Research on Knowledge Management of Intangible Cultural Heritage Based on Linked Data

At present, the protection of intangible cultural heritage has received more andmore attention from all levels of society. Intangible cultural heritage is a treasure of national culture. It is an indispensable part of Chinese civilization, the crystallization of the wisdom of Chinese civilization, and represents the country’s soft power.­e eective organization and management of intangible cultural heritage knowledge is the premise and foundation for the protection, dissemination, and inheritance of intangible cultural heritage. Ontology and linked data technology provide a new method and realization path for the organization and management of intangible cultural heritage knowledge. In this paper, the intangible cultural heritage knowledge is organized reasonably semantically based on the method of linked data, and the purpose is to use the structure of linked data to express the resource data of dierent structures in a structured manner. ­is paper rst introduces the meaning and background of the research and analyzes the relevant research at home and abroad. Second, it introduces the related knowledge of linked data, analyzes and sorts out the elements and semantic relationship of knowledge in the eld of intangible cultural heritage, and designs and constructs the ontology model of intangible cultural heritage knowledge, Finally, based on linked data technology, the process of intangible cultural heritage knowledge organization and linked data set construction is studied, including key steps such as entity to RDF, entity association, linked data storage, and publication.­e application of linked data technology in the eld of intangible cultural heritage knowledge organization and management can promote the standardization and standardization of intangible cultural heritage knowledge management and is of great signicance to the protection and inheritance of my country’s intangible cultural heritage culture.


