This is a brief introduction to the relationships between taxonomies, thesauri and ontologies, and similar \u00c2\u2018things\u00c2\u2019. It doesn\u00c2\u2019t contain definitive, scientific definitions, it is a personal interpretation of some fairly complex structures. It aims to give you a fairly clear what these \u00c2\u2018things\u00c2\u2019 are, so lay readers will find more easy information.<\/p>\n
2.\tControlled vocabulary: this is the only collective name for these \u00c2\u2018things\u00c2\u2019 that I can find. It\u00c2\u2019s not exactly sharp and snappy, but it does the job. A controlled vocabulary is list of terms or headings, each one having an assigned meaning. It is designed for classifying, indexing and searching for information resources. Essentially, if everyone uses the same name for the same concept, things become much easier to find.<\/p>\n
3.\tList: in this context, a straightforward list of words, in some logical order, usually alphabetical.<\/p>\n
4.\tTaxonomy: a structured list, or \u00c2\u2018tree\u00c2\u2019, formed into a hierarchy with broader terms at the top. Taxonomies were once used by biologists to classify living things into species, genera, families, etc. Ideally, each item (or taxon) in a taxonomy should be mutually exclusive and unambiguous, so if \u00c2\u2018bats\u00c2\u2019 appears in one place referring to flying mammals, it shouldn\u00c2\u2019t turn up in another place referring to sports equipment. For each term the broader and narrower terms are indicated, and related terms can also be included.<\/p>\n
5.\tOne of the problems with new technology is that its complexity stretches the use of language almost beyond recognition. When web designers needed to map out the structure of a web site, they called these maps \u00c2\u2018taxonomies\u00c2\u2019. However the pages on a site are interrelated in many ways, and can usually be reached from many other places, so the structure was more complex than a traditional taxonomy allowed. On top of this \u00c2\u2018taxonomy\u00c2\u2019 is now used to describe almost any type of controlled vocabulary. This is where much of the current confusion arises.
\nA traditional taxonomy \u00c2\u2013 each item exists only once, and in one place. Relationships are vertical.<\/p>\n
6.\tPolyhierarchical taxonomy. Who needs buzzwords when you can casually drop a \u00c2\u2018polyhierarchical taxonomy\u00c2\u2019 into a discussion? Half way between a traditional taxonomy and a fully-fledged thesaurus, it\u00c2\u2019s a taxonomy with additional dimensions, in particular an item can appear in more than one place, so you can reach a given taxon by different routes (hence \u00c2\u2018polyhierarchical\u00c2\u2019).<\/p>\n
7.\tThesaurus: not wanting to be outdone by the techies, librarians borrowed the word \u00c2\u2018thesaurus\u00c2\u2019 from Roget and his ilk, and used it to refer to structured sets of the terms used to index information. However, the original concept, a set of synonyms used to make you look more literate, is still the most widely used meaning of \u00c2\u2018thesaurus\u00c2\u2019. This does nothing to reduce the confusion surrounding controlled vocabularies.<\/p>\n
8.\tA thesaurus is taxonomy with extras, it shows lateral connections (\u00c2\u2018related\u00c2\u2019 and \u00c2\u2018see also\u00c2\u2019 terms), has an underlying index showing words that might spring to mind but which you shouldn\u00c2\u2019t use (\u00c2\u2018non-preferred\u00c2\u2019 terms), and tells you what you should use instead (\u00c2\u2018preferred\u00c2\u2019 terms). It can be polyhierarchical, and usually contains scope notes to indicate exactly what the term means. A thesaurus lists concepts \u00c2\u2013 the actual words used aren\u00c2\u2019t of paramount importance. While a taxonomy is designed to classify things, a thesaurus is designed to help you find the right words or phrases to describe what you are ultimately looking for. There is an ISO standard for thesauri, and creating a proper one is a major undertaking. The standard (ISO-2788) as well as national standards such as ANSI\/NISO Z39.19. Successive updates since then have led most recently to the publication of ISO 25964-1, Thesauri and interoperability with other vocabularies. They are an excellent place to go if you want the \u00c2\u2018official\u00c2\u2019 definitions of terms such as \u00c2\u2018taxonomy\u00c2\u2019 \u00c2\u2018thesaurus\u00c2\u2019 \u00c2\u2018paradigmatic relationship\u00c2\u2019 and \u00c2\u2018enumerated classification scheme\u00c2\u2019.<\/p>\n
Thesaurus. Behind-the-scenes notes show some of the additional features of a thesaurus. It gives a variety of related terms, and shows that \u00c2\u2018apples\u00c2\u2019 can be fruit (as in this entry) or computers.<\/p>\n
9.\tOntology: the new kid on the block, \u00c2\u2018ontology\u00c2\u2019 comes from the study of philosophy, and came to its current use by artificial intelligence workers. There is comfort to be gained from knowing that there is no definition that everyone agrees on.<\/p>\n
10.\tOntologies are designed to allow computers to really interact with each other. For an electronic \u00c2\u2018agent\u00c2\u2019 to crawl around the web to find the best holiday, or get you a great deal on a computer, information has to be marked up in such a way that machines can understand it. Remember at this point that, although your PC can calculate pi to the nth degree, it can\u00c2\u2019t tell the difference between a computer and a granny smith if they are both referred to as \u00c2\u2018apples\u00c2\u2019.<\/p>\n
11.\tAn ontology is therefore a thesaurus gone mad. It is more specific in defining a concept or item and its relationships. A dog will be a noun, and an animal, and a mammal, and possibly a pet. Instead of having \u00c2\u2018puppies\u00c2\u2019 as a narrower term of \u00c2\u2018dogs\u00c2\u2019, it might have puppies as offspring of dogs. Ontologies often cover all elements of metadata, not just other subject terms. For example, a document could be connected to a person by the relationship \u00c2\u2018created by\u00c2\u2019, this person may then connect to an organisation by the relationship \u00c2\u2018works for\u00c2\u2019; to an address by \u00c2\u2018lives at\u00c2\u2019 and other documents by \u00c2\u2018author of\u00c2\u2019. This can lead directly from one document to others written by the same person, or by others working for the same organisation that the author works for. Ontologies can be machine generated from good metadata. The semantic web will be built on ontologies.<\/p>\n","protected":false},"excerpt":{"rendered":"
Taxonomies, Thesauri and Ontologies This is a brief introduction to the relationships between taxonomies, thesauri and ontologies, and similar \u00c2\u2018things\u00c2\u2019. It doesn\u00c2\u2019t contain definitive, scientific definitions, it is a personal interpretation of some fairly complex structures. It aims to give you a fairly clear what these \u00c2\u2018things\u00c2\u2019 are, so lay readers will find more easy […]<\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"open","ping_status":"open","template":"","meta":{"footnotes":""},"_links":{"self":[{"href":"https:\/\/data.lawin.org\/wp-json\/wp\/v2\/pages\/96"}],"collection":[{"href":"https:\/\/data.lawin.org\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/data.lawin.org\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/data.lawin.org\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/data.lawin.org\/wp-json\/wp\/v2\/comments?post=96"}],"version-history":[{"count":0,"href":"https:\/\/data.lawin.org\/wp-json\/wp\/v2\/pages\/96\/revisions"}],"wp:attachment":[{"href":"https:\/\/data.lawin.org\/wp-json\/wp\/v2\/media?parent=96"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}