Use in thesaurus and taxonomy development
Classification schemes and thesauri could be used beneficially to develop organisational taxonomies. This is demonstrated by Wang et al. (2008), who developed taxonomy in the information studies domain for the Division of Information Studies at the Nanyang Technology University, Singapore. This taxonomy was built by using the Dewey Decimal Classification, the Information Science Taxonomy, two information systems taxonomies, and three thesauri (ASIS&T, LISA and ERIC). The classification and thesauri proved useful in creating structure and categories related to the subject of the taxonomy. To get more details about the procedure of building taxonomy based on classification, it is worth referring to this 34-page article in the Journal of Documentation.
Đang xem: What is classification schemes
Depth classification schemes structure and list almost all micro aspects of the concerned subject. Speciators are the qualifiers added to isolates so as to make them semantically rich. Kumbhar (2005) discussed the advantages of using a speciator-based faceted depth classification scheme for the construction of a thesaurus. Use of a faceted classification scheme particularly helps in deriving related terms for the thesaurus. LIS professionals have always shown interest in combining the various vocabulary control tools and/or their principles so as to produce a new, more efficient vocabulary control tool. Jean Aitchison’s thesaurofacet is one such example. Realising the benefits of integration of knowledge organisation tools and as a result of the influence of the Western classification theory, especially Aitchison’s Thesaurofacet, China developed the Classified Chinese Thesaurus. This is developed by integrating classification and subject headings in China. Liang (2007) discusses the history and background of this integration effort.
View chapterPurchase book
In Classification in Theory and Practice (Second Edition), 2014
Enumerative and faceted classification
Classification schemes basically fall into two types – enumerative and faceted. However, major schemes do not fall neatly into one category or the other, rather they fall somewhere on a continuum between the strictly enumerative and the strictly faceted. It is useful to examine the qualities of enumerative and faceted classification by outlining the major classification schemes in the context of their place on the continuum.
An enumerative classification scheme attempts to enumerate, or list, all subjects. There are obvious problems associated with this. Apart from the difficulty of listing everything and the resulting size of the publication, in a strictly enumerative scheme the schedule (listing of subjects) will be very long. Another problem is that subjects change and new subjects emerge that could not have been anticipated when the scheme was devised. Enumerative schemes that attempt to list all fields of knowledge accommodate new subjects by leaving unused notations here and there to fit those new subjects into. However, it is clearly very difficult to place each new subject in its proper position among the existing subjects – the gaps in the schedule may be in the wrong places.
The most important example of an enumerative scheme is the Library of Congress Classification (LCC) which is examined in detail in Chapter 2. LCC is an exceptionally popular scheme, very widely used in the United States. An attractive property of LCC is that it is relatively easy to use because the classifier simply chooses notations for topics from a comprehensive list of subjects, dispensing with the need for in-depth subject analysis.
Another example of an enumerative classification scheme is Dewey Decimal Classification (DDC), also examined in depth in Chapter 2. DDC, however, possesses features that place it closer to faceted schemes on the continuum. DDC’s schedule enumerates or lists both basic and compound subjects, but the notations for many more compound subjects can be constructed using notations from other parts of the schedules or with the aid of the tables that list common concepts such as geographical area and language.
Universal Decimal Classification (UDC), examined in depth in Chapter 3, is, at first glance, very similar to Dewey, from which it was developed. There are, however, important differences between the two schemes. UDC incorporates more features of faceted classification and allows for much more detailed classification than Dewey. As well as using tables of common concepts and lists of special, subject-specific, concepts to build notations, UDC allows for a notation to be built by linking the notations for independently classifiable concepts. UDC provides the classifier with much more freedom and flexibility in linking concepts than DDC. UDC was initially designed for classification of metadata (bibliographic records), rather than for classification of books, but it was quickly adopted for use in more specialised libraries; it provides for a greater minuteness of detail because of the flexible way in which subjects can be linked together.
Faceted classification schemes do not attempt to list, or enumerate, compound subjects. Instead, classmarks are constructed using notations for basic subjects together with notations for common and subject-specific concepts, often termed common and special isolates. This type of classification scheme is properly called analytico-synthetic, the name reflecting the two major operations involved in its use – analysis of subject and synthesis of notational elements to fully express that subject. The schedules of a faceted scheme can be quite short, because there is no attempt to list every topic. Faceted schemes are much more flexible and allow for much greater precision in expressing complex subjects than enumerative schemes and they are therefore more suitable for specialised library collections. It is relatively easy to construct a faceted classification scheme for a special collection and this is explored in Chapter 3. For the classifier faceted classification requires more thought. Enumerative schemes provide a list from which a notation can be selected, whereas faceted schemes usually require a notation to be constructed, but the faceted approach obviously provides much more depth, freedom and flexibility. An example of a faceted classification scheme is Colon Classification (CC), which was developed by Ranganathan, whose influence on library classification is examined in Chapter 3. Another example of a faceted scheme is Henry Bliss’s bibliographic classification (BC).
