Abstract
Introduction
Intangible cultural heritage (ICH) is an important part of human civilization that research is an effective way to explore. The universal value of ICH convention is gradually becoming socially accepted (Lenzerini, 2011). Kuutma (2011) believes that cultural heritage offers individuals the potential to gain access to social and political capital, and it also provides a channel for economic resources and plays an increasingly important role in politics, particularly in the international arena, where one can observe major resonances. Kirshenblatt-Gimblett (1995) called for the research of cultural heritage to be theorized more than 20 years ago.
However, many folklore scholars express criticism or concern regarding viewing cultural heritage as a set of global cultural and value frameworks. Noyes (2008) states that the exploration of culture should not come from a grand theory but instead from the achievements of scholars through ethnography or more grounded theories. Hafstein (2018) argues that some culture is called “intangible heritage” not to describe it but to intervene in it, with potentially disturbing results. ICH is not only a theoretical problem but also a part of daily life, with the potential to exacerbate political, ethnic, and ideological differences (Foster & Gilman, 2015). Hafstein (2008) considered the criteria and determinants of the Representative List of ICH, discussed at the 2003 ICH Congress, as the most controversial issue at the Congress. Bendix, Eggert, and Peselmann (2012) provide a valuable and extensive comparative study of how United Nations Educational, Scientific, and Cultural Organization’s (UNESCO) current heritage policies are implemented in individual countries. In some countries, many violations of the UNESCO Convention have not been sanctioned in the practice of protecting elements on the ICH representative list (Sano, 2005). Even in the selection of ICH projects, the dispute regarding ICH has never ceased, and many questions continue to be asked: Is an ICH project really of artistic value? Is the practice really endangered? Does it really have a full protection plan (Labadi, 2013)?
Although there remain certain controversies in the field of ICH, the research results in this field continue to emerge. By conceptualizing a macro view of the overall development characteristics and trends of the academic community through analyzing and reviewing the existing literature, we can better obtain a complete picture of the field of ICH research. Therefore, based on the above motivation, this study intends to analyze the following problems using bibliometrics analysis: (a) identify the nature of the research frontier, (b) annotate the research field, and (c) identify new trends and mutations over time. That is, this study seeks to identify which literature in this field is groundbreaking and iconic, which literature has played a key role in the development of the field, which topics occupy the mainstream position across the whole field of research, what the knowledge base is in this field, and how research frontiers have evolved. The data best matched to CiteSpace come from the Web of Science (WoS) database, but WoS does not have the widest coverage and collects primarily journal articles, excluding books published by scholars, and thus, the data collected may be limited in representativeness (Chen, 2017). The scientific measurement method used in this study is obviously different from a general literature review, and this study will also explore the differences between the results of the two different methods. As a review and analysis of previous studies, this study provides other researchers with useful references such as choices of research topics, determination of the research direction, and research methods.
Current State of Research
This article summarizes the research on ICH from three aspects, namely, the value of ICH, the protection and inheritance of ICH, and the utilization of ICH.
Tapping the Value of ICH
The exploration of the value of ICH has continued since the origination of the concept. Value has also become the basis for scholars to advocate protection and inheritance, as well as the development and utilization of ICH. First, ICH reflects the lives of people and human cultural activities in a certain area and in a certain period. With historical value that cannot be ignored, ICH is complementary to the omissions and inadequacies of official historical records and promotes a more authentic, comprehensive, and fair understanding of history (Alivizatou, 2016; Blake, 2015; Foster & Gilman, 2015). Second, ICH has scientific value (Yelmi, 2016). For instance, the traditional Chinese “feng shui” represents the wisdom of the ancients in interpreting nature (S. Zhang, 2018). Third, the rich aesthetic value of ICH has a positive effect on enriching people’s spiritual lives (Howard, 2016; Hwang, 2017; Lombardo, Pizzo, & Damiano, 2016). Fourth, economic value is the most important part of ICH in satisfying people’s lives and is also the greatest driving force for its continued existence (Fallon, 2016; Lee, 2015; Wright & Eppink, 2016).
