Research On Intelligent Learning Resource Recommendation Based On Knowledge Graph Of Disciplinary Courses

With the progress and development of society,the country attaches increasing importance to education,which is the foundation of national development and the source of talent cultivation.Higher education,as a focus of education,is receiving more and more attention.Teaching and learning are the fundamental and core research contents of higher education.College students are the main group of learners in higher education and the main contributors to future society.Therefore,how to make it more convenient,easy,and accurate for college students to learn professional knowledge and thus contribute their professional skills to society has important theoretical value and practical significance.There are two problems in current college students’ learning.On the one hand,there is a large amount of disciplinary professional knowledge in higher education,and it is difficult to closely connect knowledge with each other due to the limitations of the curriculum,resulting in disciplinary barriers.How to break through the disciplinary barriers of professional courses,connect disciplinary professional knowledge closely,and help college students understand the relationship between different knowledge and quickly establish a knowledge structure system is a problem that currently exists.On the other hand,due to the impact of the COVID-19 pandemic in the past two years,online teaching has become the norm.Online teaching is difficult to supervise,and students find it challenging to obtain the content they need from a vast amount of information.How to help college students achieve smart learning under the current technology of artificial intelligence is another problem that currently exists.This thesis conducts an in-depth investigation into the above-mentioned problems and studies the knowledge of academic disciplines and intelligent learning using popular technologies such as knowledge graphs,and natural language processing in recent years.The main research contents are as follows:(1)To address the current problems in the knowledge graph of academic discipline courses,such as incomplete coverage of courses and knowledge points,unclear display of knowledge associations,and course barriers,a cognitive theory-based model for constructing knowledge graph of academic discipline courses is proposed.To validate the effectiveness of the model,a knowledge graph of academic discipline courses is constructed based on this model.(2)To solve the problems of over-reliance on the association relationship in a single knowledge graph-based recommendation model and the lack of interpretability in text-based recommendation models,a resource recommendation method that integrates semantic information with the knowledge graph is proposed.Based on the above knowledge graph of academic discipline courses,the knowledge graph is processed using two sub-models:a graph model and a text model.The advantages of the graph model and text model are combined by reasonably using the node association relationship and node text attributes of the knowledge graph.The effectiveness of this method is demonstrated through experiments.(3)To address the lack of complete intelligent service platforms for resource recommendation and knowledge question answering in higher education,this article conducts preliminary exploration of the intelligent learning service platform.Based on the knowledge graph of academic discipline courses,a prototype system for intelligent service platforms,such as learning resource recommendation,intelligent question answering,course search,and knowledge point search,is implemented.