Showing papers in "Knowledge Organization in 2006"

Knowledge organization : A new science?

Abstract: In ISKO's name, the term "Knowledge Organization" (KO) denotes already the object and the activity area significant for the existence of any science. Both areas are outlined and their specific contents shown. Also a survey of its special subfields is given. The science-theoretical foundation of Knowledge Organization as a new scientific discipline is based on the propositional concept of science. Within a universal system of the sciences, KO has been regarded as a subfield of Science of Science. Concludingly it is proposed to find the necessary institution for work in concerted effort of scientists, knowledge organizers and terminologists on the collection, definition, and systematization of concepts of all subject fields, utilizing the Information Coding Classification (ICC) as the necessary categorizing structure.

Folksonomies : (Un) controlled vocabulary?

Abstract: Folksonomy, a free-form tagging, is a user-generated classification system of web contents that allows users to tag their favorite web resources with their chosen words or phrases selected from natural language. These tags (also called concepts, categories, facets or entities) can be used to classify web resources and to express users' preferences. Folksonomy-based systems allow users to classify web resources through tagging bookmarks, photos or other web resources and saving them to a public web site like Del.icio.us. Thus information about web resources and online articles can be shared in an easy way. The purpose of this study is to provide an overview of the folksonomy tagging phenomenon (also called social tagging and social bookmarking) and explore some of the reasons why we need controlled vocabularies, discussing the problems associated with folksonomy.

Semantic interoperability and metadata quality : An analysis of metadata item records of digital image collections

Abstract: This paper is a current assessment of the status of metadata creation and mapping between cataloger-defined field names and Dublin Core (DC) metadata elements across three digital image collections. The metadata elements that evince the most frequently inaccurate, inconsistent and incomplete DC metadata application are identified. As well, the most frequently occurring locally added metadata elements and associated pattern development are examined. For this, a randomly collected sample of 659 metadata item records from three digital image collections is analyzed. Implications and issues drawn from the evaluation of the current status of metadata creation and mapping are also discussed in relation to the issue of semantic interoperability of concept representation across digital image collections. The findings of the study suggest that conceptual ambiguities and semantic overlaps inherent among some DC metadata elements hinder semantic interoperability. The DC metadata scheme needs to be refined in order to disambiguate semantic relations of certain DC metadata elements that present semantic overlaps and conceptual ambiguities between element names and their corresponding definitions. The findings of the study also suggest that the development of mediation mechanisms such as concept networks that facilitate the metadata creation and mapping process are critically needed for enhancing metadata quality.

Potential and prospects of taxonomies for content organization

Abstract: While taxonomies are being increasingly discussed in published and grey literature, the term taxonomy still seems to be stated quite loosely and obscurely. This paper aims at explaining and clarifying the concept of taxonomy in the context of information organization. To this end, the salient features of taxonomies are identified and their scope, nature, and role are further elaborated based on an extensive literature review. In the meantime, the connection and distinctions between taxonomies and classification schemes and thesauri are also identified, and the rationale that taxonomies are chosen as a viable knowledge organization system used in organization-wide websites to support browsing and aid navigation is clarified.

Records classification and functions : An archival perspective

Abstract: The design of records classification systems has been subject to various and often inconsistent approaches. Subject matters, record types and forms, structures and functions of the organization: archivists and records managers (whether consciously or unconsciously) have traditionally referred to any of these elements, or to a mix of some or all of them, when developing their classification tools. Only in recent times has the concept of function become central to the theory, method, and practice of records classification. I discuss advantages and limitations of the functional approach applied to records classification by showing how it has been interpreted in the literature of different countries.

Functionality and flexibility of traditional classification schemes applied to a Content Management System (CMS): facets, DDC, JITA

Abstract: Paper presented at the 2nd ISKO Italy-UniMIB meeting, Milan, June 9 2006: http://www.iskoi.org/doc/milano06.htm Different classification schemes may be used for web indexing. The authors analyze three weblogs (Biblioatipici, Letture and Andrea Marchitelli’s blog) to demonstrate that different contents may be classified using the appropriate scheme. Biblioatipici is a weblog about temporary workers in libraries and Italian documentation centres, indexed with a faceted scheme, home made by authors. Letture, a diary about reading and books, is indexed by DDC; finally, Andrea Marchitelli’s blog, a weblog about digital libraries and open access, is indexed by JITA, the scheme used for indexing e-prints in some different open archives. The three applications are presented starting from the most complex scheme (i.e. the faceted one) to the simple but least functional (JITA), passing through a traditional bibliographic classification scheme, the Dewey Decimal Classification. The analysis demonstrates that different web contents can efficiently be classified with different schemes. In particular, with the facets one (Biblioatipici) the indexer is able to generate the needed classes in the indexing phase. Moreover with a small number of facets and foci one can obtain an exponential number of classes. Finally, one of the advantages for the user is that the faceted scheme allows multiple accesses on the basis of different information requirements, in addition to being coherent and intuitive. DDC scheme can be more suitable than the first web pages or blogs performing a sort of digital library. In this case in fact the DDC scheme constitutes a consolidated classification standard, widespread in a huge quantity of libraries and the use of a different scheme might get confusion. JITA scheme, finally, can find a huge applicability in web pages (or blogs) in LIS field showing several advantages: it’s very simple and essentialy pragmatic, intuitive and coherent. It’s a conservative scheme because structurally closed, in fact it does not allow the classifier to insert new LCS’s categories. If, from a certain point of view, this is a disadvantage, vice versa this feature is really useful because get a complete matching between classes of all the web pages implemented with it.

UDC in subject gateways: experiment or opportunity?

