative filtering-based model, personalized recommendation can be better carried out for new users, and the model performance is improved.

Description

technical field [0001] The invention relates to the fields of recommendation systems and machine learning, in particular to a recommendation meta-learning method for cold-start users. Background technique [0002] With the rapid development in recent decades, the Internet has become an important way for people to obtain information and is widely used in various fields of social life. However, the contradiction between users' ability to process information and the ever-increasing amount of information on the Internet has become increasingly prominent. Personalized recommendation systems have achieved remarkable results in alleviating this information overload problem. One of the most representative models is the collaborative filtering model. [0003] As the most popular method in recommendation system, collaborative filtering is widely used in recommendation systems, but the collaborative filtering model has a serious cold start problem, that is, the interaction of new use...

AI Technical Summary

Problems solved by technology

However, auxiliary information is not always available due to reasons such as user privacy concerns, which makes collaborative filtering methods only rely on limited interactions. Besides, for real-world situations, it is necessary to provide accurate and fast dynamic recommendation

Images