What’s taxonomy?
A taxonomy is a list of terms in a classification structure. Although the term “taxonomy”
is used rather indiscriminately lately to describe any sort of classification, a formal
taxonomy is typically expressed in a hierarchical structure (parent/child) which
emphasizes context. It organizes concepts, products, functions, or objects to make
them easier to find, identify, and study.
When taxonomies are coupled with metadata, the combination provides a robust
content management solution that not only aids in search and retrieval, but enhances
navigation and discovery. In any given metadata model, there will be some elements
that will need to be populated by a controlled vocabulary. The controlled vocabulary is
retrieved from the structured taxonomy. Each term inherits intelligence from the
taxonomy.
A taxonomy serves two main purposes:
- Harmonize different dialects or expressions. Example: a search on the term
courgette would return everything catalogued with zucchini or C. pepo..
- Connect and associate related concepts to each other. Example: cooking is
related to nutrition. If someone did a search on either of these terms, the
taxonomy would guide the user to other information tagged with the related term.
This concept can be used to manage media assets as well as processes. If applied to
an intellectual property, a taxonomy would assist in the navigation and “findability” of
modules such as episodes, segments or personal profiles, and relate these to the
images, graphics, copy, or streaming video relevant to the query.
What’s a controlled vocabulary?
In order to build a viable framework for a media asset management system, the
consistent use of descriptors is crucial.
A controlled vocabulary, or authority control, is an established list of terms from which
an indexer, cataloguer or documentalist may select when assigning descriptors to a
metadata element within a record representing an asset. This eliminates guesswork,
misspellings and other free associations when indexing digital assets. Consistency in
the use of language is only one, but very important, component in sharing and retrieving
digital assets.
What’s a metadata element?
Metadata models and schemas are a combination of “elements” or fields of information
that help describe a digital asset. Title, Media Type, and Date Created are common
metadata elements.
What does a taxonomy do?
The structure of the taxonomy describes what exists and how it is known. By evaluating
the placement of a term in the hierarchy or classification structure, a logic is inferred of
what kind of content the taxonomy supports. In biology, the classification of organisms
in a hierarchy from general to particular, genus/species for example, is fairly straight
forward. For creative content, however, the taxonomy reflects a culture and a conceptual
schema. This schema must provide sense and meaning in order to assist users in
posing a query, and then interpreting the information retrieved.
• Taxonomies normalize descriptive terms
• Taxonomies structure and organize concepts
• Taxonomies bridge dialects
• Taxonomies optimize metadata models
• Taxonomies offer pathways of discovery
• Taxonomies provide intelligence through context
Why do we need one?
Large organizations are discovering that information and content repositories exist in
various regional businesses and continue to grow in parallel with each other. They are
not, however, sharing common guidelines or content management standards which
result in a wide variation of how systems and digital assets are managed. As digital
content proliferates, the division between these silo repositories threaten to widen
eventually impeded any effort toward interoperability. If left unaddressed, regional
offices end up duplicating efforts and/or constructing ad-hoc solutions further
compounding organizational ineffectiveness.
Taxonomies help configure the global infrastructure by uniting businesses and partners
in their quest to find, track and exchange digital assets in a rapidly evolving new
marketplace. By sharing common metadata practices, utilizing a taxonomy to bridge
dialects and concepts, assets and information can be accessed and leveraged
efficiently. Content can be re-used, re-purposed, workflows streamlined, and
collaboration supported through common practices.
How many different kinds of taxonomies are there?
Taxonomy, defined loosely here, can be represented in various ways depending on the
content or knowledge base it is classifying. Some examples of taxonomy structures are:
- Authority lists
- Reference guides
- Synonym rings
- System maps
- Hierarchies
- Faceted classification
- Thesauri
- Ontologies
A faceted classification taxonomy is widely used in business circumstances because of
its flexibility, scalable qualities, and polyhierarchical ability.
What’s a polyhierarchy?
Polyhierarchy means that a single term or concept can reside in more than one
classification. This removes the constraint of locking a term under one heading
allowing more subjectivity in interpretation. This means that the term courgette could be
classified under vegetables, garden proliferates, and edible blossoms.
How do you build a taxonomy?
Building a taxonomy requires three major components. They are:
1. Metadata model or schema: A set of data fields for describing content to
enhance information access.
2. Controlled Vocabularies: An established list of terms from which an indexer or
cataloguer may select when assigning descriptors to content.
3. Application Profile / Technology integration: the formal representation of
metadata & vocabularies. Many media asset management systems can only manage
and advantage taxonomy through a third party application. As such, taxonomy, no
matter how well thought out, is useless unless it can be applied to the system.
|
More
Taxonomy Community of Practice
Taxonomy Community of Practice
Wiki Reference page
A Framework of Guidance for Building
Good Digital Collections
How to Find Anything
Organizing Knowledge: Joseph Busch
and Dave Clark on Taxonomies
Elastic List Nobel Prize Winners
|
