Design And Implementation Of Service Platform For Vertical Knowledge Graph Visualization

With the continuous development of Internet technology and Big Data,Internet information increased exponentially,so that more and more data acquired by people.Facing massive data,people with different roles in industry who are facing different situations have different application requirements.But the currently offered knowledge graph visualization service exists the following problems:1)the productions rarely integrate the knowledge graph and visualization technology,which causes low intuitionistic degree,low interaction and insufficient data utilization.Facing to different situation needs different tools for visualization,the interface data can be heterogenerous.2)can't make the most sufficient utilization of vertical knowledge graph.There is lack of a platform which can one-stop visualize the knowledge graph,support knowledge graph data's edition and statistic visualization and finally generate high interactive industry report.The low speed of generating industry report greatly affects the reports's timing and interactive.3)the vertical knowledge graph visualization platform has the character of huge amount of data.When requesting data directly from database,especially when there's a huge request amount,can cause enormous pressure to the database.the database's ability of processing data leads a great limit to system's throughput,even causes downtime.According to the above problems,this paper researched the relevant systems,designed and realized a vertical knowledge graph visualization service platform to offer services for industry users,like checking entities'relationship,exploring knowledge graph data analysis and creating industry report,etc.This design mainly includes the following works:(1)design and release the knowledge graph visualization tool Kgv.js.Encapsulate ECharts and D3.js secondary,so that it can help visualization developers' easier invoking,in order to improve developing ability.That can achieve the purpose of decoupling the visualization process from data and ensure the separation of the front-end's presentation and back-end's system business logic.(2)design and implement a vertical knowledge graph visualization service platform that supports dynamically changing data sources.This platform includes a knowledge graph data management system based on knowledge graph visualization,knowledge dervice system based on knowledge graph and visualization technology and industry report generating system based on knowledge driving.The platform includes multi-source data connection,knowledge graph visualization,check entity relationship,explore data analysis and generate industry report,etc.functions.(3)design and implement a response optimizing mechanism based on Redis.Accelerate the vertical knowledge graph visualization service platform's response and reduce database's pressure with the optimization of Redis' master-slave mode.Design cache's persistence and high availability to ensure the system's best using status.Verify the effectiveness through functional and performance tests.The vertical knowledge graph visualization service platform implemented by this paper supports multiple data source like dynamic changing data source,loading in data by interacting with web page and uploading data file,and then realized industry statistical data information visualization.By using hybrid database,realize the storage of industry data and industry report and the system's quick response cache.This platform provides different function modules for contenting knowledge graph data managing,data analyzing and information browsing.