Customized educational resource recommending method and apparatus

Abstract

The invention provides a customized educational resource recommending method and apparatus. The method comprises the following steps of: (1) determining a current initial recommendation vector of a user according to current initial preference information of the user about educational resources; (2) obtaining a current customized recommendation vector of the user according to a learning record and the current initial recommendation vector of the user; (3) calculating a correlation r between the current initial recommendation vector and the current customized recommendation vector; and (4) if the correlation r is greater than a designated threshold value s, performing filtering according to the current customized recommendation vector to obtain a recommended resource; or if the correlation r is less than or equal to a designated threshold value s, prompting the user to re-select an interest category and returning to step (1). According to the technical scheme, accurate customized educational resources can be recommended to different users, and therefore users can quickly find a content that matches interests of the users.

Description

technical field [0001] The invention relates to the field of data processing, in particular to a method and device for recommending personalized educational resources. Background technique [0002] In today's Internet era, the manufacture, reproduction and dissemination of information are becoming easier and easier, making all kinds of information show an explosive growth trend. But in the face of massive data, users have to waste a lot of time on information screening. To solve this problem, various personalized information recommendation methods have attracted more and more attention. [0003] The current information recommendation technology is mainly divided into two categories. One is the collaborative filtering technology for information recommendation based on the interests of neighbor users, that is, to find the neighbor users with the highest similarity with the user, and recommend the information that the neighbor users are interested in to the user. This recomm...

AI Technical Summary

Problems solved by technology

However, in the case of a large resource item cardinality, the computational complexity of this method is very high

Moreover, most of the resources in the medical education system are video-based, feature words are difficult to extract and the accuracy rate is low, and the obtained feature words are difficult to accurately describe the actual content of the resources. Therefore, the medical resources recommended in this way will There is a large deviation from the user's interest

[0005] In addition, the cores of these two types of technologies are based on historical data (using the user's past behavior records to estimate the user's interest or find neighbors). When the system uses this type of technology to recommend new users, since the system does not have the user Historical data, recommendation with a large blindness

Images