Book recommendation method, related device, equipment and storage medium

Abstract

The invention belongs to the field of e-commerce recommendation, and particularly relates to a book recommendation method, a related device, equipment and a storage medium. The method comprises the steps of acquiring and preprocessing user information data, book information data and user behavior data ; calculating to obtain user book interestingness, user activeness and book popularity; controlling user nodes to walk randomly based on user book interestingness, obtaining embedded vector representation of the user nodes , carrying out similarity calculation, and aligning the user nodes in different book city platforms ; performing data compensation by using the aligned scoring data; and taking the user activeness and the book popularity as related factors of a gray model, processing the compensated score data matrix by using the gray model to obtain a score prediction sequence of the user nodes, and recommending the top-ranked books. According to the method, hidden information between the users and the books can be mined, the problem of data sparsity can be effectively solved, information fusion is carried out on the aligned users, and relatively accurate recommendation data can be provided.

Description

technical field [0001] The invention belongs to the field of e-commerce recommendation and relates to user behavior analysis, in particular to a book recommendation method, related devices, equipment and storage media. Background technique [0002] Personalized recommendation is a hot topic in current research. Personalized recommendation is to analyze the user's interests based on the user's attribute information and user historical purchase behavior information. Realizing personalized recommendation can improve the transaction volume of e-commerce platforms. In the current e-commerce era, the recommendation system has become a more active and modern way of filtering information. With the research and development of recommendation technology, its application fields are also increasing. For example, news recommendation, business recommendation, entertainment recommendation, study recommendation, life recommendation, decision support, etc. The innovation, practicability, re...

AI Technical Summary

Problems solved by technology

Although the above method has certain accuracy, the real heterogeneous e-commerce bookstore network data usually faces a series of data problems such as data anonymization, data sparseness, data missing, uneven data distribution, data heterogeneity, and data difficulty in obtaining.

Images