Protection and Inheritance of ICH
Protecting ICH is a common appeal made by almost all scholars; only the approaches, methods, and strategies of protection are different. As a carrier of heritage, the community is considered to be an important force for protection. Many scholars believe that the community is a stakeholder of heritage and should actively participate in its protection (Cominelli & Greffe, 2012; Keitumetse, 2016; Kwon, 2017). Some scholars believe that the key to protecting ICH is to protect its inheritors and that archives should be established to protect heritage (Alivizatou, 2016; Fraser, 2015; McDonough, Kendall, & Bonn, 2017). How to inherit the ICH whose value cannot be reflected in contemporary times, such as local music, opera, and handicrafts, has also been a research topic (Ren, 2017; J. X. Wang, Yao, & Feng, 2017; Z. Z. Zhang, 2017). Another protection measure that should not be ignored is education regarding ICH. Only when ICH personnel are effectively trained can heritage protection work be sustainable (Dagnino, Ott, Pozzi, & Yilmaz, 2015; Ott, Dagnino, & Pozzi, 2015; G. J. Wang, 2017). With continued advances in technology, relying on technological means to protect heritage has become a current trend (Cozzani, Pozzi, Dagnino, Katos, & Katsouli, 2017; Piccialli & Chianese, 2017).
Utilization of ICH
In the current process of heritage protection, a school of thought has emerged with the belief that reasonable tourism development is an effective way to protect ICH and is also a means to generate economic value from heritage. After all, protecting ICH cannot rely solely on government support, inheritors, and the persistence of the community because their strengths are limited. To date, extensive research has been conducted on the tourism development of ICH, and the research has yielded considerable results (Esfehani & Albrecht, 2018; Jin, Hu, & Wang, 2017). Cultural creativity is also a way to develop ICH. The combination of ancient culture and modern creativity has given heritage a new vitality that has channeled the cultural essence of heritage, providing new support to people’s spiritual lives (Simeon & Martone, 2014; Thipwong, Ting, Huang, Chen, & Huang, 2018).
Research Methods and Tools
Related Theories and Conceptual Terms
Scientific knowledge map
Based on science, knowledge mapping is a cross-disciplinary field of applied mathematics, information science, and computer science and is a newer development in the field of science metrology. The purpose of mapping and analyzing scientific knowledge is to extract and visually reorganize the knowledge from a large number of previously published scientific research documents and to carry out knowledge discovery (Chen, 2003).
Research frontier
The concept of a research frontier was first proposed by Price and was used to describe the dynamic nature and ideological status of the research field. In general, a research frontier consists of approximately 40 or 50 recently published articles; studying the changes in a relatively small literature network such as this can be helpful to track the trajectory of an uncountable amount of literature (Price, 1965). Chen (2006) believes that the frontier of research reveals the emergence of theoretical trends and new topics.
Intellectual base
According to Persson (1994), from a bibliometric perspective, citations form the frontier of research, and the cited documents form the knowledge base. Chen (2006) considers the co-citation network formed by the cited classical literature as forming the intellectual base.
The concepts and relationships related to the research frontier and the knowledge base can be found in Figure 1.

Concept model of CiteSpace (Chen, 2006).
Betweenness centrality
It represents the number of shortest paths through the node in a network. In the knowledge map, the greater the centrality of a node, the greater the role it plays in the communication between other nodes.
Data Collection
By carefully compiling the world’s most important literature index, WoS has become the gold standard for research discovery and analysis, linking publications and researchers to databases spanning various disciplines by citing and controlling indexes. Reference searches can be used to track previous research and monitor recent developments in content over the 100 years that are fully indexed. 1 The data format processed by CiteSpace software is also based on the WoS data download format (Chen, 2015). Because WoS provides more complete references, indexes, and researcher relationships than other databases, the WoS database is used as the data source in this study.
In the process of collecting documents, two problems should be considered: recall rate and precision rate. Choosing a simpler retrieval strategy will help improve the recall rate, but the result may reduce the precision rate, while a more stringent retrieval strategy will reduce the recall rate but will significantly improve the precision rate. Although WoS provides a variety of search strategy combinations, CiteSpace primarily tracks and analyzes the evolution of a topic, so the targeted topic retrieval results are better (Chen, 2015). The sources of the subject words in WoS include titles, abstracts, keywords, and full text. Because articles are more academic in nature than conference proceedings, using a longer publication span helps to collect more a complete body of literature. This study used the core set of the WoS database as the data source and the search strategy of “Topic = Intangible cultural heritage + Document type = Article” to collect a total of 249 articles on April 13, 2018. To ensure data accuracy, the study individually reviewed the titles and abstracts of all the articles; all data collected met the requirements. Article data were then stored in the form of “Full Record and Cite References” using the “Plain text” format for subsequent analysis. Although the amount of data collected in this study is small, it still meets the requirements for CiteSpace analysis (Chen, 2017).
