Ontology Construction And Empirical Research Of Documentary Heritage Information Resources From The Perspective Of Digital Humanities

The development of digital protection of documentary heritage has generated a large number of real and valuable information resources of documentary heritage,which has important activation and application value.However,many information resources are still in the unexploited stage,and people have not consciously examined the utilization and dissemination of documentary heritage information resources from the perspective of digital humanities.Therefore,it is necessary to mine the specific content of documentary heritage information resources,and construct ontology by extracting entities and clarifying the relationship between entities,so as to guide the knowledge organization,inheritance and utilization of documentary heritage.Digital humanities and memory theory provide a theoretical basis for the ontology construction of documentary heritage information resources.Named entity recognition,keyword extraction,ontology construction and other methods provide technical support for this research.Documentary heritage network information resources provide the data source for this study.On this basis,the research on the ontology construction of documentary heritage information resources is carried out.This research mainly consists of two parts: First,we sorted out the related concepts,theories and technologies of the ontology construction of the documentary heritage information resources.On this basis,the process of the ontology construction of the documentary heritage information resources is elaborated.Second,taking the "Chinese Archival Document Heritage List" as an example,This paper conducts an empirical study on the ontology construction process of documentary heritage information resources from six aspects: acquisition of archival documentary heritage information resources,entity extraction,analysis of core concept objects,construction of application ontology,ontology evaluation and ontology construction mode based on crowdsourcing.The main innovations and features of this paper are:(1)This paper put forward the research idea of constructing the ontology of documentary heritage information resources by using the ontology as the core idea and using technologies such as named entity recognition and keyword extraction,and collect and compare existing documentary heritage information resources,define data sources as information resources such as archives official websites,newspapers,etc.,and perform operations such as collection,preprocessing,entity recognition,and keyword extraction of multi-source heterogeneous documentary heritage data.The study found that for highly specialized corpus with a large span of language time and space,the use of unsupervised algorithms in machine learning can better solve the problem of entity extraction of documentary heritage information resources,and based on this,a seven-step method is used to construct domain ontology.(2)Drawing on the principle of FAIR,it integrates the reusable CIDOC CRM model in the field of cultural heritage,the Dublin core element set,the Chinese archives classification method and thesaurus,the selection criteria and declaration form of the memory project,and the relation vocabulary REL,etc.By analyzing the seven core categories related to documentary heritage(documentary heritage item,item type,geographic location,time,physical characteristics,event,responsible person)and several subcategories,the domain ontology of documentary heritage information resources is formed.And through the application of digital humanities tools(Protégé)to establish and visualize the ontology of "Qing Dynasty ‘Yangshilei’Drawing Archives",the feasibility and rationality of the ontology and entity extraction method of documentary heritage information resources are verified.