Knowledge organization

Literally KO is organization of knowledge. Organization (or order) being an arrangement, or organized structure for arranging or classifying. This organization is made in order to facilitate the use of documents or recorded knowledge (or other units). "The term "knowledge organization" taken literally means making knowledge an "organum" (Greek = instrument, aid) for particular purposes" (Kiel, 1994).

In this definition is KO defined as processes. It should be recognized that KO is the study of both knowledge organizating processes and knowledge organization systems (KOS). Both aspects may be considered in a narrow perspective within Library and Information Science (LIS) and within a broader, interdisciplinary perspective.

1) In the narrow sense (within LIS) is KO concerned with information in bibliographical records.

In the narrowest sense of KO are bibliographical records parts of library catalogs (such as MARC-records). There are somewhat different traditions concerning such records in different kinds of libraries. For example have public libraries often universal classification schemes as the basis for subject data in their records while special libraries such as National Library of Medicine may use special classification schemes.

In a more general sense within LIS are also records in documentation databases such as Medline and the Citation Indexes considered. These databases are based on other traditions and standards compared to library catalogs (this field used to be termed documentation but changed its name to information science about 1968). For example is the concept of "a work" important in the library catalog tradition but not in the documentation database tradition. The thesaurus is an example of a knowledge organizing system (KOS) developed in the culture of online databases as distinct from the library tradition. The citation databases are connected to the field of bibliometrics, which exemplify how different designs of bibliographic records are connected to different research traditions within LIS. (Bibliometrics is concerned with patterns of citations between documents as indexed in a special form of bibliographical records).

In a yet broader sense is KO concerned with records in archives and museums. These institutions build their cataloging practices on other traditions compared to both library catalogs and documentation databases. (One famous organizing principle unique to archives is the principle of provenance). In some countries (e.g. in Sweden) are museology and archival sciences independent from library and information science, but there are also trends which try to view such "memory institution" as a common theoretical frame. While it is understandable that museology and archival sciences are different from library science, it is a bit paradoxical that they are also different from LIS because the word "information" in the term LIS should denote a general field not limited to libraries and documentation databases (Many people in e.g. archival science consider this field a subfield of information science). Concerning subject analysis it is important to consider that the principles for ordering knowledge in archives, museums and libraries are basically all derived from the same epistemological principles. (Ørom, 2003, for example, demonstrates how knowledge organization in arts museums, in arts libraries and in arts literature are all depending on the same fundamental conception of art).

In a yet broader sense is KO concerned with full-text documents and Internet documents and their representation for retrieval. Although much of this last field is developed outside LIS and KO it has to be considered as a part of the object of KO if this field should play any important role in the future. Such full-text representations may - with some generalization of the concept - be considered a part of the concept of the bibliographical record. Bibliographical records are then what defines the focus of KO within LIS.

In 1957-1962 the Cranfield I experiments claimed that traditional forms of classification and indexing (e.g. the UDC) were rather inefficient compared to retrieval based on simple "uniterms" or alphabetical subject headings (cf., Ellis, 1996, 3-6):

Uniterm 82,0% recall

Alphabetical subject headings 81,5% recall

UDC 75,6% recall

Facet classification scheme 73,8% recall

Findings from Cranfield I (after Ellis, 1996, 3-6)

Although a lot of interpretations and subsequent experiments have been made, have traditional forms of classification, indexing and KO since then mostly been regarded as less efficient, and the researchers in the traditional area of KO (such as Bliss, Ranganathan and Mills) have not, for example, been visible on bibliometric maps displaying researchers in LIS (e.g., White & McCain, 1998). The dominant and visible trend in LIS have mostly been associated with experimental research in IR such as Cranfield or TREC. This is important to consider in relation to the teaching of KO as an academic subject, although other interpretations of the state of the field can be made and should be made (results are still rather inconclusive).

Practical education in KO tends to teach one kind of KO associated with one of the traditions mentioned above (also many researchers tend to be familiar with only one or a few traditions). This is not the aim of an academic education in KO which must introduce all traditions and explain why different kinds of systems have been developed to different kinds of tasks. The convergence of media on the Internet also makes it more important to know these different traditions in order to evaluate their respective possibilities and limitations and to learn potentially appropriate principles from each of them.

2) In the broader sense is KO about how knowledge is organized in, for example, scientific disciplines, higher educational institutions, encyclopedias, dictionaries, languages, genres, theories, minds etc.

KO in the broad sense is necessary to understand in order to understand KO in the narrow sense. This is the case because the organization of, for example, library catalogs such as the Dewey system, is basically made according to the system of scientific concepts and disciplines. It is the system of scientific disciplines that are used to organize bibliographical records in DDC, while this broader system must somehow be described with LIS/KO if a proper theoretical framework for LIS is to be established. Because of this fact is the sociology of knowledge, among other fields, to be regarded as part of the teaching of KO in LIS.

"The relation of a system to its culture means that no system is an innocent expression of self-evident universal relationships . . . decisions about which kinds of relationships belong in a system are taken outside the system and thus of necessity reside, more or less obviously, in both the syntactic and semantic devices of the system that are in turn derived from its cultural warrant" (Beghton, 2001, 105).

From the point of view of the system of professional communication as modeled in the UNISIST model (cf., Fjordback Søndergaard, Andersen & Hjørland, 2003; Hjørland, Fjordback Søndergaard & Andersen, 2005), the classification of bibliographical records belong to the secondary literature/services, while classification of concepts belong to the primary literature. This is important for LIS because knowledge organization is normally based on literary warrant, which means based on relations discovered or constructed in the primary literature of a domain.