Research Tools
This study adopted CiteSpace 5.2.R2, a bibliometric tool developed by Prof. Chaomei Chen, and Carrot2, a tool that assists with text visualization, developed by Audilio Gonzales. Using these tools, the study performed a bibliometric analysis of the literature and displayed the results visually.
CiteSpace 5.2.R2 is a citation visual analysis software, developed from the background of scientometrics and knowledge visualization, which is specifically used to identify potential knowledge contained in scientific literature. This software can help researchers understand the basic knowledge of the discipline, find classical literature in the field, discover research frontiers, and clarify the context of research evolution (Chen, 2016).
CiteSpace transforms research domain concepts into mapping functions between research frontiers and intellectual bases, and the three concepts created within the framework of this mapping function concept are important to solve the following three problems: (a) identify the nature of the research frontier, (b) annotate the research field, and (c) identify new trends and mutations over time. CiteSpace has been used by users in more than 100 countries, and more than 15,000 papers have been published using the CiteSpace tool. 2
Carrot2 is an open-source search result clustering engine with a powerful clustering function that can aggregate small collections of documents into corresponding topic categories and can help researchers to cluster text documents (Singh, Suryawanshi, Gupta, & Saindane, 2016).
Research Methods
This study adopted a bibliometric method to conduct a scientific research collaboration analysis on the collected documents, including author collaboration, institutional collaboration, and national and regional collaboration. In addition, a literature and keyword cluster analysis was performed, and literature co-citations, author co-citations, and journal co-citations were analyzed.
Cluster analysis
Clustering is the process of dividing a set of physical or abstract objects into multiple groups. The group generated by clustering is called a cluster, which is the set of objects. There is a high similarity between the elements within a cluster and a higher degree of dissimilarity among different clusters (Chen, 2003). In CiteSpace, the clustering label is derived from the citation document and can be extracted from the title or index or summary of the cited document (Chen, 2015).
Cooperative analysis
The most basic organizational unit of scientific exploration has changed from the individual to the research group, and scientific cooperation has become the main form of scientific research. In scientific metrology research, the analysis of cooperative networks is mainly conducted to analyze the co-authorship of scientific and technological achievements of researchers (institutions and countries).
Co-citation analysis
The study of co-citation of literature was proposed by Henry Small (1973). The main principle of co-citation of literature is to measure the similarity between two papers by the number of times they are cited together, including periodicals, documents, and authors.
Research Results
Research Yield
The search found that the earliest ICH publication in the WoS database was the paper “Critical Natural Capital: A Socio-Cultural Perspective” published by Chiesura and de Groot in

The number of published papers on intangible cultural heritage (2003-2018).
Composition of Knowledge Group
Analysis of the knowledge group in a certain scientific field is an effective way to understand the basic knowledge, research field, and research direction. First, CiteSpace was used to convert the original data into the analyzable data format for Carrot2. Then, using Carrot2’s literature clustering function to cluster the literature, 3 the following set of cluster indicators was obtained: Cluster count base = 40, Cluster merging threshold = 1.0, Size-Score sorting ratio = 0.3, Phrase label boost = 5, Size-Score sorting ratio = 0.3, Phrase length penalty start = 3, and Phrase length penalty stop = 8 (Carpineto, Osiński, Romano, & Weiss, 2009; Osiński & Weiss, 2007). Clusters are displayed in the bubble diagram in Figure 3. The larger the size of the bubble in the graph is, the greater the number of publications included in the cluster. Notably, the topics that have attracted the most attention from scholars are heritage space, the intangible nature of cultural heritage, the internationalization of heritage, heritage protection, landscape heritage, heritage education, heritage management, and cultural policy. These results show that exploration of the essence of ICH and research on its protection are still the mainstream of research. The expression of artistic heritage, the development of heritage resources, and memory heritage have also attracted relatively more attention from scholars. The economic value of heritage and the sustainable development of heritage have become areas of scholarly focus. It is worth noting that with the development of computer technology, research on the digitalization of heritage has gradually become an important research topic, indicating that the protection of heritage is closely tied to the development of the times as well as to science and technology. Table 1 shows the detailed numbers for each cluster.

