Social contact and consumption joint recommendation system and method, storage medium and computer equipment

Abstract

The invention belongs to the technical field of recommendation systems applied to deep learning, and discloses a social contact and consumption joint recommendation system and method, a storage mediumand computer equipment. Consumption preference features are extracted from a rating matrix R, social contact preference features are extracted from a social contact matrix S, and joint consumption characteristics and joint social characteristics are obtained through a reciprocal graph neural network, so that potential consumption and social possibility of a user are predicted. The social contactand consumption joint recommendation system comprises a self-attention space layer, a self-attention spectrum layer, a mutual benefit analysis layer and a prediction layer. According to the invention,the limitation of sparse matrix decomposition is broken through; the introduction of a self-attention model fully considers individual differences, so that the features extracted through the first two layers are more suitable for the real attributes of the user; the introduction of a mutual benefit mechanism gives full play to the interactivity of the original information of the two recommendation systems, and improves the recommendation accuracy, recall rate and NDCG index of the prediction layer.

Description

technical field [0001] The invention belongs to the technical field of recommendation systems for deep learning applications, and in particular relates to a social and consumption joint recommendation system, method, storage medium, and computer equipment. Background technique [0002] At present: Many sociological studies have shown that people's consumption behavior and social behavior are closely related. However, most current studies either only consider the influence of social interaction on consumption recommendation, or explore the recommendation of user relationships, or use one of the recommendations as an auxiliary reference for the other to improve accuracy. Only a few works treat social and consumption recommendation as a common problem, and these solutions are usually based on matrix factorization and neural networks, and then simply use cascade or weighted sum to aggregate user consumption and social behavior information, making the two The interaction relatio...

AI Technical Summary

Problems solved by technology

[0004] (1) The existing technology uses matrix decomposition technology to extract consumption preference features and social preference features. The recommendation system will inevitably encounter the problem of sparse data for new user recommendations and new data storage. However, matrix decomposition is not sensitive to sparse matrices. The consumption preference feature or social preference feature extracted by matrix decomposition cannot solve the recommendation problem when the data is sparse;

[0005] (2) The existing joint recommendation technology uses direct cascade or weighted cascade to summarize and use the features extracted from the above process. The research shows that the consumption behavior of users is easily influenced by their friends. It is also easier for people to establish social connections, and simply using cascades or weighted sums to aggregate features cannot fully reflect the interaction between the two;

[0006] (3) The existing joint recommendation technology ignores the role and influence of individual subjective cognition differences in the recommendation system. There must be differences in the influence of friends with different degrees of intimacy on their consumption behavior. will inevitably affect their social behavior

If individual differences cannot be taken into account, the accuracy of prediction results will inevitably be affected

Images