Introduction
With the development of society and the changes of the times, my country has made great achievements in both economic and technological development. In the long-term exchanges with countries around the world, the Chinese government has deeply realized the importance of cultural soft power in enhancing the overall national strength and has begun to strengthen the cultural identity of the Chinese people. As an important part of the excellent traditional culture of various ethnic groups, intangible cultural heritage shows the characteristics of the region, contains the wisdom of the group, and is a nonrenewable cultural wealth for every country. Intangible cultural heritage has various manifestations, covers a wide range of elds, and contains rich knowledge. Structural representation and semantic organization of knowledge in the eld of intangible cultural heritage is an inevitable requirement for the protection of intangible cultural heritage and knowledge dissemination in the era of knowledge. However, in the management of intangible cultural heritage knowledge, the China Intangible Cultural Heritage website and the intangible cultural heritage dissemination platforms of various provinces and cities only display the information of intangible cultural heritage items and representative inheritors in the form of a list, lacking knowledge of projects and inheritors, projects and regions, and so on. e lack of e ective organization and management of knowledge in the eld of intangible cultural heritage results in the fragmentation of information, which greatly limits the integrity and dissemination of knowledge. At the same time, there is a lack of professional knowledgebased linked data sets in the vertical field of intangible cultural heritage [1], and a mature intangible cultural heritage encyclopedic knowledge base has not yet been formed. Although some achievements have been made in the research of Chinese encyclopedia knowledge map in the general field, in the field of Chinese knowledge encyclopedia, there is a lack of large-scale encyclopedic knowledge data sets such as DBpedia and Freebase. Since the data in the field of intangible cultural heritage are still dominated by semistructured data and basic databases, there are many fragmented data, and there is a lack of high-quality open data sets for intangible cultural heritage; the single linear organization of intangible cultural heritage knowledge cannot reflect the diversity of intangible cultural heritage culture. characteristics, which cannot meet the needs of multidimensional disclosure and knowledge retrieval. is limits scientific research on intangible cultural heritage as well as the dissemination of intangible cultural heritage knowledge and the inheritance of intangible cultural heritage.
Foreign research on linked data is early. In January 2007, the Open Linked Data Project was launched, and the research on linked data began. In May 2010, the World Wide Web Consortium (W3C) established the Linked Data Incubation Group in the field of libraries. e main purpose is to help libraries use their data information to establish linked data, realize the effective organization and utilization of resources, and provide users with more comprehensive information. Intelligent retrieval services improve the relevance and interoperability of digital resources among libraries. In 2008, the Swedish National Library actively participated in the development of linked data, organized the Swedish Union Catalogue (UBRIS), and released it in the form of linked data. During the publishing process, the existing vocabulary was reused, which contains vocabularies of library domains such as DC, FOAF, and SKOS [2]. With the development of the Semantic Web, some domestic researchers have introduced ontology and linked data theories and methods into the field of intangible cultural heritage knowledge organization and knowledge management and have carried out a lot of research on ontology construction, resource integration, and resource organization. In terms of knowledge structure analysis and ontology construction research in the field of intangible cultural heritage, Lu et al. analyzed various elements of intangible cultural heritage from the perspective of systems theory and constructed a series of intangible cultural heritage items, people, things, events, documents, and so on. e ontology conceptual model is composed of core concepts and formulated metadata standards and specifications for intangible cultural heritage resources [3]. Huang et al. analyzed the difficulties in the construction of intangible cultural heritage knowledge ontology and took folk dance as an example to design an intangible cultural heritage knowledge ontology construction system for text and multimedia data [4]. Zhou et al. took the ontology method as a research perspective and designed a framework for the organization and retrieval of intangible cultural heritage information resources from the perspectives of ontology representation, semantic organization, and semantic retrieval and realized the ontology of intangible cultural heritage information resources by taking drama intangible cultural heritage projects as an example [5]. Ou and Tang proposed to divide natural language questions into simple sentences involving one data set and complex sentences involving multiple data sets in the study of automatic question-answering in the face of multiple RDF data sets and convert them into structured questions. e SPARQL query, and then integrate the answer, through experiments prove that the answer is more accurate [6]. Li and Zhao proposed an intangible cultural heritage archives resource development model based on linked data [7]. Dong used linked data technology to organize intangible cultural heritage knowledge semantically and realized the semantic disclosure and organization of intangible cultural heritage resources and their relationships [8].
e traditional knowledge management method of intangible cultural heritage projects is mainly classified and filed according to the subject, region, and level, unable to fully express and reveal the relationship between knowledge elements. With the deepening of knowledge organization and knowledge management research, the management of knowledge in the field of intangible cultural heritage has developed into new organizational methods such as topic maps, knowledge maps, and semantic organization, and more emphasis has been placed on the multidimensional disclosure of the relationship between domain knowledge [9]. In particular, the ontology and linked data technology in the Semantic Web technology stack provide new solutions for the semantic organization of knowledge in the field of intangible cultural heritage. Linked data is regarded as a lightweight implementation of the Semantic Web, which aggregates heterogeneous information resources in the form of linked open data, which can improve the visibility, sharing, and openness of resources and knowledge. e organization and management of intangible cultural heritage knowledge based on linked data are to express the knowledge elements and their attributes in the field of intangible cultural heritage in a structured and formalized manner under the guidance of a standardized and unified domain ontology model and to build the semantic relationship between the knowledge elements, then achieve the purpose of knowledge semantics and order, and provide open data acquisition and knowledge services.
Based on this background, because of the effective management and knowledge association of intangible cultural heritage knowledge, this paper proposes a method system for intangible cultural heritage knowledge organization and intangible cultural heritage knowledge association database construction based on linked data, including the design of intangible cultural heritage knowledge ontology model to the construction of linked data, the whole process of storage and publishing; research the knowledge association between the internally linked data sets in the field of intangible cultural heritage and the external knowledge databases of DBpedia and Geo Names; and establish an intangible cultural heritage knowledge related database with rich semantic relationships and knowledge interconnection and sharing, providing domain knowledge. Orderly organization and visualization of knowledge: It is helpful for the excavation and discovery of intangible cultural heritage knowledge as well as the international dissemination of my country's intangible cultural heritage culture.

Concepts and Principles of Linked Data.
e concept of linked data comes from the "Father of the Internet," Tim Bemers Lee, who proposed the Web of Data when he analyzed the development and evolution of the Web in "Design Issues for the World Wide Web" published in 2006. In 2007, linked data was formally proposed in the Linked Open Data Project submitted by Chris Bizer and Richard Cyganiak to W3C. Different scholars have put forward their views on what is linked data. Tim Bemers Lee believes that linked data defines a URI specification that enables people to directly obtain digital information resources through HTTP and URI mechanisms. Wikipedia recommends linked data as a best practice for publishing, sharing, and linking all kinds of data, information, and knowledge on the Semantic Web using URI and RDF. It can be seen from this that linked data is a description specification, which describes data in the triple RDF format of "subject-predicate-object." URI determines the uniqueness and relevancy of resources in the data. Link association builds a new data network to realize intelligent applications. Because linked data solves the problem of linking distributed and heterogeneous data in the form of semantic association, it is recommended by W3C as the "Best Practice of the Semantic Web." Linked data has realized the extensive linking of data in different fields in the network and promoted the sharing and widespread use of various types of data. Users can quickly discover a lot of relevant data through data links. Figure 1 shows the situation of the Linking Open Data Cloud Diagram (LOD) as of March 2019. e data set already contains 1239 data sets and 16,147 links. Each node in the figure represents the data published by each data set. e lines between them represent the relationship between the data sets.
Tim Bemers Lee pointed out that the principles of linked data [10] mainly include four aspects: the first principle is to use URI as the name of the resource to identify things. All resources on the World Wide Web are assigned a unique URI as the resource identifier, which stipulates that everything in the World Wide Web uses a URI as a unique identifier, which is the primary condition of the Semantic Web. Using URI to standardize the name of things can avoid ambiguity and confusion; the second principle is to use HTTP URI to locate and find corresponding resources, that is, all URIs can be accessed and retrieved. It is stipulated to use HTTP URI as the link specification, which is conducive to discovering the required information resources according to the data link; the third principle is that when the URI of the resource is accessed, the information related to the resource is provided, and unified standard to represent this information, people can view data information and other classes and attributes through URI; the fourth principle is to provide relevant URI as much as possible to help people discover and obtain more information with potential use value and promote the sharing and utilization of information resources, truly realize the globalization of information.
ese four principles are important indicators of the semantic association of digital information resources on the World Wide Web and are where linked data fully reflects its ability to solve the problem of distributed and heterogeneous digital information resources on the Semantic Web. If these four principles are not fully satisfied, they may also play a certain role in data association, but it will greatly reduce the potential of data association and reduce the value-added benefits of data aggregation.