Visualization of the paper cluster foam tree.
Summary of the Cluster Types.
The above characteristics have been reflected in the research of many scholars in the field of ICH, such as Hoelscher (2006) explored the nature of heritage, Dümcke and Gnedovsky (2013) reviewed the economic and social values of ICH, Rössler (2006) explored the evolution of UNESCO undefined strategy for heritage landscape conservation.
Analysis of Scientific Research Collaboration
Scientific research collaboration is one of the primary means of interdisciplinary research and is the foundation for overcoming scientific challenges and achieving major breakthroughs in research. Katz and Martin (1997) defined scientific research collaboration as the working together of researchers to achieve the common goal of producing new scientific knowledge. There are many forms of scientific research collaboration. The research collaboration network mentioned in this article refers to the authors, research institutions, or countries (regions) that appear in the same paper. This article converts the raw data into the Jigsaw.jig file format, and then imports the data into the Jigsaw 0.54 software. To calculate collaboration, the following procedures are performed: select List function, use double column display mode, select Document name arrangement in the left column, select Author name arrangement in the right column, and load all the data. The number of various forms of cooperation in the research results for ICH is calculated: 116 articles have a single author, 57 have two authors, and 76 have three or more authors. The collaboration rate is 53.41%.
In the CiteSpace cooperative analysis, the relevant thresholds for CiteSpace are set as follows:
Because the time span of the data collected in this study is 2003-2018, 15 years, the time span is smaller than what Chen (2015) suggested, so we have selected 1 year as a time slice.
In this study, 249 pieces of data were collected, which forms a small sample. To obtain better atlas display, combined with the experience of Chen (2015), the node threshold of the top 50% was selected.
CiteSpace provides two network pruning modes, the path finding method and the minimum spanning tree. Due to its completeness (unique solution) and prominent structural characteristics of important networks, this study adopts the path finding method for network pruning (Chen, 2015).
In Threshold Interpolation setting, C, CC, and CCV are set to 2. C refers to the frequency of citation or occurrence of data, CC refers to the frequency of co-citation or co-occurrence between two data, and CCV refers to the rate of co-citation or co-occurrence between data (Chen, 2015).
The other indicators are based on the work of Chen (2015), using default values.
Co-author network
The co-author network is used to analyze the joint research in a certain research field. As shown in Figure 4, the CiteSpace analysis results show a total of 25 nodes, indicating 25 highly prolific authors, and 11 cooperative lines, indicating that the collaboration rate between the authors is very low. No relatively stable and large academic teams have been formed. There are two teams with three authors and five teams with two authors. The larger the nodes are in the figure, the higher the number of publications published, and the thicker the connecting line is, the closer and stronger the collaboration. In terms of papers, the most prolific authors have three papers published: These authors are S. Rwawiire, B. Tomkova, C. Grant, and J. Schofield.

Visualization of the co-author network.
Co-institute network
The CiteSpace analysis shows that there are 40 research institutes and nine collaboration links in the co-institute network. Figure 5 shows that most of the nodes are isolated points, indicating that almost all the research results have been completed by a single author. Only a few organizations have had collaborative experiences, and the intensity of collaboration is very weak. In terms of the number of publications, the top five are the University of Melbourne, University of Seville, University of Queensland, University of York, and Griffith University. Based on the previous research, the frequency of occurrence of data must be higher than 2 on the network (Chen, 2015). While in this study, the maximum number of publications is only 6: It is a relatively small number of publications for one institute, indicating that the institutes’ research on ICH has not gone deep enough.

Visualization of the co-institute network.
Co-country (region) network
The CiteSpace analysis revealed a total of 40 countries and regions as well as 44 collaboration links (Figure 6), indicating that in the research process, cross-country (region) links are more frequent. As shown by the collaboration lines in the network, although there are frequent links between countries and regions, the intensities of links are weak, indicating that such cross-country and cross-region links may be a collaborative attempt but are not yet sufficiently stable. Because the CC value is 2, while the co-occurrence of most countries is also at the level of 2, just meeting the basic requirement of analysis (Chen, 2015).

The visualization of the co-country network.