|Arithmetic specialization||Plug-in applic.-spec. unit||Viterbi ALU||Dual multipl.-accumulator|
|Data type||Fixed point||Fixed point||Fixed point||Fixed point|
|Memory structure||Harvard with two data||Harvard with two data||Harvard with two data||Harvard with four data|
|Address. modes with||Address. modes with||Address. modes with||Address. modes with|
In Classification in Theory and Practice (Second Edition), 2014
Faceted classification schemes designed for use in general libraries (CC and BC) are unlikely to provide a useful classification for specialised collections: they tend to be rather complicated and are not revised frequently enough to keep pace with new areas of knowledge. A practical alternative for a library wishing to organise a special collection of materials would be to create a new classification scheme based on faceted principles.
Such a scheme can be devised relatively quickly and will be tailor-made to meet the demands of the subject area. Choices can be made to ensure the scheme meets the needs of both the collection and its users. If a scheme is to be devised for a children’s collection, notation can be designed for brevity, simplicity and literal mnemonic qualities, for example. If a scheme is to be designed to meet the needs of expert users, then depth of classification can be emphasised.
Rajendra Kumbhar, in Library Classification Trends in the 21st Century, 2012
Methods of constructing classification schemes
Construction of a classification scheme is a complex and intellectual task. It involves many steps and also needs to apply various theories and principles. Amongst them, sameness and difference are important guiding principles (Olson, 2001). Zins (2004) explored the epistemological foundations of knowledge organisation and discussed the implications of knowledge organisation for the development of classification schemes and knowledge maps. Thomas (2004) provided a useful reading in classification, facet analysis as well as thesaurus construction, including deconstruction and reconstruction of sections of classification systems and thesauri. The literature reviewed in this section deals with various approaches and methods and rationality for constructing classification schemes. It particularly deals with the use of postulates, formation of hierarchies, professional and naive classification schemes, design of work-centred classification schemes, use of circulation logs and speciators, planes of work, etc.
A classification scheme must have a useful structure. Morphology for the design of a classification structure is given by Gopinath (2001b), who further explained the usefulness of a lexicographic approach to the design and development of a classification scheme. According to Gopinath, while designing a universal classification scheme, one has to keep in mind the 10 qualitative requirements, i.e. simplicity, brevity, freedom from bias, ease of updating, flexibility, ability to regenerate, ability to educate, representation, portability and philosophy.
Pierre de Keyser, in Indexing, 2012
Libraries use classification schemes to order documents. These produce a systematic classification of documents which are similar or which differ only slightly from each other. A researcher can browse the stacks (or an electronic version) and make discoveries which are not possible in the way in which electronic documents are ordered nowadays. Moreover, a search engine confronts us with so many thousands of results that we quickly lose all courage to browse them.
Thomas Mann gives the following example: he was once looking for the title of the first book published in French and couldn’t find the answer in Google. He then went to the bookshelves hoping to find a certain catalogue of French books, which was missing; the book next to its empty place on the shelf attracted his attention. He took it from the shelf and found in it the information he was looking for .
In Classification in Theory and Practice (Second Edition), 2014
Like any classification scheme that aspires to cover all knowledge, DDC has its faults. Coverage in some areas, such as religion, even after revision in the current edition, demonstrates an antiquated bias reflecting the cultural context within which the scheme was created. Class numbers can be very lengthy, even for quite simple topics. Its rigid structure and preponderance of rules can seem overly restrictive at times. Newer disciplines such as computing and electrical engineering have not been accommodated very successfully. Interdisciplinary topics such as women's studies and media studies are unhelpfully dispersed across the schedules (and across the library shelves).
Nevertheless, DDC is deservedly very widely used in public and academic libraries. In general it provides a well-structured and helpful arrangement of materials on the library shelves. It is a very familiar scheme, and one that is quite easy for the library user to grasp. Its hospitality can be criticised, but new subjects are accommodated. Regular revisions of the hard-copy version and access to the very latest changes to the scheme via WebDewey ensure that the scheme keeps pace with new areas of knowledge.
