Though a keyword searching has been public recently, major subject heading lists have been used as subject access tools to support users' information retrieval. Subjects have been expressed by complex nouns without subdivisions in subject heading list...
Though a keyword searching has been public recently, major subject heading lists have been used as subject access tools to support users' information retrieval. Subjects have been expressed by complex nouns without subdivisions in subject heading lists in Korea. Also, properties and structures of headings have not been defined in proper. To solve these problems, subject headings should be categorized in facets, and synthetic rules should be defined among facets.
The purpose of this study is to develop a faceted application model of subject headings for the current information environment in Korea. A faceted model is designed to reflect the fundamental properties of subject headings and the characteristics of the Korean language. For this study, literature review, case study, and facet analysis were carried on.
In literature review, the concepts of subject access and subject, the structure of the subject headings, the characteristics of Korean headings, the definitions of facets and the procedures of facet analysis were investigated. As a result, the categories of subject headings were divided into Aboutness and Isness. Also, it was required to apply the faceted model in subject headings to solve the structure’s complexity of traditional subject heading lists and to express various combinations. In addition, the facet analysis was consisted of 6 steps including subject field definition, facet formulation, facet amplification and structuring, creation of scope note, facet arrangement and definition of facet coordination rule, and has been used in developing or assigning subject headings.
In case study, six cases including OCLC's FAST, PRECIS, Thesaurus construction and use, CC 7th edition, BC 2nd Edition, and UDC 3rd Edition were analyzed to focus on configuration and combination of facets. As a result, in Aboutness category, the number, the name and the segmentation of facets were various in 6 cases. Also, the number and the type of facets were different from cases in Isness category. The Form facet was the common facet in all cases, and Language facet, Genre facet, and Person facet were treated partially. In the aspect of facet combinations, basically postcoordination was applied among the different facets and then postcoordination was applied in the same facet in some cases. The citation order was different from the cases. In Aboutness category, the order of facets was defined as CC’s PMEST except Geography facet and Chronology facet. Whereas, in Isness category, the citation order of facets was various in many cases.
As facet analysis, 8,231 headings of the National Library of Korea Subject Headings (NLSH) were classified based upon the initial faceted models, and were analyzed the type of combinations of precoordinated headings. The facets in the initial model were consisted of 22 facets including 8 facets of FAST and 14 facets of the other five cases.
As a result, first, in the aspect of the terminology properties, most of headings were Chinese characters and foreign words were quite included. Also, academic and scientific terms were coexisted with general and public terms, and there were a lot of compound and complex headings. Second, in the aspect of facet distribution, 6,911 headings (83.96%) were classified in Topical facet, and the other 1,320 headings (16.04%) belonged in other facets. Third, the number of headings in Things/Entity facet was 3,926 (56.81%) compared with 2,985 headings (43.19%) in Action facet.
Fourth, 26 subject headings (0.32%) were classified in Event facet, and the subproperties of Event facet were the names of historical events, wars, and meetings and conferences. Fifth, 225 subject headings (2.73%) belonged to Geography facet, and then subproperties were the names of continents, countries, cities, rivers/oceans/ports, mountains/mountain chains, and ancient countries. Sixth, 30 subject headings (0.36%) were classified in Chronology facet, and the subproperties were the names of general subdivision of periods, periods in Korea and foreign countries. Seventh, as proper nouns, there were 76 subject headings (0.93%) in Personal Name Facet, 21 subject headings (0.26%) in Corporate Name Facet, and 150 subject headings (1.82%) in Title Facet.
Eighth, 109 subject headings (1.32%) were classified in Form facet, and they were divided into miscellaneous, dictionaries, lectures/speeches, serials, educational materials, series/collections/anthologies, and biographies. Ninth, 185 subject headings (2.25%) were included in Genre facet with the specific styles of art and literature, and the subproperties were separated by the general subdivisions of genre for art and literature, the specific genres of music, fine art, movie/play, broadcasting, and literature. Tenth, 45 subject headings (0.55%) were classified in Language facet, and subproperties were the specific languages and letters divided by language families, and artificial languages. Eleventh, 453 subject headings (5.50%) belonged to Person facet, and gender/age, ethnic groups, social status/classes, and profession/qualification were suggested as the subproperties.
