Recommending method and recommending system of electronic book

Abstract

The invention provides a recommending method and a recommending system of an electronic book. According to the recommending method, data released on a website by a user is taken as the basis, the preference of the user to be analyzed for the electronic book can be judged by means of comprehensive survey and analysis of the visit frequency of the website by the user and relevant users and the data released on the website by the user. The recommending method and the recommending system of the electronic book have the advantages of being capable of overcoming the defects that the preference of the user can be deduced only by means of identity information registered by the user in the prior art, and solving the problems that the reading preference of new users of a reading platform can not be determined due to the fact that the reading historical data of the new users can not be obtained.

Description

technical field [0001] The invention relates to the technical field of text data analysis, and specifically designs an electronic book recommendation method and system based on user social network characteristics. Background technique [0002] There are many methods for recommending e-books to users in the prior art, and a brief description is given below: [0003] The first method is to infer the user's reading preference based on the user's previous reading history data, and then recommend corresponding e-books to the user according to the inferred reading preference. However, in this method, there are certain deficiencies in the inference of users' reading preferences, mainly because the categories of books provided by the reading platform are unevenly distributed, and there are many original books, which usually focus on romance, time travel, fantasy, etc. category; at the same time, since the main user groups of mobile phone reading users are students, migrant workers ...

AI Technical Summary

Problems solved by technology

However, this method has the following shortcomings: the existing system mainly reverses the user identity through the age, gender, occupational characteristics, product brand and mobile phone type of the registered user. The reason is that there are many inaccuracies in the data, which leads to a very low accuracy rate of user identity category determination, which in turn causes a large error in the inference of user preferences, which affects the effect of book recommendation

However, for users with fewer access to the reading platform system, there may not be enough historical data information; and the current system lacks a reliable basis for judging user identities due to various restrictions, and the accuracy of user category judgment is low

Therefore, there is a large error in the inference of user preferences.

Images