Categories of Controlled Vocabularies
Ironically, terminology about controlled vocabulariesA managed list of terms. In the context of vocabularies, management typically includes careful selection of terms, maintenance of terms over time (i.e. addition, deprecation, modification), and presentation of the vocabulary in an accessible format. Related Guide can be somewhat confusing. Much time has been spent understanding the differences between the categories of controlled vocabularies, and necessary conditions for each category. Of course, if you're reading this guide, you probably know this first hand. In general, there are three broad categories of controlled vocabularies: flat, multi-level and relational. Within these three categories, there are a variety of types of controlled vocabularies. (Click on image to view larger size.)
Flat Vocabularies
Authority File * Glossary * Dictionary * Gazetteer * Code List
All flat vocabularies contain a labelA descriptor for a metadata value. This can be thought of as a question to which the value is providing an answer. For example, for the metadata label "date", the metadata value could be "March 16, 2008". and a value. These lists match the acceptable valuesMetadata values are the content connected to metadata labels in a metadata element. For example, if the metadata label is "date", the metadata value could be "May 13, 2007". Related Guide with the appropriate metadataData about data. Metadata provides a context for research findings, ideally in a machine-readable format. It enables discovery of data via an electronic interface, and correct use and attribution of findings. Related Guide label. Some flat vocabularies will build upon this foundation, and add a definition or additional information about each value. These types of vocabulary really put the focus on the values only. No relationshipsConnections between metadata terms within a vocabulary. These relationships can connect terms by scope, provenance, or other well-defined criteria. are established, no hierarchies are setup, and no complicated matrices are necessary. Flat vocabularies are sets of three or four pieces of information - a label, a value, and possibly a definition and/or additional information.
Multi-Level Vocabularies
Taxonomy * Subject Heading
A multi-level vocabulary is essentially a way to group terms into classesGrouping of metadata values, based on shared criteria. Related Guide. A classification tells more about the terms by placing them into well-thought-out categories.
Think of a classification as a tree - start with a trunk, and follow the trunk to a limb, then follow the limb to a branch, then follow the branch to a leaf. If you look at an individual leaf on the tree, you can back-track to the branch, which will lead you to the limb, which will eventually lead you to the trunk.
In a multi-level vocabulary, you can examine in which category a term belongs, and you can examine the relationships between categories as well. In some Multi-Level Vocabularies, the only connection between categories is a narrower-than / broader-than comparison (taxonomy). In others, you can compare like categories across broader categories (subject heading).
Relationship List
Thesaurus * Semantic Network * Ontology
Relational VocabulariesManaged list of acceptable terms that makes use of relationships between metadata terms. Relational vocabularies include thesauri, semantic networks and ontologies. Related Guide contain a mechanism to connect terms. The relations are prescribed by various standards and protocolsA strategy for transmitting data between systems. A protocol can be used not only over the internet, between computers, but also between applications running anywhere. Examples: FTP, SNMP, SSH. . There are a variety of relationships described for thesauriA type of relational controlled vocabulary which provides a list of terms, with specific relationships between the terms. Related Guide in the ANSIAmerican National Standards Institute /NISONational Information Standards organization Z39.19 - 2005 standard, including broader than, narrower than, used for, and related. In a relationship list, there are critical connections that are made in a standard way.
Think of a relationship list as seashells on the coast. You might find one on the San Diego coastline that looks exactly like one in Monterey Bay, except the San Diego shell has not been degraded by extreme waves. These two are probably related (similar to broader than). Or, perhaps you see a shell on the Oregon coast, and it looks similar to a shell in Woods Hole, Massachusetts. The two aren't exactly the same though, because they're different colors. There is still a relationship to be made (similar to related). Perhaps you find a message in a bottle in Maryland. It's originally from Europe, and there's a seashell inside. In this case, you would want to relate the shell to other European shells (similar to used for).