Abstract: The article has been reviewed and accepted for publication in Knowledge Organization 33 (2006)

Ontologies, Different Reasoning Strategies, Different Logics, Different Kinds of Knowledge Representation: Working Together

Abstract: The recent experiences in the building, maintenance and reuse of ontologies has shown that the most efficient approach is the collaborative one. However, communication between collaborators such as IT professionals, librarians, web designers and subject matter experts is difficult and time consuming. This is because there are different reasoning strategies, different logics and different kinds of knowledge representation in the applications of Semantic Web. This article intends to be a reference scheme. It uses concise and simple explanations that can be used in common by specialists of different backgrounds working together in an application of Semantic Web.

The alternative model of classification of belles-lettres in libraries

Abstract: Several systems for the classification of fiction have been proposed to date, but experience shows that such classification partially depends on the culture and literary education in specific countries. Slovenian public libraries have traditionally used UDC as the only subject information on belles-lettres. Research has shown that users would prefer richer subject information. Therefore an attempt was made to enhance UDC by adding subject information, that should be helpful to users and librarians. The newly proposed Alternative Model system contains lists of verbal and alpha-numerical denotations for the basic groups of belles-lettres book material (main-genres: lyrics, drama, epics) and all other categorical criteria (language of the original literary work, literature to which the work belongs, genre, sub-genre) and half-categorical (accessibility of the content of literary works, origin within the periods of literary history, the century in which the literary work was written, the rhythm of the language). All these lists are available, but not included in this paper. The idea of the Alternative Model system is to show the possibilities of making the classification of belles-lettres in libraries more helpful, efficient and exact.

A methodology for noun phrase-based automatic indexing

Abstract: The scholarly community is increasingly employing the Web both for publication of scholarly output and for locating and accessing relevant scholarly literature. Organization of this vast body of digital information assumes significance in this context. The sheer volume of digital information to be handled makes traditional indexing and knowledge representation strategies ineffective and impractical. It is, therefore, worth exploring new approaches. An approach being discussed considers the intrinsic semantics of texts of documents. Based on the hypothesis that noun phrases in a text are semantically rich in terms of their ability to represent the subject content of the document, this approach seeks to identify and extract noun phrases instead of single keywords, and use them as descriptors. This paper presents a methodology that has been developed for extracting noun phrases from Portuguese texts. The results of an experiment carried out to test the adequacy of the methodology are also presented.

Governing the professional and intellectual potential of a modern organization : Sociologic approach

Abstract: Governing the professional and intellectual potential is an interdisciplinary field of scientific research using a systematic process of developing innovation technologies for transforming individual knowledge and specialists' experience in such a way that would apply the knowledge and experience to the processes, services and products offered by an organization to reach its strategic goals. From the technological standpoint, governing the professional and intellectual potential represents modeling, forming, using and developing the corporate system of governing the professional and intellectual potential. We consider structuring knowledge using this model rather valuable during the stage of forming the governance system of professional and intellectual potential. Understanding, i.e., explicit definition of these factors, would allow for constant observation of the behavioral trends and for organizing the activity in a way conducive for influencing the favorable change of these factors. In addition, the presence of the critical management factor (CMF) system enables one to check the significance of any activity (i.e., any processes within a company) against these factors.

Terminology structuring for learner's glossaries

Abstract: This study presents a methodology for compiling corpus-based learner's glossaries designed for non-specialist translators and Language for Specific Purposes (LSP) learners. The need for such bilingual microglossaries on subsections of a subject field for LSP teaching and translation purposes is emphasized in the Introduction. The concept"learner's glossary' is delimited among other types of terminological collections such as a specialised dictionary, a thesaurus and a term bank in Section 2 where the information categories in an entry are specified (keyterm plus translation equivalent(s), definition of keyterm and exemplary context, narrower terms with synonyms, definitions and translation equivalents; special phrases based on keyterm collocations with translation equivalents and exemplary contexts). Section 3 describes the principles and working methods of modern terminography as well as the source materials available for the glossary compilation. A hybrid term extraction technique is also described in that section and is used to extract the candidate terms with subsequent manual initial processing of the results. In Section 4 the theoretical grounds and methodology for analysing the conceptual relations in a terminological system are presented including the expert validation of the automatically extracted terms as a first phase in that process. Rationale for applying a lexico-semantic analysis in identifying collocational information on the keyterms is provided in Section 5 which is proposed to involve descriptions of the actantial structure of a keyterm for identifying verbal (T+v) collocations and of the paradigmatic and syntagmatic morpho-syntactic relations applicable to a keyterm. Finally, a model for structuring a learner's glossary entry is proposed in Section 6.

Improving information retrieval of subjects through citation-analysis

Abstract: Citation-chasing is proposed as a method of discovering additional terms to enhance subject-search retrieval by broadening and prioritizing the results. Subjects attached to records representing cited works are compared to subjects attached to records representing the original citing sources, and to the subjects yielded by chasing see-also references from the latter group of headings. Original citing sources were yielded via a subject-list search in a library catalog using the subject heading "Language and languages - Origin." A subject-search was employed to avoid subjectivity in choosing sources. References from the sources were searched in OCLC where applicable, and the subject headings were retrieved. The subjects were ranked first by number of citations from original sources, then by total citation-frequency. The results were tiered into four groups in a Bradford-like distribution. A similar rank and division was performed on the subjects representing the original citing sources, and those yielded by chasing see-also references. Both in terms of subject frequency and topic type, positive comparisons between citation-chasing and see-also references show a confirmation of different methods of yielding alternative subjects. Exclusive results suggest potential mutual complementary value among these different methods.