The top countries in number of publications are China, the United Kingdom, Australia, the United States, Italy, Spain, and Turkey. In particular, it is worth noting that Australia is not the country with the most publications, but its citation burst is 2.90, which the other countries do not have. This result shows that Australia has made significant contributions in this field. Using node centrality as a metric, European and American countries have played a key role in this research field. For example, the centrality value for the Netherlands is 0.54, and the centrality value for the United States is 0.39 (see Table 2).
Top 10 Countries in the Study of Intangible Cultural Heritage.
Keyword Co-Occurrence Analysis
Keyword co-occurrence analysis is an effective way to elucidate the structure of scientific knowledge, explore research hotspots, and discover research trends. Figure 7 is arranged by time series and depicts the keyword maps of the research on ICH from 2003 to 2018. A total of 67 high-frequency words and 191 connection lines are presented. Modularity

Visualization of keywords by time series.
Table 3 shows the high-frequency keywords for each year according to Figure 6 (frequency >5), which shows that the research on cultural heritage began in 2006, and during this period, only one keyword, “cultural heritage,” occurred with a relatively high frequency and large centrality, indicating that the main research at this time was centered on the concept of cultural heritage. In 2011, there were more high-frequency words. The keyword of “intangible cultural heritage” had the highest frequency (40) and the highest centrality value (0.44). Words such as “culture,” “identity,” and “China” also showed a high frequency, indicating that the cultural identity of ICH became a research hotspot in this period. Since 2012, there have been many high-frequency words, but their centrality is relatively low, indicating that no additional hotspots have formed. The reason for this lack of hotspots is that the study of ICH in this period is in the stage of encouraging open and free ideas.
High-Frequency Keywords in the Study of Intangible Cultural Heritage.
Through the CiteSpace clustering function, keywords were used to extract information, and the log-likelihood ratio (LLR) was used as the calculation method to obtain clustering results. Five effective clustering tags were obtained (silhouette score >0.5). Table 4 presents detailed information for each cluster. To better explain the clustering results, data are randomly selected in each cluster as illustrative examples.
Clusters by Keyword in the Study of Intangible Cultural Heritage.
Intangible heritage
Figure 6 shows that the theme of intangible heritage began to attract scholars’ attention in 2011. The research on intangible heritage has mainly focused on new methods and new perspectives. For example, through new subjective methods, Swensen and Saeter (2011) applied the Mall Method test to examine subjective appreciation of urban cultural heritage from the perspective of residents, and Casey (2013) took the Okinawa Tarama Village’s declaration of the August Dance Festival as ICH as an example and adopted a more global, system-oriented approach to articulate the interaction between overlapping areas of heritage, tourism, and politics.
Authenticity
Research on authenticity has mainly focused on the theoretic paradigm of authenticity (Carroll & Wheaton, 2009; Chambers, 2009). Some scholars have conducted empirical research on the authenticity of various heritages, such as the empirical study of Kam Big Song in Southwest China (Ingram, 2012) and the empirical studies from the background of the Vanuatu Cultural Centre and the research institutions of the Melanesia Islands (Alivizatou, 2012). Paddock and Schofield (2017) studied the authenticity and adaptability of ICH based on Mongol Ger, which is an ancient civilization that gradually disappeared.
Creative tourism
The protection of heritage and tourism development has always been a focus of academic debate. Many scholars advocate that ICH should be appropriately developed and that tourism development is an effective way to sustain the development of heritage. Creative tourism is a way to develop ICH tourism. For example, archeological tourism is a creative method (Ross, Saxena, Correia, & Deutz, 2017). With the development of digital technology, digitalization of ICH has been accomplished. The digital technology and development of creative tourism in heritage have become a current model (Marques & Borba, 2017).
Economic benefits
Economic factors have a key impact on the effective protection of heritage. Many ICHs are not effectively protected. One of the reasons for this lack of protection is that their social value cannot be brought into play in contemporary society. How to tap the value of ICH is an important research topic that includes tapping the market value of cultural heritage (Lazrak, Nijkamp, Rietveld, & Rouwendal, 2014), the social benefits of intangible resources (Buckley & Graves, 2016), the improvement of the life quality of residents of heritage sites, and the stimulation of inheritors’ enthusiasm (MacRae, 2017).
