Book recommendation method

Abstract

The invention relates to a book recommendation method. The book recommendation method comprises the following steps: step A. establishing a book label vector of each book in a library according to the author, three-level classification, four-level classification and universal label of the book; step B. establishing the text label vector of each user according to the authors, three-level classifications, four-level classifications, universal labels of all deeply read books in the book reading records of the user and corresponding weighted values of the authors, the three-level classifications, the four-level classifications, the universal labels of all deeply read books in the book reading records of the user respectively; and step C. calculating the label scores of all books in the library corresponding to each user according to the corresponding weighted values of the authors, the three-level classifications, the four-level classifications, the universal labels in the book label vectors of the books respectively in the text label vectors of the users. The book recommendation method belongs to the technical field of network application, books can be individually recommended for the users according to the contents of the books and the reading habits of the users, and books meeting the user habits are recommended for the users; and therefore, the degree of satisfaction is increased, and the user experience is improved.

Description

technical field [0001] The invention relates to a method for recommending books and belongs to the technical field of network applications. Background technique [0002] With the rapid development of mobile and Internet technologies, the digitization of books has become an inevitable trend. More and more book reading platforms have received high attention from users and achieved rapid development, and have become an important way for people to obtain information and knowledge. [0003] There are usually a large number of digital book resources on the book reading platform. How to effectively use these rich and valuable resources, so that users can find and make full use of them more quickly is very important. A very important function of the platform. [0004] At present, the methods widely used by book reading platforms in book recommendation include: [0005] 1. A recommendation method based on book classification or type. Through book classification, books of the same ...

AI Technical Summary

Problems solved by technology

[0005] 1. A recommendation method based on book classification or type. Through book classification, books of the same classification or the same type as the books that the user has read are recommended to the user. This method does not take into account the content topics and keywords of the books;

[0006] 2. The recommendation method based on the content of the book, using the TF-IDF (term frequency-inverse document frequency) formula to recommend based on the text content of the book, because many books on the book reading platform are fictional or original Books, the protagonists and events involved are fictitious, so the applicability of the recommendation method based on text content is not good, and it cannot recommend books that are more in line with user preferences to improve user satisfaction. poor experience

Images