Design And Implementation Of Knowledge Gragh System For Companies Listed On NEEQ

With the rapid development of internet technology in recent years,information interaction is gradually becoming faster and data resourses are more abundant.Faced with such big data,lots of time is spent to search the summary data which is most directly related with targeted entities.Meanwhile,to discover the relationship between entities and find out the similarities among them,one's time is also taken up to browse and compare data repeatedly.By utilizing the knowledge gragh,this process becomes more efficient and easier.In the area of NEEQ(National Equities Exchange And Quotations),compared with the main board market,since the amount of stocks is bigger,the sources of them are vaster and the information for those companies is more diverse,the process of searching information about them can be made more speedy and intuitive by the technology of knowledge gragh.The background of this subject is companies listed on NEEQ.The static information of companies is extracted by the technology of crawler;the extracted knowledge is tagged and classified;the knowledge is stored in the form of gragh.Also,reference third party knowledge base,different structure knowledge is cleaned and merged.Knowledge gragh is constructed by using gragh database.Updating module for companies is contianed by knowledge gragh system;login module,fixed model searching module and collecting module for visitors are also contained.Using cypher query language,visitors can get access to information of companies,senior executives,major shareholders and other financial information.The technology of NLP is used by the intelligent searching module of this system to recognize the named entities from the nature language of visitors.Then by the operation of tagging,user-defined searching are supported.Users are also supported to discover the relationship between companies and the searchi ng route can be assigned by visitors.The idea of event evolutionary gragh is consulted,connections between companies which have relationships on industry chain are built.Then combining with the searching history and collection list,the preference of visitors and recommend stocks which visitors may be interested in are speculated.Finally,through a series of functional and non-functional testings,the structure of knowledge gragh and system itself are proven to meets the requirements.Tests show that this system can be put into use.