Culture
Culture is a basic attribute of heritage. The exploration of the cultural attributes of heritage is the essence of intangible culture. Either a macro interpretation or an empirical study of specific cases is an in-depth excavation of cultural connotation. Bideau and Kilani (2012) believe that multiculturalism and cosmopolitanism are natural attributes of ICH. Morel used the Tango, the “MILONGA THAT ERASES FRONTIERS,” as an example to explore strategies for heritage inheritance (Morel, 2011). Beardslee took the ICH of Jemaâ el-Fna in Marrakesh as an example to explore whether the rights of residents would be weakened in the process of empowering the community (Beardslee, 2016).
Analysis of Co-Citations
Analysis of literature co-citations
Co-citation cluster analysis is frequently used in the bibliometric studies. When the quality indicators such as Silhouette score and Modularity
The analysis graph generated by CiteSpace shows that there are 62 highly cited papers and 120 co-citation connection lines in the analysis results of the literature co-citations. Similarly, keywords were used to extract information, and the LLR calculation method was selected for clustering, resulting in three cluster tags (Figure 8). Modularity

Visualization of the reference co-citation network.
Summary of the Reference Clusters.
The highest frequency of citations occurred with the two books published by Smith in 2006 and 2009. The first book introduced the utilization of heritage (Smith, 2006). The second book gave a relatively comprehensive introduction to ICH (Smith & Akagawa, 2009). The third most cited work was Kirshenblatt-Gimblett’s (2004) paper published in
Top Five Most Cited Papers, With Co-Citation Frequency.
CiteSpace can also be used to analyze the frontier literature of this research field. Table 7 shows part of the frontier literature of each cluster. Most of the papers were published in the
The Frontier Literature on Intangible Cultural Heritage.
Examining each cluster revealed that the papers in Cluster0 are relatively new and are the results of the past 2 years. The tag for Cluster0 is “Nationalism,” which shows that in this field, scholars paid more attention to the global nature of heritage. Aykan believed that heritage is a common property of people all over the world and does not have national borders; nonetheless, the nationalist approach to intangible heritage is the main obstacle to cross-country heritage. Using the Karagöz Shadow Theater as an example, Aykan (2015) called for international collaboration to protect the world’s intangible heritage. Diettrich (2015) used the performing arts of the Federated States of Micronesia in the Western Pacific Ocean as an example, analyzing ICH and music ecology theories through the colonial perspectives of historical protection to explore the relationship between heritage policy making and colonialism. When studying the cultural landscape of Bali, Indonesia, MacRae discovered that there was a gap between the understanding of the global world heritage system and the local community, which caused a series of unexpected consequences. MacRae (2017) argued that it is not enough to rely on the local government to protect heritage.
In Cluster2, most of the papers were published from 2012 to 2014 and were relatively concentrated, indicating that the attention paid to this theme was relatively short. The tag for Cluster2 is “Heritage,” indicating that in this field, scholars paid more attention to the nature of heritage itself, such as the multicultural characteristics of heritage and the characteristics of the interaction between heritage and people. Research by Bideau and Kilani (2012) showed that under the coexistence of Malaysian multi-ethnic people (Chinese, Malaysians, and Indians), the tangible construction of social models is based on the construction of the ICH of different ethnic groups. Bille’s study on the interaction between Bedouins and heritage in Jordan led to the finding that heritage played a “decorative” role in the process of Bedouin resettlement, as seen in the example of the inscription on heritage. Such interactions between people and heritage have preserved heritage (Bille, 2012). Roberts linked specific examples of popular music with the broader debate over cultural heritage and cultural heritage industrialization, calling for a larger discussion on the use of popular music as a cultural heritage discourse. The essence of Roberts’s (2014) study was to explore how to protect heritage as a cultural property.
Notably, “Fraser Island” became a clustering tag in Cluster3. Although there are only four papers in this cluster, they are closely related, and the themes are highly consistent. The specific literature shows that they are studies on the cultural and historical significance of ICH from archeological perspectives (Schaepe, Angelbeck, Snook, & Welch, 2017; Weisse & Ross, 2017; Fernandez & Moshenska, 2017).
