The creation and display of data records for a database comprises a process that involves a sequence of steps requiring both semantic and systems analysis. An essential early step in this process is the choice, definition, and naming of data elements. In the explication of a data element, the specification of attributes is fundamentally a terminological activity. This paper presents a systematic description of data elements as a three-stage process.
The attributes and the values of a data element, which provide the full meaning of the element, remain associated with them from their birth as a concept to a generic data element that serves as a template for final application. Therefore, terminology, as the methodology of concepts, is centrally important to the entire data management process. For a data element to be readily comprehended, smooth mapping from natural language to a database is needed, and, consequently, terminology standardization is needed from the outset of database work.
In this paper the semantic aspects of data elements are analyzed and discussed. Seven kinds of data element concept information are considered, and those that require terminological development and standardization are identified. The four terminological components of a data element are the hierarchical type of a concept, functional dependencies, schematas showing conceptual structures, and definition statements. These constitute the conventional role of terminology in database design.
A different problem exists for more complex or more abstract entities that change in time. Complex entities such as “product” or “department” are comprised of sets of data elements associated with one or more databases. For such entities new data can be created by the very act of using existing data as well as normal additions to the data collection. The use of a taxonomic approach at the outset can offer aid to the database designer. Sound terminology obtained by standardization beginning with each data element concept, followed by compre-hensive definition of the data elements throughout the process, should provide appropriate links to a data system and promote definability of complex entities. The use of structured sets of data element types to define complex concepts and those that evolve (have new attributes added over time) is described in terms of a taxonomic approach.