The term "Knowledge Organization" seems to be established around 1900 by people like Charles A. Cutter, W. C. Berwick Sayers and Ernest Cushington Richardson. Also Henry Bliss' book (1929) The organization of knowledge and the system of the sciences represent one of the main intellectual contributions in establishing the field. All of these authors argued that book classification is based on knowledge organization such as it appears in science and scholarship, why they founded knowledge organization in the broader sense. KO within LIS is dependent on knowledge produced outside LIS. LIS belongs to the field of meta-science. In 1993 the journal International Classification changed its name to Knowledge Organization. International Journal for Concept Theory, Classification, Indexing, and Knowledge Representation, which contributed to consolidate KO as the label of a field of research.

The broader field of KO is an interdisciplinary area covering, among other fields, linguistics, philosophy, psychology, semiotics and sociology. An example of a book based on this broad understanding of KO is Oleson & Voss (1979) The Organization of knowledge in modern America, 1860-1920. It is important that the necessary elements from those broader fields of KO become available in a compact integrated form, on which LIS-researchers can draw.

Different approaches to library and information science may draw more or less on different fields. Some have argued that the human mind naturally organizes its knowledge of the world into conceptual systems. This view is often based on the epistemology of Immanuel Kant and is related to cognitive views in Library and Information Science (LIS) as well as in cognitive science. Anthropologists, on the other side, have observed that taxonomies are generally embedded in local cultural and social systems, and serve various social functions. This view is related to the domain analytic view. Cognitive views, for example, emphasizes how knowledge is organized in the mind or brain (e.g. in "the mental lexicon"), while the domain-analytic view emphasizes the social and intellectual organization of knowledge (e.g. in disciplines and theories).

It is important to realize that no general knowledge can replace subject knowledge in knowledge organization. Knowledge is always about something specific. The structures in specific domains are discovered or determined by professionals in those areas. Often (if not always) are there conflicting views on how a specific domain should be organized. Any given KOS (such as the UDC) is always more or less reflecting a certain view of the fields it covers (cf., Ørom, 2003, about the Arts). The construction, use and evaluation of KOS is primarily about how a given field (e.g. art) is conceptualized and represented in the systems. This is the reason why many high-quality systems (such as Medline) demand high levels of domain-specific knowledge (e.g. advanced degrees in biomedicine, cf., National Library of Medicine, 2005). The domain-analytic approach to LIS is a suggestion on how to teach necessary domain knowledge within LIS).

It is important, however, to remember that the bibliographical record (in a broad sense, including metadata in full-text records) is what defines KO within LIS. KO in the broader meaning is only relevant for LIS to the degree that it contributes to the design, use and evaluation of bibliographical records.

Anderson, J. D. (1996). Organization of knowledge, in: Feather, John & Sturges Paul (Eds.): International encyclopedia of information and library science, London & New York: Routledge, pp. 336-353.

Beghtol, C. (2001). Relationships in classificatory structure and meaning. IN Bean, C. A. & Green, R. (Eds.). Relationships in the organization of knowledge. Dordrecht: Kluwer Academic Publishers. (Pp. 99-113).

Dahlberg, I. (2006). Definitionen aus dem Begriffsfeld „Wissensorganisation“ vorgelegt von Dr. Ingetraut Dahlberg, Bad König, 12.9.2006. (in German and in English). Click: Dahlberg_2006_Knowledge Organization

Ellis, D. (1996). Progress and Problems in Information Retrieval. London: Library Association Publishing.

Fjordback Søndergaard, T.; Andersen, J. & Hjørland, B. (2003). Documents and the communication of scientific and scholarly information. Revising and updating the UNISIST model. Journal of Documentation, 59(3), 278-320.

Hjørland, B. (2003). Fundamentals of knowledge organization. Knowledge Organization, 30(2), 87-111.

Hjørland, B; Fjordback Søndergaard, T. & Andersen, J. (2005). UNISIST model and Knowledge Domains. IN: Drake, Miriam (Ed.). Encyclopedia of Library and Information Science. New York: Marcel Dekker. (Electronic publication. Available for subscribers on: http://www.dekker.com/servlet/product/DOI/101081EELIS120024989

Johansson, J. (2006). Svensk kunskapsorganisationsforskning 1990-2002. [Swedish research in Knowledge Organization 1990-2002]. Borås: Högskolan i Borås. (Magisteruppsats i biblioteks- och informationsvetenskap vid Biblioteks- och Informationsvetenskap/Bibliotekshögskolan 2006:66). https://dspace.hb.se:8443/dspace/bitstream/2320/1481/1/06-66.pdf

Kiel, E. (1994). Knowledge organization needs epistemological openness: A reply. Knowledge Organization, 21(3), 148-152.

McIlwaine, I. C. (2003). Trends in knowledge organization research. Knowledge Organization, 30(2), 75-86.

Spang-Hanssen, H. (1974). Kunnskapsorganisasjon, informasjonsgjenfinning, automatisering og språk. In: Kunnskapsorganisasjon og informasjonsgjenfinning. Oslo: Riksbibliotektjenesten, pp. 11–61. Core Concepts in LIS/Spang_Hanssen_1974.pdf

White, H. D. & McCain, K. W. (1998). Visualizing a discipline: An author co-citation analysis of information science, 1972-1995. Journal of the American Society for Information Science, 49(4), 327-355.

Ørom, A. (2003). Knowledge Organization in the domain of Art Studies - History, Transition and Conceptual Changes. Knowledge Organization, 30(3/4), 128-143.