Author co-citation analysis
Analysis of the author co-citations generated a total of 218 highly cited authors and 616 network lines, as shown in Figure 9. The larger the node on the graph is, the higher the citation frequency. The results of UNESCO’s research are the most cited, which is in line with the nature of its work and shows that it has played a tremendous role in promoting research on ICH. Further processing of the data produced Table 8, which shows that UNESCO, Smith, and Kirshenblatt-Gimblett were the most cited authors, indicating they have made substantial contributions to the research in this field. The centrality values of Smith, Kirshenblatt-Gimblett, Byren, and Appadurai all exceed 0.1, indicating that they have played a relatively large role in this field. The citation rates and centrality values of Smith and Kirshenblatt-Gimblett are highly significant, indicating these two authors have played a prominent role in this field. As shown by the citation bursts in Table 9, the citations of the papers by UNESCO, Kirshenblatt-Gimblett, and Appadurai have all grown rapidly over the period under study, demonstrating that the opinions of their papers have generally been accepted. To sum up, Kirshenblatt-Gimblett has made the largest contribution in this field.

The visualization of the author co-citation network.
High-Frequency Author Citations in the Study of Intangible Cultural Heritage.
Top Three Cited Authors With the Strongest Citation Bursts.
Journal co-citation analysis
CiteSpace journal co-citation analysis extracted a total of 251 highly cited journals and 733 network lines. As shown in Figure 10, the most cited journal is the

The visualization of the journal co-citation network.
High-Frequency Author Citations in the Study of Intangible Cultural Heritage.
Discussion
Comparison Between the Scientific Metrology Method and the Traditional Literature Review Method
The biggest advantage of scientific metrology lies in analyzing massive data, combing the complexities of cooperative and citation relationships, extracting the classical literature and the frontier literature, and clearly illustrating the evolution path of the subject visually. The title, abstract, keywords and so on produce the thematic subject words, with which cluster analysis can be produced. Because the data collected in this study form a small sample, it cannot completely demonstrate the advantages of scientific metrology. Nonetheless, in some mature disciplines, the advantages are obvious when the number of studies in the literature reaches the thousands or even tens of thousands, and this type of analysis can be repeated among different researchers. As long as the criteria are the same, the results will certainly be the same. For example, Prof. Chaomei Chen, the CiteSpace developer, has used research on terrorism (1990-2003), mass extinction (1981-2003), and regenerative medicine as teaching cases (Chen et al., 2002; Chen et al., 2014), and the results have been repeatedly verified.
The tradition of the literature summary method relies on the scholar to carry out manual collection, extensive reading, and inductive summary. Literature collection is based on personal standards, the literature may originate from many databases, and the types of literature can vary. Moreover, due to scholars’ limited time and energy, the data that are collected by manual analysis are quite limited, and even with the same literature, different author analysis may result in different results. In addition, it is difficult for scholars to analyze the relationship between studies from massive data, and the analysis of the evolution of disciplines is far less intensive than that produced by scientific metrology. After all, the data of artificial analysis are quite limited.
However, scientific metrology is not impeccable. Scientific metrology relies on the support of database structure, the process of data collection, and the relative mechanization of the extraction of subject words by computer. There may be a certain gap between the results of econometric analysis and the results of empirical analysis (Chen, 2006; Chen, 2010). Although there are weaknesses and shortcomings in citation analysis in science metrology, such as the balance between recall rate and precision rate mentioned in the previous data collection section, the information it studies is difficult to replace, and the results of the study are relatively certifiable, in that they can be verified and audited (Chen, 2017).
Discussion of the Results of This Study
Yield
The results show that the number of high-yield authors is lower and the output of high-yield authors is less. There are several primary reasons for this phenomenon. First, the number of ICH journals included in the WoS database is relatively small, which can be confirmed by the co-citation analysis of journals. Second, as an emerging discipline, ICH is younger than other disciplines, such as biology, philosophy, chemistry, and so on, and thus, the yield will be quite different. This characteristic is the most substantial difference between this research method and traditional human reading literature analysis, and also the greatest difficulty: analysis is greatly affected by the database. Third, many articles published by scholars are not included in the WoS database, and furthermore, the results of many scholars’ research are not published in the form of papers but may be published in books, reports, or conference papers.
