The defining characteristic of taxonomies is the presence of hierarchical relationships. Information specialists recognize several types of hierarchical relationships.
Generic relationship – As explained in ANSI/NISO Z39.19-2005 (page 47), “This relationship identifies the link between a class and its members or species.” The generic relationship is generally called the Broader term/Narrower term relationship. In most cases, it’s easy in traditional biological taxonomy of organisms, because of its well-established and well-known groupings of phyla, genera, families, species, and so forth, i.e., Rodents and NT Squirrels.
It is sometimes referred to as an “isA” relationship, in reference to the kind of relationship that the narrower term has to its broader term. A squirrel “is a” rodent.
Instance relationship – This is also an “isA” relationship. The difference between this kind of relationship and the generic relationship type is that in an instance relationship, the narrower term represents a single instance in the category identified by its broader term. In other words, the narrower term is a class of one. An example of this relationship follows:
Seas
NT Baltic Sea
NT Caspian Sea
NT Mediterranean Sea
In the NISO/ANSI standard, the narrower terms are called Narrower Term Instances, or NTI. Some software uses “NTI” to indicate the narrower terms in this kind of relationship, although this differentiation is of limited usefulness for most applications.
French cathedrals
NTI Chartres Cathedral
NTI Rheims Cathedral
NTI Rouen Cathedral
NT Gothic cathedrals
The instance relationship is a way to accommodate specific people, places, and things into the most appropriate place in the hierarchy.
Whole-part relationship – The name of this relationship type is somewhat self-explanatory. As described in Z39.19-2005, “This relationship covers situations in which one concept is inherently included in another, regardless of context, so that the terms can be organized into logical hierarchies, with the whole treated as a broader term.” (Z39.19-2005, page 49)
Three types of terms that Z39.19 identifies as often involving whole-part relationships are 1) systems and organs of the body; 2) geographic locations; and 3) hierarchical organizational, corporate, social, or political structures. Many other types of terms can be in whole-part relationships, as well. In a sense, a sub-discipline of an academic discipline could be considered to have a whole-part relationship with the broader discipline, although this could also be viewed as a generic relationship.
Body systems and organs
Ear –> Middle ear
(Note: Taxonomy terms for body systems and organs are customarily given in singular form, rather than the usual plural form. This seems more natural, probably because of the more generic nature of the concepts being represented. Here, the actual concept is “the middle ear” in an anatomical sense, rather than the collective world of people’s and creatures’ middle ears.)
Geographic locations
Bernalillo County –> Albuquerque
Hierarchical social etc. structures
Cricket leagues –> Cricket league divisions –> Cricket teams –> Cricket players
Historical or geochronological time spans
Mesozoic era
–> Triassic period
–> Jurassic period
–> Cretaceous period
Fields of study
Geology –> Physical geology
The abbreviations BTP [broader term (partitive)] and NTP [narrower term (partitive)] are sometimes used to identify the terms of a whole-part relationship.
Polyhierarchical relationships – These are also referred to as “multiple broader terms.” As explained in ANSI/NISO Z39.19: “Some concepts belong, on logical grounds, to more than one category. They are then said to possess polyhierarchical relationships. . . . These relationships may be represented using the BTG and BTP notation.” (pages 49-50)
Nurses
–>Nurse administrators
Health administrators
–> Nurse administrators
Careers
–> Accounting
Finance
–> Accounting
Polyhierarchy is a relatively recent development in the world of classification. It takes advantage of computer capabilities to go beyond the physical limitations of the non-digital library or museum collection. It is also a necessity for the adequate indexing of digital objects in large databases.
Marjorie M.K. Hlava, President, Access Innovations
| Data Harmony is an award-winning semantic suite that leverages explainable AI. |
|
This posting is one of a series based on a workshop, “Thesaurus Creation and Management,” that Marjorie Hlava presented in December of 2012.
