The Role of Classification in Knowledge Representation and Discovery - 1

Library Trends, Summer, 1999 by Barbara H. Kwasnik

ABSTRACT

THE LINK BETWEEN CLASSIFICATION AND KNOWLEDGE is explored. Classification schemes have properties that enable the representation of entities and relationships in structures that reflect knowledge of the domain being classified. The strengths and limitations of four classificatory approaches are described in terms of their ability to reflect, discover, and create new knowledge. These approaches are hierarchies, trees, paradigms, and faceted analysis. Examples are provided of the way in which knowledge and the classification process affect each other.

INTRODUCTION

Developments in our ability to store and retrieve large amounts of information have stimulated an interest in new ways to exploit this information for advancing human knowledge. This article describes the relationship between knowledge representation (as manifested in classifications) and the processes of knowledge discovery and creation. How does the classification process enable or constrain knowing something or discovering new knowledge about something? In what ways might we develop classifications that will enhance our ability to discover meaningful information in our data stores?

The first part of the article describes several representative classificatory structures--hierarchies, trees, paradigms, and faceted analysis--with the aim of identifying how these structures serve as knowledge representations and in what ways they can be used for knowledge discovery and creation. The second part of the discussion includes examples from existing classification schemes and discusses how the schemes reflect or fail to reflect knowledge.

KNOWLEDGE, THEORY, AND CLASSIFICATION

Scholars in many fields, from philosophy to cybernetics, have long discussed the concept of knowledge and the problems of representing knowledge in information systems. The distinction is drawn between merely observing, perceiving, or even describing things and truly knowing them. To know implies a process of integration of facts about objects and the context in which the objects and processes exist. Even in colloquial usage, knowledge about someone or something is always expressed in terms of deep relationships and meanings as well as its place in time and space. To know cars means not only understanding car mechanics but also knowledge of the interplay of the mechanical processes and perhaps even factors such as aesthetics, economics, and psychology.

The process of knowledge discovery and creation in science has traditionally followed the path of systematic exploration, observation, description, analysis, and synthesis and testing of phenomena and facts, all conducted within the communication framework of a particular research community with its accepted methodology and set of techniques. We know the process is not entirely rational but often is sparked and then fueled by insight, hunches, and leaps of faith (Bronowski, 1978). Moreover, research is always conducted within a particular political and cultural reality (Olson, 1998). Each researcher and, on a larger scale, each research community at various points must gather up the disparate pieces and in some way communicate what is known, expressing it in such a way as to be useful for further discovery and understanding. A variety of formats exist for the expression of knowledge--e.g., theories, models, formulas, descriptive reportage of many sorts, and polemical essays.

Of these formats, science particularly values theories and models because they are a "symbolic dimension of experience as opposed to the apprehension of brute fact" (Kaplan, 1963, p. 294) and can therefore be symbolically extended to cover new experiences. A theory thus explains a particular fact by abstracting the relationship of that fact to other facts. Grand, or covering, theories explain facts in an especially eloquent way and in a very wide (some would say, universal) set of situations. Thus, Darwinian, Marxist, or Freudian theories, for example, attempt to explain processes and behaviors in many contexts, but they do so at a high level of abstraction. There are relatively few grand theories, however, and we rely on the explanatory and descriptive usefulness of more "local" theories--theories that explain a more limited domain but with greater specificity.

CLASSIFICATION AS KNOWLEDGE REPRESENTATION

How are theories built? How does knowledge accumulate and then get shaped into a powerful representation? There are, of course, many processes involved, but often one of them is the process of classification. Classification is the meaningful clustering of experience. The process of classification can be used in a formative way and is thus useful during the preliminary stages of inquiry as a heuristic tool in discovery, analysis, and theorizing (Davies, 1989). Once concepts gel and the relationships among concepts become understood, a classification can be used as a rich representation of what is known and is thus useful in communication and in generating a fresh cycle of exploration, comparison, and theorizing. Kaplan (1963) states that "theory is not the aggregate of the new laws but their connectedness, as a bridge consists of girders only in that the girders are joined together in a particular way" (p. 297). A good classification functions in much the same way that a theory does, connecting concepts in a useful structure. If successful, it is, like a theory, descriptive, explanatory, heuristic, fruitful, and perhaps also elegant, parsimonious, and robust (Kwasnik, 1992b).