NKOSis devoted to the discussion of the functional and data model for enabling knowledge organization systems/services(KOS),such as classification systems,thesauri,gazetteers,and ontologies,as networked interactive i...NKOSis devoted to the discussion of the functional and data model for enabling knowledge organization systems/services(KOS),such as classification systems,thesauri,gazetteers,and ontologies,as networked interactive information services to support the description and retrieval of diverse information resources through the Internet.These tools help to model the underlying semantic structure of a domain for purposes of information retrieval,knowledge discovery,language engineering,and the Semantic Web.NKOS workshops have been held since 1997 in conjunction with related professional and digital library meetings in the U.S.,Europe and Asia.The purpose of the workshops is to bring together KOS researchers and practitioners to share work on projects,good practices and innovations,and to discuss and critique this work.Workshops focus on topics including domain modeling,terminology development,validation,automated indexing,annotation and enrichment,and ethics.This JDIS special issue includes a selection of papers developed from presentations at the NKOS Workshop held at the Korean National Library in Seoul on September 26,2019 as part of the International Conference on Dublin Core and Metadata Applications 2019(DCMI-2019).In the spirit of the NKOS workshops,these papers include research in process,reports on projects,and“thought experiments”.展开更多
Automated metadata annotation is only as good as training dataset,or rules that are available for the domain.It's important to learn what type of data content a pre-trained machine learning algorithm has been trai...Automated metadata annotation is only as good as training dataset,or rules that are available for the domain.It's important to learn what type of data content a pre-trained machine learning algorithm has been trained on to understand its limitations and potential biases.Consider what type of content is readily available to train an algorithm-what's popular and what's available.However,scholarly and historical content is often not available in consumable,homogenized,and interoperable formats at the large volume that is required for machine learning.There are exceptions such as science and medicine,where large,well documented collections are available.This paper presents the current state of automated metadata annotation in cultural heritage and research data,discusses challenges identified from use cases,and proposes solutions.展开更多
文摘NKOSis devoted to the discussion of the functional and data model for enabling knowledge organization systems/services(KOS),such as classification systems,thesauri,gazetteers,and ontologies,as networked interactive information services to support the description and retrieval of diverse information resources through the Internet.These tools help to model the underlying semantic structure of a domain for purposes of information retrieval,knowledge discovery,language engineering,and the Semantic Web.NKOS workshops have been held since 1997 in conjunction with related professional and digital library meetings in the U.S.,Europe and Asia.The purpose of the workshops is to bring together KOS researchers and practitioners to share work on projects,good practices and innovations,and to discuss and critique this work.Workshops focus on topics including domain modeling,terminology development,validation,automated indexing,annotation and enrichment,and ethics.This JDIS special issue includes a selection of papers developed from presentations at the NKOS Workshop held at the Korean National Library in Seoul on September 26,2019 as part of the International Conference on Dublin Core and Metadata Applications 2019(DCMI-2019).In the spirit of the NKOS workshops,these papers include research in process,reports on projects,and“thought experiments”.
文摘Automated metadata annotation is only as good as training dataset,or rules that are available for the domain.It's important to learn what type of data content a pre-trained machine learning algorithm has been trained on to understand its limitations and potential biases.Consider what type of content is readily available to train an algorithm-what's popular and what's available.However,scholarly and historical content is often not available in consumable,homogenized,and interoperable formats at the large volume that is required for machine learning.There are exceptions such as science and medicine,where large,well documented collections are available.This paper presents the current state of automated metadata annotation in cultural heritage and research data,discusses challenges identified from use cases,and proposes solutions.
