The Analytical-Synthetic Classification, also known as Facet Analysis/Classification (FA/FC) is the grouping and cataloguing of library information resources by breaking them into their constituents and subsequently resynthesizing them with consistence. Being the father of Indian librarianship, Ranganathan invented colon classification that has its basis on the principles of the analytico-synthetic classification. Ranganathan, in developing this classification, aimed at analyzing the first subject into its constituents, synthesizing the class number as ready-made ones are not assigned to fields or topics, and developing schedules with particular standard units. These standards corresponded to the standardized pieces of Meccano apparatus as the purpose of the colon is like the bolts and nuts of Meccano set. It is also worth noting that facet classification aims at combining the different classes in the scheduled separate units when assigning combinations and permutation. In other words, the postulates of facet classification, design the analytical synthetic classification scheme that provide phase, round, level and facet analysis.
In order for classification work to qualify as Analytico-Synthetic Classification, it has to be presented through the five stages mentioned next as proposed by Cleveland & Cleveland (2001). First, the subject must be analyzed into its entire phases as well as in phases so that the isolates in each facet can be identified. The second step is to rearrange the names of the above isolates so that they conform to the syntax of the classificatory language applied. The third stage is change of name of each isolate to translate to the standard terminology. In the fourth step, librarians translate the name of the isolate from every facet into the isolate number with the help of the facet schedule and applicable devices. Lastly, there is synthesis of the basic class number and the different isolate numbers into the already class numbers aided by connecting symbols. The work done in the first two three stages is an entire idea plane so that the fourth and fifth stages become the verbal and notational planes (Denton, 2003).
The Problems of Analytico-Synthetic Classification
Facet classification has two discernible weaknesses, both of which relate to the manner by which library users may interact with facet classification. For instance, it is quite difficult and time-consuming when to rank documents in their order of relevance because it is unclear how one should allocate the appropriate relevance across the different facets when a sets of documents origonate concurrently from multiple facets (Spiteri, 1998). In such a case, there is possibility that a particular document will be more relevant to one section and less relevant. However, the item remains significant to another separate section.
The other problem worth noting is that concurrent selections, which are infeasible to practice, can be made across separate facets. This happens, by way of example, when the set of the documents returned under one selection for one facet fail to intersect appropriately with that set of documents returned from a concurrent selection from a different related facet (Denton, 2003). This second weakness of the occasional infeasibility of facet analysis, as well the first problem with ranking documents in the order of their relevance, may be approached for resolution by application of the concept of Adaptive Matching that are discussed in the next section.
Along the same lines, Kawamura (2011) reckons that the influence of new factors on the scene of facet classification is responsible for the unresolved problems of analysis which include the problem of continuous updating to achieve an objective and currently relevant representation of the objective reality. The other problem recognized here is the facet classification’s lack of an appropriate organizing philosophical base for current research in classification to enhance the development of a system that is entirely open-minded and possesses utmost hospitality in concept capture, chain and array. Furthermore, this mode of classification currently lacks universality. The problem here is that almost all research papers and other research resources in libraries are described using a different mode of facet classification in every center when it comes to broad subject ordering, announcement, and storage. Instead, the case should be that a universal classification system is available for identity and common use with any national system around the world.
The other problem of information retrieval from resources classified using analytico-synthetic system is the difficulty in transforming the concepts and their relationships, owing to the difficult expression in the language of those documents. This is due to the lack of a standardized language in which the syntactic structures are simplified and the synonyms controlled (Barre, 2006). In this case, some researchers in library information science have proposed the utilization of a mechanized thesaurus based on consistent networks of related meanings in order to routinize and mechanize the translating techniques to convert ordinary language in different centers into a regularized language to enable universal coding of library information (Foskett, 2003).
Current Approaches in Facet Classification
Ranganathan’s Theory of Facet Classification
Ranganathan, the father of Indian librarianship, formulated three planes namely Canons, Postulates and Principles, in theorizing the concept of facet analysis that is still applied in librarianship to date across the world. The Canons are to follow specific rules of facet analysis, while the Postulates and Principles are designed to be executed in accordance with recommended procedures in the application of facet analysis (Leigh, 2008). Specifically, the ideal plane is comprised of fourteen Canons, thirteen Postulates, and twenty-two Principles, the Verbal Plane of four Canons, and the Notational Plane of nineteen Canons (Ranganathan, cited in Lancaster, 2003).
In order to avoid the problems of apparent infeasibility of some elements of facet classification, especially in the ranking of documents in the order of their relevance, it is advisable to divide each entity into its constituent elements according to the Canon of Differentiation. What should follow this is choosing of the appropriate facets into which to divide the constituent elements of each entity in accordance with the Canon of Relevance, which reflects the subject, purpose, and scope of the system of classification (Lancaster, 2003). The Canon of Relevance is followed by the Canon of Ascertainability that stresses the importance of choosing facets that are ascertainable. This is followed by the Canon of Permanence where Ranganathan explains that facets used should continue being used without alteration (Wild, Giess & McMahon, 2009). The Canons are then differentiated into the corresponding planes mentioned earlier.
On completion of classification into Canons, the entities are then classified into the various Principles. The Principles for Helpful resource organize entities into foci according to their order in the array. The Principle of Later-in-Time comes next to arrange items in a progressive sequence of time or the Principle of Later-in –Evolution that states that items belonging to different evolution stages should be ordered in their proper sequence of evolution (Wild et al., 2009). In the same beat, items are also classified by two Principles of Quantitative Measure describing them as increasing or decreasing in quantity.