Related Technologies of Linked Data.
Linked data can reflect the objective entities such as data and concepts and the relationship between them, which is an important factor for it as an effective way to achieve semantic relationships on the Semantic Web.
A URI is a string used to identify a resource, allowing users to interact with any resource through a specific protocol. Linked data is based on Web technology, mainly involving HTTP, HTML, and URI. HTTP is a hypertext transfer protocol, which is a standard for client and server requests and responses. HTML is a hypertext markup language, which is stored on the server. Web page files: HTML uses markup symbols to identify and standardize various parts of web page content; URI is a uniform resource locator, which realizes the unified positioning of network resources at the access address; and associated data further define and expand these three technologies, using URI at the same time. To solve the problem of naming and positioning [11], at the same time, the biggest feature of URI is that the identification is stable, the resource path is regarded as a part of the resource name, and it is not allowed to change at will so that the resource does not have information dislocation due to the change of the attribute, which is helpful to Stable Links to Achieve Semantic Associations in Dynamically Changing Network Environments. URI is the most critical technology for linked data. Using URI, anything in the World Wide Web can be given a unique identification name, and "anything" is different, mainly divided into information resources and noninformation resources. Information resources refer to the information resources on the Internet. Digital resources that can be found, such as pictures, videos, web pages, audio, and so on, exist in various physical objects outside the network, including nature, human society, and human meaning, such as mountains, rivers, people, and so on. e fourth principle of linked data refers to the use of relevant URI as much as possible to realize the RDF description of resources. It can be seen that URI can also be used for URI reuse while ensuring the uniqueness of resource identifiers. at is if a resource has been identified by authoritative URI and other data creators, the URI identifier of the resource can be used to ensure the uniqueness of the resource identifier. At present, institutions such as the Library of Congress and W3C provide terminology services and representative use cases based on linked data. At the same time, authoritative vocabulary sets such as DBpedia and FOAF also provide conceptual terms and associations for linked data. erefore, the use of URI can effectively and uniformly identify resources, which not only has a clear indication function for resources but also ensures effective management of resources, which is conducive to identifying distributed and heterogeneous related resources in data association and realizing the spanning of linked data. Purpose of development: URI simply creates a unique identification name and ensures the stability of identification, thus enabling information resources to form a stable associative aggregation in the World Wide Web [12].
RDF is a resource description framework, which provides a way to flexibly describe diverse network resources. It is a markup language used to describe resources on the Web. It is a basic semantic format that can be understood by machines. RDF data can realize data are exchanged between computers with different types of operating systems and application languages so that each data in the network can be shared and utilized to the greatest extent. RDF is to describe resources by using the triple pattern of the "subject-predicate-object" structure under the condition of strictly following the network structure, in which the resource is the subject, the attribute of the resource is the predicate, and the attribute value is the object. e resource is defined by URI, and the predicate represents the relationship between subject and object, which is also represented by URI. Data can be clearly and accurately described through RDF. e resources in RDF contain links identified by URI, which can link to other related resources. At the same time, these links contain semantic relationships, indicating the relationship between resources. Data can be composed of many pieces of RDF, described by a resource identified by URI. e object of RDF can be a numerical attribute or an object attribute, that is, the resource identified by URI; the predicate between the subject and the object indicates the relationship between subject and object and can also be mark with URI.
ese URIs can come from normative vocabularies, such as FOAF, DC, SKOS, and so on.
SPARQL is a query language and data acquisition protocol especially developed for RDF. Using SPARQL, all information resources represented by RDF can be retrieved and queried on the network. e ultimate goal of SPARQL is to retrieve the Semantic Web in the same way that SQL retrieves relational databases. SPARQL can query RDF data between different data sources. e data source can be the data in the RDF format of the entity, or it can be the data in the virtual RDF format through the middleware. e SPARQL language can query based on graphs, and its query results can also be returned in the form of RDF graphs or data sets. An SPARQL query statement can be composed of five parts: statement, query form and result set, data set, graph schema, and result decoration. e relevant query statements are shown in Table 1.