Elias Sanidas, Stéphane Carlier, in Intravascular Ultrasound, 2020
A tissue classification scheme based only on the analysis of the integrated backscattered (IB) signal, using a simple surface scanner on carotid samples was primarily described in 2001.58 This methodology was developed with the integrated, rotating, 40-MHz IVUS catheter from Boston Scientific (Fremont, California, United States). IB-IVUS was applied to 18 samples of coronary artery and the results were compared with the corresponding histological findings. The resulting IB-IVUS values were divided into five categories so that coded color maps could be constructed: thrombus, intimal hyperplasia or lipid core, fibrous tissue, mixed lesions, and calcification. The initial comparisons between angiography and IB-IVUS showed that the angioscopically colored surface of the plaque reflected the thickness of the fibrous cap more than the size of the lipid core.59 Initially developed with the Boston Scientific Clearview system, IB-IVUS (IB-IVUS, YD Co, Ltd., Nara, Japan) is now used with the VISIWAVE platform from Terumo (Tokyo, Japan). With the ViewIT 40-MHz catheter (Terumo, Tokyo, Japan), a comparison of IB-IVUS with histopathology shows that the sensitivity in the classification of calcification, fibrous tissue, and lipids was 90%, 84%, and 90%, while specificity was 97%, 96%, and 86%, respectively.60
Data Classification Scheme (SOX)
Establish a structure for the classification of data (i.e., categories, security levels, etc.). Although this should be done, depending on your company it might be a monumental task. Therefore, keep in mind the scope of what you need to accomplish for SOX–systems that are materials in the financial reporting process.
Security Levels (SOX)
Security levels should be defined and maintained for any security access above general access, such as systems that are material in the financial reporting process.
As mentioned above, models are represented as first-order logic τCBSD-statements; these are statements that use relation symbols from τCBSD among others. This means that additional relation symbols can be defined and used if necessary. As an example of a statement representing a model, consider the UML design model cutout of Common Component Modeling Example (CoCoME) depicted in Figure 7.8. It shows that the component StoreGUI requires the interface StoreIF. This informal statement can be translated to a first-order logic statement over τCBSD-structures:
whereas ex is the entity representing the element x in the model. In the following, we will denote with ϕM the set of τCBSD-statements for a model M. Logical statements are evaluated over relational structures as introduced, and we will denote by S(ϕM) the set of systems that fulfill the set of formulas ϕM.
22.214.171.124 Classification of models
According to the classification scheme by Eden et al. (2006), we can distinguish architectural models, design models, and implementations by their sets of statements and properties of their sets of satisfying structures. A statement is called extensional if and only if each of the structures satisfying the statement is closed when addition and removal of elements. This means that
adding entities or tuples containing new entities to the structure and
removing entities or tuples that are not directly referenced in the statement
lead to structures satisfying the statement as well.
Statements that are not extensional are called intensional. Consider the statement examples in Figure 7.3. The depicted structure satisfies them all. Only the leftmost example is extensional, and neither adding entities nor removing other entities than C1 or C2 from the structure will change the evaluation of φ1. φ2 would evaluate to false if all interfaces were removed; φ3 would evaluate to false if we added a component providing no interface.
Moreover, we call statements that are at least closed under addition local; such statements claim properties of systems that, once established, cannot be lost. Statements that are not local are called non–local; a system that satisfies such a statement can in general be extended in a way that the resulting system does not satisfy the statement any more. In Figure 7.3, only φ3 is non-local.
Eden argues that statements being both intensional and non-local are those defined by architectures; such statements define constraints that are “system-global” constraints in the sense that their impact cannot be locally isolated. For example, the layering of a system is non-local; although a system may be correctly layered at some point, a component could be integrated and related to other components such that dependencies would form a cycle and the layering would be violated. We call constraints like that architectural rules. In contrast, consider two classes/components realizing an observer-observable relationship. The constraint about what makes them a “valid” observer or observable is local to the two components; it describes the interaction between these two alone, which stays the same if further components are added to the system in any case (see Eden et al., 2006 for details).
Hence, we define two partitions ΦM ≔ ϕMext ⨃ ϕMint for the set of statements for a model M, which contain the extensional and intensional statements of M, respectively. Furthermore, we call a model M(a) architectural if its set of intensional statements contains non-local statements (architectural rules), (b) a design model if its set of intensional statement contains only local statements, and (c) an implementation if its set of intensional statements is empty.
126.96.36.199 Transformation of models
The transformation of models into τCBSD expressions can be specified by transformation definition depending on a model's meta-model. Formally, we can express the transformation definition TL for a given meta-model L as
whereas ClassL denotes the set of meta-classes defined in L and EXTτL denotes the set of extensional first-order logic expressions containing navigable expressions over τL ⊆ τCBSD and INTτLthe set of intensional statements. Navigable expressions mean that the structure of the meta-model can be exploited to refer to objects in the model that has to be transformed. For example,
that for each component in a UML model according tuples of providesInterface and requiresInterface are generated for each interface that is referred to in the model by this.
provides (whereas “this” refers to the current component, see also OMG (Object Management Group), 2010).