| From the present situation,the main source of the study data for the listing Corporation annual report is the CSMAR Solution,or manual collection of something that is not yet integrated information.Research on text data is still not perfect,basically have to manually collect,such as R&D information collection.From the information perspective,the organization and access technology of annual report remain to be improved,which is mainly manifested in the lack of semantic management tool,relatively old technology of retrieval and the user's retrieval key word only regarded as a simple character.Therefore,it is an effective means of improving annual report service that to establish a semantic description mechanism of the annual report(such as R & D information),which can endue the computer with semantic understanding.Ontology is a concept modeling tool which can describe the information system on the semantic and knowledge layer.At present,along with the expansion of ontology application,it appeared a lot of domain ontology for a specific event,such as financial management domain ontology,telecom domain ontology,high-speed railway domain ontology,route planning domain ontology.This paper introduced ontology into R & D information and tried to build R & D information ontology,which can be a powerful semantic tool of the annual report information processing,organizing and using,reach a consensus between man and machine,and promote the man-machine communication.This paper analyzed the characteristics of R&D information in the annual report,tried to build R & D information ontology learning from the experience of information science field,combining with the achievements of ontology construction technology and the results of machine learning and natural language processing technology.Finally,based on the foundation of domain ontology,it designed and developed a body inspection information extraction model based on domain ontology. |
PDF Full Text Request