Technical Feasibility Analysis.
e core of linked data is to name data resources with URIs under the premise of following the HTTP protocol and organize and standardize data resources into the RDF format of the "subject-predicate-object" relationship to reflect the relationship between attributes and attribute values, and finally use HTTP URI to locate and query each data resource. e application development of linked data has promoted the generation of tools for constructing and publishing linked data, which mainly include three types.

Relational Database RDF Conversion Tool.
e database is the main place for storing resources. e database is a data collection that organizes, stores, and manages data according to a certain data structure. e independence of data is conducive to the centralized control of data, provides convenience for massive resource storage, and promotes the combination of linked data and resources. Databases can be divided into hierarchical databases, network databases, and relational databases. e relational database is based on a relational model that reflects the relationship between entities and entities, which meets the requirements of relational data mining for associations between data resources. At present, the commonly used relational database management system is MySQL, which saves data in different data tables instead of storing all data in one database. Each data table has a primary key attribute that is not a null value. Both the key attribute and the primary key have a certain limited association relationship. Each data table associates the data attribute values in different data tables through the invocation of the primary key and the foreign key. Using MySQL can easily call the data of each database; even if each database is updated, it will not affect the values of other databases. Based on determining the relationship between data resources, data resources need to be processed, including the creation and publication of RDF [13]. D2R is a tool for publishing relational databases into linked data. It mainly includes three parts: D2RQ Mapping language, D2RQ Engine, and D2R Server. e main function of the D2RQ Mapping language is to convert relational data into mapping rules in RDF format. D2RQ Engine uses the D2RQ mapping file to convert the data in the relational database into RDF format. It does not convert the relational database into real RDF data but maps it into a virtual RDF format. is file is used to access the relational database. When generating data, the query language SPARQL of RDF data is converted into SQL language, and the SQL query result is converted into RDF format or SPARQL query result. D2R Server provides a query access interface for RDF data, which is convenient for SPARQL clients and traditional HTML browsers. e operation flow of D2R is shown in Figure 2.

Linked Data Tools at Directly Generate RDF Data.
Such tools mainly include Virtuoso Universal Server and Sparq Plug. Virtuoso Universal Server is a commercial-grade linked data tool that enables XML, Web server, and network. e ideal carrier between them can use the SPARQL side to convert the data into RDF format. Sparq Plug uses SPARQL query language and HTML DOM to convert traditional HTML data into RDF form.

Other Linked Data Tools
at Publish RDF Data. Pubby is a front-end linked data for SPARQL that provides a linked data interface for resources in RDF format. Pubby is used to connect the linked data interface and the SPARQL side. It can provide a linked data interface to connect to a local or remote SPARQL protocol server. At the same time, it can also provide simple HTML calls to the database to display available resources. Its working principle is shown in Figure 3.
In RDF, resources are identified by URI, but in most SPARQL data sets, URI cannot be directly referenced, that is, they cannot be accessed directly in a web browser. At this time, by establishing a Pubby server on the SPARQL side, the method of URI mapping is adopted to obtain the original URI information by connecting to the SPARQL terminal and returning the result to the client so that the user can obtain the URI information that can be used directly.