Then, the results of the facet classification about 600 subject headings were verified for twice and problems and improvements were gathered with three professors in information organization field and three librarians in technical service. As a result, first, the facet identity was improved in the second verification better than earlier. In particular, Geography, Event, Chronology, Personal and Corporation Name, Title, and Language facets were highly identifiable, but Topical, Form and Genre facets were relatively low. Second, the name of Action facet was required to be changed to 'Action/Status facet' because it was difficult to associate the status nouns with the facet. Third, it was necessary to allow multiple classification or to integrate Things/Entity and Action/Status facets into Topical facet in cases of science and history, and disease, because the definitions of subject headings in the dictionary could be classified in entity as well as in action. Fourth, it was recommended to allow topical headings including specific languages and subdivisions of individual languages to classify both Topical and Language facets.
Next, precoordinated 755 subject headings with two or three facets were analyzed in aspects of the combination method and citation order. As a result, in combination of two facets, Topical facet was most actively combined with other eight facets to form the precoordinated headings, and then, Genre, Form, and Geography facet were combined gradually. In combination of three facets, there were twenty-five subject headings and five different types of combinations.
Based on the results, the final faceted application model of subject headings in Korea was proposed in two phases. In the first phase, three groups were divided according to the fundamental properties of subject headings referred to the IFLA's FRSAD model. Also, facets and subcategory in three groups were suggested. Group 1 was made up of Topical, Event, Geography, and Chronology facet related to Aboutness. Group 2 was composed of Personal and Corporate Name, and Title facet as proper nouns, and Group 3 was consisted of Form, Genre, Language, and Person facet related to Isness. Then, Topical facet in Group 1 was subdivided to five types based on various combinations of precoordinated headings.
In the second phase, eight subfacets and forty-four subproperties of the eleven facets were proposed. As top facets, Topical, Event, Geography, Chronology, Personal and Corporate Name, Title, Form, Genre, Language, and Person facets were defined. Next, the subfacets were Topical-Thing/Entity and Topical-Action/Status facet, Part, Kind, Property, Whole, Patient, and Agent facets associated with two facets.
In case of subproperties, there were three types of historical event, war, meeting and conference in Event facet. In Geography facet, seven types of continent, country, city, river/ocean/port, mountain/mountain chain, ancient countries, and overland route, and in Chronology facet, three subproperties corresponding with the general subdivisions of periods, periods in Korea and foreign countries were defined. In Personal Name facet, three properties of domestic and foreign personal name, and family name were subdivided, in Corporate Name facet, domestic organization, and foreign & international organization were presented as two subproperties. As well, the five subproperties of Title facet were domestic and foreign books, scriptures, songs, and products/standards.
The seven subproperties of Form facet were miscellaneous, dictionaries, lectures/speeches, serials, educational materials, series/collections/anthologies, and biographies. Then, in Genre facet, six subproperties including the general subdivisions of genre for art and literature, the specific genres of music, fine art, movie/play, broadcasting, and literature were divided. The three subproperties of Language facet were the specific languages and letters divided by language families, and artificial languages. Four subproperties such as gender/age, ethnic groups, social status/classes, and profession/qualification were categorized in Person facet.
Next, the scope notes of facets were made up of the definitions and classification principles of facets. The subproperties and examples of headings were added in the definitions based on results of facet analysis to the initial model. Then, the principles and guidelines related to three types of problems that were stated in facet analysis and verification process were provided.
Finally, the rule of citation order suitable to Korean subject headings was suggested, and the syntax was applied to build and search subject headings in virtual. The citation order was defined as Geography, Language, Person, Chronology, Genre, Topical, and Form, and it was parallel to language structure of Korean. Then, faceted structure made and expressed complex and compound subject headings more consistent and systematic without subdivisions. Also, it facilitated to access subject headings using coordinate search or filter and limitation of search results.
This study was valuable in that the facet structure of subject headings in Korea was revealed, and the citation rule of facets conforming to the order of Korean language was defined. Based upon the result of this study, the traditional subject heading lists in Korea can to be developed for a better quality and to play a key role as a subject access tool, and to increase subject accessibility of users.
|
