Research On Fragmentation Knowledge Integration And Push Based On Knowledge Graph

The development of mobile Internet has promoted the rapid growth of multisource data,accelerated the spread of knowledge,and its mode of transmission is also diversified.Human beings have entered an era of fragmented knowledge,and people can use smart devices to publish or acquire meaningful pieces of knowledge on the Internet platform anytime and anywhere.One type of typical big data is user-generated content(UGC).Fragmented knowledge in the UGC is the product of a large number of human behaviors,thinking and interaction activities,which facilitates the acquisition and sharing of knowledge and helps to grasp the frontier information of the field in a timely manner.However,the “fragmentation” of knowledge leads to fragmentation of knowledge resources,rapid knowledge iteration and lack of authoritative verification,which is not conducive to learners to build a reasonable knowledge framework and deep learning.Therefore,how to deal with the fragmented knowledge of this new era to promote efficient learning is an urgent problem to be studied.This study takes the fragmentation knowledge in the UGC platform as the research object,and uses the knowledge graph technology and the idea of swarm intelligence to integrate the fragmentation knowledge of a certain field in the UGC platform.The paper focuses on the process of fragmentation knowledge integration based on knowledge graph to build a complete knowledge system.Firstly,through the mining and screening of fragmented knowledge,the relevant knowledge points are obtained,and then the individual knowledge is integrated into group knowledge.Then,a continuous loop feedback update model of “user individual knowledge set—fragmented knowledge point set—individual knowledge graph—group knowledge graph” is formed.Then,the knowledge is pushed on the basis of the obtained group knowledge graph,and the learning order of each knowledge point is sorted by proposing a central degree calculation method based on the importance and weight of the fragmented knowledge points.And thus the learning path recommendation of knowledge points and the related knowledge push method are obtained.Finally,the text of the topic related to the "blockchain" is captured from the Zhihu platform for case study.The group knowledge graph is constructed for fragmentation knowledge integration through the method proposed in this paper,and the results are stored and visualized by Neo4 j.The Cypher language is used to implement various knowledge services such as knowledge point query and relationship query,and the specific user is used as an example to carry out knowledge push.Through case studies,it is found that the user's individual knowledge graph only involves a certain module of the subject knowledge,and the integrated group knowledge graph has a relatively complete structure.Matching the user's own knowledge with the group knowledge system to push the knowledge to the user,which is beneficial for the user to learn from different perspectives,can effectively expand the knowledge,and accelerate the dissemination and utilization of knowledge.Compared with other methods,it shows that the central degree calculation method proposed in this paper is more accurate and the time complexity is lower.