Design and Construction of Intangible Cultural Heritage
Knowledge Ontology Model. Ontology is regarded as a clear formal specification of shared conceptual models. In the field of information science and computer, ontology can be regarded as a model, which is a formal expression of objectively existing objects or concepts, their attributes, and related relationships. For the effective organization and management of knowledge in the field of intangible cultural heritage, it is first necessary to clarify the structure of intangible cultural heritage knowledge, its constituent elements, and internal relationships. en, based on referring to the international general ontology model, a knowledge ontology model in the field of intangible cultural heritage is established according to the knowledge characteristics in the field of intangible cultural heritage. Intangible cultural heritage knowledge ontology is a formal conceptual model formed by a highly abstract summary of the intangible cultural heritage connotation and its constituent elements [14]. Based on ontology theory and existing research results, according to the ideas and steps of domain ontology design and construction, and by analyzing the knowledge structure and constituent elements of intangible cultural heritage projects, this paper constructs an intangible cultural heritage Mobile Information Systems knowledge ontology model. Description and knowledge association provide a uni ed and standardized knowledge representation model and data model. At present, for the protection and inheritance of intangible cultural heritage culture in my country, a four-level intangible cultural heritage protection system of "national-province-city-county" has been formulated. erefore, the basic knowledge about intangible cultural heritage can be regarded as composed of intangible cultural heritage items, inheritors, relevant institutions, project types, geographical locations, and other elements. [15]. Intangible cultural heritage project (ICH Project) is an abstraction of intangible cultural heritage projects, and its example refers to each speci c project in the four-level list of intangible cultural heritage protection established by my country. An instance of an intangible cultural heritage item is a composite object, which not only has its connotative attributes but also includes related entities such as inheritors and regions.
Based on analyzing the basic structure and relationship of intangible cultural heritage knowledge, the ontology model of intangible cultural heritage knowledge designed and constructed in this paper is shown in Figure 4. e ontology model reference draws on the ontology models such as CIDOC CRM, FOAF, Geo Names, person relationship vocabulary (Relationship), and the Dublin Core (DC) metadata standard [16]. According to the core elements of knowledge in the eld of intangible cultural heritage, the knowledge ontology in the eld of intangible cultural heritage is abstracted into four core categories: intangible cultural heritage project (ICH project), representative inheritor (Person), geographic location (Place), and project type (Category). Each core class de nes corresponding data properties, and the relationship between entities is described and revealed through Object Properties.   Item type category (Category) is used to construct the classi cation system of intangible cultural heritage items; by using the object attributes (skos: broader, skos : narrowe, etc.) that represent the relationship between the upper and lower levels of the concept, a multiperspective, multilevel, scalable intangible cultural heritage item classi cation system. Representative inheritors of intangible cultural heritage are an important part of intangible cultural heritage. e de nition of the agent class (Agent) in the ontology model de ned in this paper reuses the FOAF ontology model, and the agent class can be divided into two subcategories: "inheritance individual" and "organization and institution"; "individual" mainly refers to certi ed national. e representative inheritors at the provincial and municipal levels are individuals, and "organizations and institutions" refer to the declaration unit of some intangible cultural heritage items. For example, the declaration unit of the 24 solar terms in the lunar calendar is the China Agricultural Museum. For the representative inheritor of an intangible cultural heritage item, its attributes include the inheritor's number, name, gender, title, ethnicity, date of birth, place of origin, and other basic information, as well as the skills and skills they have acquired. In addition, the attributes in the relationship vocabulary (Relationship) are reused to better express the intricate inheritance relationship and inheritance lineage between inheritors. e geographic location class is de ned in the ontology model. On the one hand, it expresses the geographical space of the distribution and circulation of intangible cultural heritage items, and on the other hand, it records information such as the place of residence and place of origin of the inheritor. And the geographic location class is given the data attributes of the administrative level of provinces, cities, counties, villages, and towns, which corresponds to the administrative divisions of our country. At the same time, associate each geographic location instance with the geographic database Geo Names to obtain information such as geographic location introduction, latitude, and longitude. Combined with geographic information system (GIS) technology, not only the spatial distribution of intangible cultural heritage can be expressed in the form of intuitive visualization of the spatial distribution of maps but also the hidden information of the spatial dimension in intangible cultural heritage can be excavated from a deep level. Using GIS spatial analysis techniques to analyze the spatial structure, evolution, and characteristics of intangible cultural heritage is of great signi cance for understanding the connotation of intangible cultural heritage [17]. e relationship between entities connects independent knowledge elements to form an intangible cultural heritage knowledge network, changing the single-clue model of traditional knowledge organization. In addition, based on these associations, by de ning rules and relational reasoning, invisible knowledge can be inferred and discovered. Focusing on the organization and management of knowledge related to intangible cultural heritage projects, this article refers to the international common ontology models such as CIDOC CRM, FOAF, Relationship, and so on, and carries out the reuse and custom extension of an ontology according to the characteristics of intangible cultural heritage knowledge. e intangible cultural heritage knowledge ontology model is oriented to the organization and management of intangible cultural heritage knowledge, providing a formal representation of knowledge in the eld of intangible cultural heritage, and meeting the knowledge management needs in the process of declaration and certi cation of intangible cultural heritage projects. e main function of building an ontology model of intangible cultural heritage is to standardize the description and formal expression of intangible cultural heritage knowledge, while the semantic transformation of data and the structured representation and storage of knowledge need to be implemented with the help of linked data technology.

