Research On Scholarly Relations Revealing Based On Linked Data

In the era of big data,the number of academic information in scientific activities is in a wide range of change and has a quickly development.Scientific activities contain a lot of scholarly entities and revealing the relations between scholarly entities would be significant.It would be helpful to master the regularity in scientific development,to know current situation of the development of disciplines,promote knowledge discovery and to make scientific progress.With the development of the digital research environment,the scholarly relations revealing has a new connotation and denotation,which needs new tools and methods to reveal the relations.Linked data use structured metadata to organize information,aiming at revealing the semantic relations between things.Linked data lead a new way in the development of information organization and attract many researchers at home and abroad.This study uses the literature survey and logic method.Firstly,based on the theory of scientific communication,information integration of internet resources and linked data,analyze scholarly entities in scientific activities and summarize into 5 types of scholarly entities,which is illuminated by Popper’s theory of "three worlds".5 Scholarly entities are work,people,data,organization and condition.Then explain the meaning of relations between scholarly entities,draw the map of relations between scholarly entities and analyze the internal relations inside the scholarly entity.Secondly,analyze the pattern and path of the scholarly relations revealing.According to the general pattern of linked data application,deduct the pattern of scholarly relations revealing and show the path to conduct the system.The four key aspects to complete the whole system are linked data publishing,interlinking,data interacting and applications.Thirdly,analyze the structure and functions of the typical project VIVO of the Cornell University by using the knowledge of scholarly entities and the pattern of the scholarly relations revealing system.At present the relations between scholarly entities based on linked data is between primary level and intermediate level,the publishing of scholarly entities with linked data,interlinking semantic framework and applications need further development both in theory and practice.This study explore the scholarly relations revealing using linked data.The summarized five types of scholarly entities makes a contribution to the study of relations in scientific activities.Using linked data to reveal the relations between scholarly entities can be seen as a good exploration to the study of using linked data to organize knowledge and promote the knowledge discovery.