Collaboration network analysis
The composition of the research team obtained by CiteSpace analysis is different from that of the real research team. For example, the well-known Sun Yat-sen University in China appeared in the analysis, but it was not prominent. Moreover, the prominent cultural heritage research team of Simon Fraser University and the cultural finance research team of Georg-August-Universität Göttingen did not even show up in this study, which clearly conflicts with the facts. There are several possible reasons for this discrepancy. First, the research results published by the research team are not included in WoS, resulting in a lack of data, thereby affecting the fairness of the results. Second, this study uses “intangible cultural heritage” as the search keyword, while similar research teams do not necessarily meet the standard of “intangible cultural heritage” when studying ICH. For example, in Japan and South Korea, ICH is defined as “intangible cultural property” (Jongsung, 2004), while some studies use “cultural heritage” as the theme, which may lead to deviations in data collection.
High yield and high influence
In the results of the analysis, the high-yield authors and the high-influence authors are not consistent with each other. In other words, a high-influence author does not necessarily have high yield, and a high-yield author does not necessarily have high influence. This result may exist in reality. However, there is a strong consistency between high-yield journals and high-influence journals. Because of the accuracy of positioning, the specialization and authority of certain journals lead to the inevitability of this phenomenon.
Conclusion
Although bibliometric work is common in other fields, it is relatively rare in ICH sector. Behles (2014) conducted a study on the citation of
Based on the topics collected from the WoS database, this study used CiteSpace software for analysis and visualization. The results possess a certain reference value. First, since 2011, the research on ICH has begun to increase substantially, and the trend is obvious. Second, studies on ICH have mainly focused on heritage itself, such as heritage space, heritage protection, heritage management, landscape heritage, and so on. There is relatively little research on the use of heritage and sustainable development. It is gratifying to note that the research on the internationalization of heritage has generally received attention from scholars, which is conducive to either the protection of heritage itself or heritage research from the macro level. Third, certain differences exist between the research results of ICH research teams and reality. These differences may be because the collection of data is not comprehensive or because the subject words resulted in the collection of limited criteria. Fourth, in terms of yield, China, Australia, the United Kingdom, the United States, and Spain ranked highest, playing a propelling role in the research on ICH. Fifth, judging from the co-citation network, high influence is strongly consistent with high-yield journals, but high-impact authors are weakly consistent with high-yield authors. Finally, there have emerged some new directions in the study of ICH, such as creative tourism of heritage, community participation in heritage protection and development, and authenticity of ICH.
Although the study of ICH has been a subject of concern by scholars, the theme is relatively loose, with the visual representation of themes akin to the blossoming of a hundred flowers, which is very helpful to the multiperspective and multilevel study of ICH. Of course, the deep-level study of heritage in developing countries cannot be ignored. The existing results show that most of the research has been concentrated in relatively developed countries such as the United States, the United Kingdom, Australia, and Spain. It is worth noting that China, as a developing country, has made a relatively prominent contribution to the research. The next step is to actively guide and encourage other developing countries to carry out the same research. Many of UNESCO’s research accomplishments have been recognized by scholars. As an international organization, it is worth looking forward to using its coordination capabilities in the future to promote research collaboration among various countries, research institutions, and researchers. These studies can provide criticism, affirmation, or debate about ICH. Many studies have also shown that heritage protection in economically backward areas faces many difficulties, and poverty alleviation in developing countries has always attracted the attention of the international community. How to help the local community shake off poverty to become more economically stable in the process of heritage protection may become a topic for future researchers.
The data for this study come from the WoS, which is a database of relatively high-quality papers, but the paper quantity in the database is relatively small. Many research results published in other journals are not included in the WoS database and were not collected in this study. Therefore, there are certain limitations to the representativeness of the results. To expand the breadth and depth of the study, it is suggested that other scholars can use other analysis tools to collect data from other databases outside the WoS for analysis and comparison. The results can be compared to identify similarities and differences so that more detailed and valuable research will be provided to scholars. Scholars also can use other analysis and visualization software to cluster, analyze, and display the data, providing readers with more detailed visual effects. In short, there is still substantial room for research on ICH, and scholars need to use different paradigms, methods, and perspectives to explain and measure ICH to further broaden the research field.
Supplemental Material
Appendix – Supplemental material for A Bibliometric Analysis of Research on Intangible Cultural Heritage Using CiteSpace
Supplemental material, Appendix for A Bibliometric Analysis of Research on Intangible Cultural Heritage Using CiteSpace by Xinwei Su, Xi Li and Yanxin Kang in SAGE Open
Footnotes
Declaration of Conflicting Interests
Funding
Supplemental Material
Author Biographies
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