Construction of Intangible Cultural Heritage Knowledge Association Data
Set. e construction of the intangible cultural heritage knowledge association data set is a huge and systematic project. First, clarify the research scope and sort out the knowledge objects in the field of intangible cultural heritage; second, extract the domain entities, clarify the various attributes of the entities, build the domain ontology model, and form thesaurus and glossary; third, according to the ontology model, the entities are RDF attribute description, establish entity links, including entity links between internal entities and external open data; finally, select an appropriate data storage and publishing platform to provide access and data interfaces for humans and machines. Following the basic principles of linked data, the construction of a knowledge-linked database in the field of intangible cultural heritage can be divided into five key steps, namely data modeling, entity naming, entity RDFization, entity association, and entity publishing [18]. Among them, the data modeling process is the construction process of the knowledge ontology model in the field of intangible cultural heritage. is paper takes the Hubei Province intangible cultural heritage project as an example to explore the construction steps and specific implementation methods of the intangible cultural heritage knowledge association data set. e technical framework of the intangible cultural heritage knowledge association data set construction is shown in Figure 5.

Entity to RDF.
Due to a large number of intangible cultural heritage projects in my country, and the dynamic changes in the identification of intangible cultural heritage projects and inheritors, the construction process of the entire knowledge base in the field of intangible cultural heritage needs to be carried out in layers and batches. ere are two main sources of data. One is the data of intangible cultural heritage project declaration and representative inheritor certification application. e collected data are preprocessed such as data cleaning and stored in the database. Many traditional intangible cultural heritage information systems mainly use relational databases for data storage.
is paper chooses to perform data semantic mapping based on D2RQ to convert the content in relational databases into linked data. Semantic mapping of data is to convert the two-dimensional table structure into associated data that is better at dealing with complex relationships and richer in semantic information; it specifically includes data table-to-class mapping, column-to-attribute mapping in data tables, and table-to-association mapping. In the mapping language, d2rq: Class Map is used to define the classes of the ontology model, which corresponds to the mapping of the data table, and d2rq: Property Bridge is used to define the attributes in the ontology model, which corresponds to the mapping of the columns and relational tables in the data table. e mapping of data in the relational database to RDF should follow the classes and attributes defined in the ontology model [19]. e five main data tables in the relational database are mapped to four core classes and their attributes, intangible cultural heritage item, inheritor, classification, and geographic location. e relational table is mapped to the "has Inheritor" object attribute, as shown in Figure 6. e fields of each data table are mapped to corresponding properties.
Finally, according to the data mapping file, use the dump-rdf tool of the D2RQ platform to convert the data in the relational database to generate an RDF/XML format file for use by other databases or third-party applications. Although D2RQ can also publish linked data, it is not flexible enough to update and manage data and has limited support for complex relationships and massive data. erefore, this paper uses D2RQ to semantically transform the data into RDF/XML format files. en, the data are stored in a special Triplestore database, and the server is configured to realize the associated publishing of the data and the data open interface. Table 2 corresponds to the mapping framework, which is the core statement for semantically mapping the item table, the inheritor, and the inheritance relationship table.

