An education content personalized recommendation system based on a knowledge graph

Abstract

The invention discloses an education content personalized recommendation system based on a knowledge graph. The system comprises a search engine and a knowledge graph module, a user behavior recordingmodule and a user portrait module, in the application, the user behavior recording module records the operation of a user in a search engine; various behaviors of the user are transmitted to the userportrait module; the knowledge graph module establishes a user portrait and a knowledge point portrait; the knowledge graph module ensures that personalized recommended contents conform to personal preference and teaching logic of a user according to the knowledge point portrait module and the user portrait module, a semantic understanding module in the search engine technology sorts recommendation results based on semanteme, and finally the recommendation results better conform to the requirements of the user. The technical level of composition is simple, the problem of cold start is solved,contents meeting personal preferences of users and meeting teaching logic can be recommended, the sequence of the recommended contents is optimized, the sequence better meets the natural language expression of the users, and therefore the method is more suitable for being applied to education scenes.

Description

technical field [0001] The invention relates to the technical field of educational system applications, in particular to a knowledge map-based personalized recommendation system for educational content. Background technique [0002] A recommendation system is a system that recommends and presents user-related retrieval content to users, which is beneficial for users to obtain relevant information. In practical applications, when more and more types of content appear on the Internet, users have to face massive amounts of data and search results. At this time, users need to spend a lot of time searching from these massive results. This process will take a lot of time for users to retrieve the content they really need. In this case, a recommendation system is born. The task of the recommendation system is to connect users and information, saving users' time and improving user experience. Improve search efficiency and create value for users. In recent years, recommendation sys...

AI Technical Summary

Problems solved by technology

[0005] In order to overcome the problems in the prior art that a single recommendation system cannot deal with cold start, when the amount of data is large, the preprocessing efficiency is low, the hybrid model of hybrid recommendation is complex and costly, and the existing recommendation system does not have Considering the teaching logic, the user’s semantic understanding is not considered in the ranking of recommended content, which is not suitable for the application of educational scenarios. A new recommendation system model that is simple and does not have the problem of cold start, which solves the problem that a single model in the traditional recommendation algorithm cannot handle cold start, when the amou

Images