Design And Application Of Functional Semantic Annotation Framework Of Scientific Papers

With the digitization of knowledge resources and the rapid development of semantic publishing,traditional methods of knowledge organization and information retrieval are increasingly unable to meet the needs of users in the new research environment for document retrieval,reading and utilization.The key issue is that the description of papers'content structure restricts the interpretation and mining of textual knowledge.Therefore,it is essential to identify and normatively describe the knowledge units and their semantic functions within scientific papers for promoting semantic publishing and improving knowledge services,which can help users locate intelligence units in scientific papers quickly and accurately as well as conduct strategic reading and comparative analysis.This study takes the semantic function of scientific papers' content as the research object,and mainly involves the following four research contents.Firstly,by literature research,this study systematically reviews the research status of domestic and international digital document structured standards,scientific paper content structure and semantic annotation.It is found that there are deficiencies in domestic research on deep semantic annotation of scientific papers' content structure and knowledge attributions.At the same time,the industry has a need for a generic,comprehensive and fine-grained semantic annotation framework.Secondly,based on the functional unit theory,this study designs a generic functional semantic annotation framework containing 28 functional units and 5 attributions from the perspective of semantic function.Thirdly,this study designs a semantic annotation experiment,which verifies that the annotation framework has good coverage in Chinese and English papers and can well describe the fine-grained knowledge content of scientific papers in depth.Fourthly,using the content analysis method,this study performs statistical analysis on the annotation data of 200 Chinese papers to explore the distribution rules of functional units and attributions within scientific papers,and to understand the organization of scientific paper knowledge units.The innovation points of this study are mainly reflected in the following three aspects.First,the functional units of scientific papers are reclassified and redefined from the perspective of semantic function,and a functional semantic annotation framework together with annotation experiments are designed.Thus,this study enriches the research perspective of scientific papers' discourse analysis and make up for the lack of domestic research on the scientific papers' semantic content structure.Second,compared to the existing semantic annotation framework of scientific papers,the functional semantic annotation framework proposed in this study can reveal the components with their semantic functions and attributions in a more detailed and comprehensive way.Third,this study deepens the understanding on scientific paper as a knowledge container and improves the semantic structure and function theory of scientific papers from the perspective of information science.