Entity Association.
Entity association is based on the RDF description of entities, uses RDF links to describe the semantic relationship between different entity objects, and establishes associations with external data as much as possible to build a linked data network. Linking data to other open RDF data sets and vocabularies is a key step in enriching the semantics of linked data. Entity linking should more semantically link internal data with external open data sets and realize knowledge discovery through knowledge aggregation across domains, disciplines, and databases. In the Semantic Web environment, with the help of the standardization and open interconnectivity of linked data and the integration of multiple knowledge bases, the richness and breadth of knowledge in the field of intangible cultural heritage can be improved, providing scientific research and knowledge dissemination of intangible cultural heritage culture. Data Foundation and Knowledge Services [20]: To enrich the knowledge in the field of intangible cultural heritage, this paper chooses to match and associate data with DBpedia and Geo Names linked data projects.
rough the association with DBpedia and its data source Wikipedia, the Chinese and English entries corresponding to the intangible cultural heritage items are obtained; through the association with the global geographic database (Geo Names), more information about the regions involved in the intangible cultural heritage field can be obtained. Use the OWL built-in attribute owl: same As to associate the internal knowledge entity with the entity in the external data set, indicating that the two linked entity objects are the same thing.
e DBpedia project is a large-scale knowledge data set based on linked data and established by extracting data from Wikipedia. It is the core of the linked open data cloud graph.
e DBpedia data set contains a large amount of information about my country's intangible cultural heritage described in Chinese and English, and entity association with it can enrich the knowledge in the eld of intangible cultural heritage and improve the visibility and generality of intangible cultural heritage knowledge [21]. is paper adopts a combination of automatic retrieval and manual inspection. First, the resource items related to intangible cultural heritage items in DBpedia are retrieved online through SPARQL language, and then the retrieval results are screened and checked by manual inspection, and nally the retrieved resources are retrieved. e URI is associated with the internal intangible cultural heritage item entity through the owl: same As attribute. In the DBpedia ontology model, the db : abstract attribute represents the abstract of the resource, and the foaf: is Primary Topic of attribute links the Wikipedia page corresponding to the resource. e SPARQL Endpoint site of DBpedia is called online, and the related concepts of the intangible cultural heritage items are retrieved by constructing SPARQL sentences to retrieve the resource items containing the keyword. Due to the imperfections of DBpedia, for example, some intangible cultural heritage items lack entries, the Chinese information provided by DBpedia is incomplete, the resource entries related to intangible cultural heritage in Wikipedia are not included in DBpedia, and because the titles of intangible cultural heritage items are not uniform, through the above methods, only some entities are correctly matched, and some are not fully matched. erefore, based on automatic retrieval and manual inspection, this paper associates as many internal intangible cultural heritage project entities with DBpediarelated resources as possible. In the end, about 1/3 of the intangible cultural heritage items are physically associated with DBpedia or Wikipedia, which also highlights the necessity of establishing an encyclopedic database of intangible cultural heritage in my country.
e Geo Names geographic database contains more than 10 million geographic names around the world and provides information such as geographic name alternatives, latitude and longitude, population, and Wikipedia. It adopts the principle of linked data to organize, de nes a unique resource URI for each geographic name, and publishes the geospatial semantic information to the Internet. To obtain more information about geographic location, enrich domain knowledge, and provide a data basis for spatial analysis of intangible cultural heritage based on geographic information, this paper combines entities related to geographic location, such as intangible cultural heritage project application areas and inheritor's residence, with Geo Names database for the association. e Geo Names data can be obtained by calling its o cial API or by using the SPARQL endpoint provided by a third party [22]. To ensure the consistency with the DBpedia data association, this paper uses the SPARQL endpoint of the Factforge website to obtain Geo Names data. e core SPARQL query statements related to data matching and association are shown in Table 3. e query sentence uses "Dark Biography" as the keyword to perform a full-text search on the Chinese tags of DBpedia and obtain the abstract information of the resource and the corresponding Wikipedia page. It is linked to the intangible cultural heritage item entities in the internal data set.  Figure 5: e technical framework for the construction of the intangible cultural heritage knowledge association data set.
Retrieve the resource entry corresponding to "Enshi Prefecture" in the Geo Names database and limit the search scope to China (coded as CN) and the resource type to country and region (coded as A). e retrieval result will return the URI of the geolocation object, <http://sws. geonames.org/1811624/>, and you can further obtain information such as the latitude and longitude of the geographic location and the Wikipedia link. In the case of the same geographic names, disambiguation can be achieved by judging the municipal and provincial geographic names (parent ADM2, parent ADM1). In the above way, the association to the geographic location entity achieves a 100% match. Finally, based on constructing internal intangible cultural heritage domain knowledge-linked data, multiple categories of entities such as intangible cultural heritage items, people, and geographic locations are associated with external databases such as DBpedia and Geo Names. e number of entities and triples in the nally constructed intangible cultural heritage domain knowledge association database are shown in Table 4. Among them, there are more than 10,050 triples, involving multiple categories of objects such as intangible cultural heritage items, inheritors, institutions, geography, and types. ere are 505 entities associated with DBpedia and 295 entities associated with Geo Names. By linking DBpedia, it can be more convenient to associate with Wikipedia, WIKIDATA, YAGO, and other resources. Moreover, the constructed intangible cultural heritage knowledge-linked data are completely open according to the W3C standard, and data services can be obtained through online retrieval, SPARQL Endpoint, and other data cabling methods.