An illustration to demonstrate the use of the Principle of Quantitative Measure is that the Principle of Increasing Quantitative Measure would arrange foci as Prime Minister at the top, Executive Party next and the Public last. On the other hand, the Principle of Decreasing Quantity would arrange them in the reverse order. The arrangement of the complexity of the items is best when done in a simple-to-complex manner such as Geography-Physical Geography-Mathematical Geography. The final aspect of classification by Principles is the Principle of Alphabetical Sequence that suggests arrangement of foci in a pure alphabetical order so as to capture all items especially where no other logical foci sequence is available (Wild et al., 2009).
As far as the Postulates are concerned in facet analysis, Ranganathan designed fifteen Postulates to direct the citation and choice of fundamental facets. The first Postulate is that of Five Fundamental Categories that hold the assumption that all subjects can be categorized into five fundamental divisions namely Personality, Matter, Energy, Space, and Time (PMEST) (Foskett, 2003). The Postulate of Basic Facet holds that, each compound subject possess a basic facet such as the case of Agricultural Diseases where the basic facet is Animal Husbandry for the subject of Caring for Cows. In the same fashion, the Postulate of Isolate Facet suggests that each facet of a compound subject manifests one and only one of the aforementioned fundamental categories (PMEST).
Still on the Postulates, those Postulates that govern the Rounds of Manifestation and the Levels of Manifestation concern the order of citing or scheduling facets (Spiteri, 1998). However, Lancaster (2003) notes that these two are the areas of significant difficulty in understanding and applying because the rather inflexible and arbitrary approach to organizing classification schedules might not clearly apparent to everybody but Ranganathan himself. Moreover, the Postulates of the Rounds for Energy, Rounds for Personality and Matter, and Rounds for Space and Time are based on the assumption that the fundamental categories follow the P-M-E-S-T sequence (Spiteri, 1998).
The Postulate of Level states that the any of the PMEST categories may be put more than once in the same round. In other words, the Postulates for Facet Sequence (First Concreteness, Facet within a Round, and Facet within Last Round) dictate the order of arrangement of facets within one round (Hassanien & Abraham, 2009). Notwithstanding, there is the danger of one being overwhelmed by the intensity of the information given in Ranganathan’s concept. Suffice it to remark that the Postulates of Levels and Rounds are not easy to comprehend and apply without giving the concept the deserved attention.
Syntax Facet: An Example
It has been mentioned above that a category may manifest itself more than once in the same facet. This is seen to be accounted for by the postulate entailing levels and rounds. A good example of this is the classification of Literature in which category of Personality (P) appears four times in the categories of Author, Language, Work and Form, each of which are at a different level. In another example of the subject of Management of Human Illnesses, both Illnesses and Management manifest in the Energy category and are also represented respectively in the first and second rounds of classification. Thus, categories are ordered in decreasing sequence of their concreteness within the facet formula, while the rounds of categories are ordered according to the Principle of dependency. This feature is even better elaborated by Ranganathan’s illustration of the Wall-Picture principle where common sense has it that there cannot be any mural in the absence of a wall, making the wall the first facet (Hassanien & Abraham, 2009).
Conclusions: Current Issues and Future Developments
The utilization of facet classification has long been used as a device for analysis of library resources as well as a way of assistance in the constructing controlled vocabularies up to the present time. It is also worth noting that facets are used as devices for searching and navigating through metadata in the authority files and Marc fields (Spiteri, 1998). An excellent example is the facets used in the North American Library catalogs that have invigorated the application of analytico-synthetic classification of resources. Moreover, facet and facet analysis concepts used as efficient tools for vocabulary construction have now been incorporated into the new ANSI/NISO Standards for Controlled Vocabularies (NISO, cited in Barre, 2006).
In consideration of the issues of difficulty and lack of universality in facet classification/analysis, it is noteworthy that researchers in library information science intensify their studies on feasible ways of making the concept even more easily comprehensive by probing into the intentions of Ranganathan who invented the mode of classification. It is also crucial that future studies be directed towards augmentation and testing of the principles of practice that are proposed in current works on the field (Leigh, 2008). Another important aspect of inquiry that needs to be focused on future works is the exploration of those tasks involved in the design and application of analytico-synthetic classification, which would necessitate utilization of longitudinal analysis and contextual inquiry over the origin and evolution of the process of facet design
Barre, K. (2006). The use of faceted analytico-synthetic theory. Champaign, IL: Kathryn La Barre.
Cleveland, D. & Cleveland, A. (2001). Introduction to indexing and abstracting. 3rd ed. Greenwood: Libraries Unlimited.
Denton, W. (2003). How to make a faceted classification and put it on the web. Retrieved from http://www.miskatonic.org/library/facet-web-howto.pdf
Foskett, D. J. (2003). Facet analysis. In M. A. Drake (Ed.), Encyclopedia of Library and Information Science, 2nd ed. (pp. 1063-1067). New York: Marcel Dekker.
Hassanien, A. & Abraham, A. (2009). Foundations of computational intelligence: Data mining. New York, NY: Springer.
Kawamura, K. (2011). Broad System of Ordering (BSO). Saitama: Koshigaya Hospital Library. Rerieved from http://arizona.openrepository.com/arizona/bitstream/10150/129413/1/Kawamura_BSO_bibliography2011.pdf
Lancaster, F. W. (2003). Indexing and abstracting in theory and practice. 3rd ed. Springfield, IL: University of Illinois.
Leigh, S. (2008). Grounded Classification: Grounded Theory and Faceted Classification. Library Trends, 47(2), 218-232.
Wild, P., Giess, M. & McMahon, C. (2009). "Describing engineering documents with faceted approaches: Observations and reflections." Journal of Documentation, 65(3), 420-445.
Spiteri, L. (1998). A simplified model for facet analysis: Ranganathan 101. Canadian Journal of Information and Library Science, 23(1/2), 1-30. Retrieved from http://iainstitute.org/pg/a_simplified_model_for_facet_analysis.php