A keyword is a finding aid used for identification, searching, and indexing purposes that records one specific term, key phrase, abbreviation, taxonomic structure, or other expression that is found within the text of a standards document, or implied (broader, narrower, etc.) from terms found within this text.
Content and Source
A keyword element (<kwd>) may contain, for example:
- a repetition of words found in the standards document,
- terms that are relevant to the standards document (broader terms, narrower terms, related terms) even if those words are not in the content, or
- taxonomic identifiers that can be used to find the standard in a search or provide semantic classification of the text by mapping a specific structure to a taxonomy.
Although similar in content to index terms, tagged keywords are part of the metadata of a standards document (inside one of the *-meta elements), while index terms (<index-term>) are scattered throughout the narrative prose of the standards documents to enable generation of an Index. Keywords may be displayed as part of the metadata, or they may be undisplayed metadata used only for searching.
Language in Keywords
Keywords are contained in <kwd-group> elements. There may be several <kwd-group>s, each of which can be identified separately by language, vocabulary, or taxonomy source. Since different languages could mean, for example, different writing directions, none of the individual keyword elements (<kwd>, <compound-kwd>, <nested-kwd>) take the @xml:lang attribute; that is reserved for the <kwd-group> element. This means that if there are keywords in several languages they should be in different <kwd-group>s.
For controlled vocabularies, two attributes can be used to link a keyword to its term source:
- @vocab-term — This attribute holds the canonical form of the keyword as expressed in the vocabulary named in the <kwd-group>. The content of the keyword element (<kwd>, <compound-kwd>, <nested-kwd>) might not be exactly the same as the canonical form, as the keyword element content might be a user-specific variant, for example, the term in another language.
- @vocab-term-identifier — This attribute holds a unique identifier and possible pointer to the specific term in the named vocabulary (typically a URI or DOI reference), but @vocab-term-identifier could be an item number or other system-specific identifier.
Models and Context
Any combination of:
- Text, numbers, or special characters
- Emphasis Elements
- Inline Display Elements
- <named-content> Named Special (Subject) Content
- <styled-content> Styled Special (Subject) Content
- Baseline Change Elements
INSPEC keywords added by SDO
... <kwd-group kwd-group-type="INSPEC" originator="IEEE"> <kwd>social networking (online)</kwd> <kwd>buyer's guides</kwd> <kwd>electronic commerce</kwd> </kwd-group> <kwd-group kwd-group-type="InspecFree" originator="IEEE"> <kwd>whiz list</kwd> <kwd>social networking sites</kwd> <kwd>online purchase</kwd> <kwd>e-commerce portals</kwd> <kwd>Web site users</kwd> <kwd>marketing strategy</kwd> </kwd-group> ...
Working Group keywords
... <std-meta> ... <kwd-group kwd-group-type="working group" vocab="uncontrolled"> <kwd>adoption</kwd> <kwd>assurance</kwd> <kwd>assurance case</kwd> <kwd>claim</kwd> <kwd>IEEE 15026-1 ™</kwd> <kwd>integrity level</kwd> <kwd>life cycle processes</kwd> <kwd>reliability</kwd> <kwd>software assurance</kwd> <kwd>software engineering</kwd> <kwd>system assurance</kwd> <kwd>systems engineering</kwd> </kwd-group> </std-meta> ...