Storage and Release of Intangible Cultural Heritage
Knowledge-Related Data. After converting all kinds of data and knowledge in the eld of intangible cultural heritage into the form of linked data, it needs to be persistently stored and published. e storage and publishing of linked data directly a ect the sharing and reuse of data [23]. ere are many ways to store and publish linked data.
is article adopts the native method for storage, con gures the server for linked data publishing, and provides open data services and knowledge services. Open Link Virtuoso is used as the storage and management database of RDF data, and Lod View is used for users to provide data browsing of the intangible cultural heritage knowledge base and use Lod Live to provide a visual representation of the intangible cultural heritage knowledge association graph. e storage scheme of linked data is roughly divided into relational data-based storage, NoSQL database storage, and Tri-plestore database storage. Among them, the Triplestore database is specially developed for the characteristics of the RDF data structure and has e cient data storage, query, and reasoning mechanisms; at the same time, because it adopts a uni ed data model, it can achieve e cient interaction between data. Considering the expansion of knowledge in the eld of intangible cultural heritage and the growth of data in the future, this paper chooses Open Link Virtuoso, which is  widely used, like the database management software. e virtuoso database is a cross-platform scalable high-performance database management software that provides SQL, XML, RDF database management functions and supports the storage and management of billions of scale triples. e virtuoso database provides multiple mechanisms such as WEB pages or ISQL commands for data import. e original data related to intangible cultural heritage is semantically described or semantically mapped to generate RDF format data and then imported into the Virtuoso data storage, and the IRI (Internationalized Resource Identifiers) of the named graph to which the data belong is specified. e intangible cultural heritage domain knowledge RDF data constructed above needs to be published through linked data to realize the utilization and sharing of data; the commonly used publishing methods of linked data are based on static RDF/XML files, based on relational data, and based on RDF data repository, using RDFa formula and other ways. Based on the Virtuoso database platform, this paper follows W3C's four principles of linked data publishing, configures a linked data publishing server, and provides services such as RDF data management, linked data browsing, and SPARQL endpoints and content negotiation. To fully and comprehensively display the attributes and relationships of each entity in the intangible cultural heritage-related data set, the Lod View tool is used to provide users with a browse of the intangible cultural heritage-related data. Lod View is a Web application developed based on Jena and Spring framework. It supports the parsing of International Resource Identifiers (IRI) conforming to the W3C standard and is a tool for converting RDF data format to HTML; configuring Lod View's SPARQL site and multimedia display and Attributes such as latitude and longitude; and returning the correct RDF data and web page description according to the content negotiation mechanism.  d2rq: Belongs to class map map: project; d2rq: property ich: has inheritor; d2rq: Refers to class map map: inheritor; d2rq: join "pro_inh．Inheritorid �> inheritor．urlcode"; d2rq: join "pro_inh．Projectid �> project．urlcode"; ． 8890/resource/ichproject/359 represents the instance of "Intangible Cultural Heritage Project Class" "Dark Biography"; the page that uses the LodView tool to display the details of this instance is shown in Figure 7.
is page shows all the properties of the "Dark Biography" instance of the intangible cultural heritage item. e upper part of the page is a multimedia display part, which visually displays multimedia resources such as pictures and videos related to it. e page lists the representative inheritors, types, distribution areas and other related objects of the project. For example, link the representative inheritor of the project through the ich: has inheritor attribute and click the hyperlink to jump to the detailed page of the inheritor object. When the user accesses, it provides an intuitive HTML page; when the application accesses, it returns the data in the corresponding format such as RDF/XML, RDF/ Turtle, JSON, and so on, according to different content requests. In the intangible cultural heritage knowledge association data set, geographic location objects are associated with Geo Names. When the user accesses the information page of the instance of the geographic location class "Shennongjia Forest Area," the location in the map is displayed according to the latitude and longitude information in the form of an online map. e middle of the page displays the specific attributes of the geographic location class and is associated with DBpedia, Geo Names and Wikipedia; at the bottom of the page, through the inversion of the ich : has Place attribute, the intangible cultural heritage items, and inheritor objects owned in the region are retrieved in reverse.

Conclusions
e development of the Semantic Web and linked data has provided new ideas and methods for the organization and sharing of knowledge in the field of intangible cultural heritage and has changed the representation and expression of intangible cultural heritage knowledge. To realize the effective  e intangible cultural heritage contains a wealth of knowledge. is article only studies and sorts out the basic knowledge elements such as intangible cultural heritage items, inheritors, regions, and inheritance relationships. e granularity of knowledge organization needs to be further refined. Intangible cultural heritage knowledge ontology model still needs to be further expanded and enriched according to actual needs. e follow-up research will expand the data sources and research objects. In the big data environment, the organization and management of the massive heterogeneous knowledge of the intangible cultural heritage will be studied.

Data Availability
e experimental data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that they have no conflicts of interest to report regarding the present study.