Using Ontology Manager

Semaphore's Ontology Manager Makes Using Taxonomies Simple, Consistent and Accurate

Ontology Manager supports a taxonomy design, initial build, validate, enhance cycle.

Our belief is that while technology can assist in taxonomy development, there is no 'silver bullet' technology that can create an intelligent structure by a mix of natural language processing and clustering algorithms. Semaphore’s design goal is to make the taxonomy developer as productive as possible.

Use Existing Lists

The 'collect' phase illustrated in Patrick Lambe’s 'taxonomy development cycle' is served by:

• Multiple import mechanisms and the ability to copy and paste branches of one model (with all their notes, etc.) to another.
• Duplicate terms are not allowed by default

The 'categorize' phase is assisted by:

• The ease of creating terms by drag and drop from any application, a browser window for example.
• Child (NT) and related (RT) relationships can be easily and visually created.
• Terms can be converted between preferred and non-preferred.
• Cyclical BT/NT relationships cannot be built.  Poly-hierarchical structures can be created.

Different classes of information can be modeled in an ontology.  This can enable the creation of a complex domain model, or provide a mechanism to map between several taxonomy 'facets'.

Validate with Users

The 'validate' phase is supported by:

• A term status model of 'candidate, approved and rejected' with a separate 'approver' role.
• Multiple standard reports to allow
• A 'Review Tool' application that allows a published snapshot of the model to be reviewed in a standard browser and comments reported back to the development team.

Easily Browse Multiple Hierarchies or Search for Terms

The tool's logical interface lets non-Information Scientists easily browse multiple hierarchies or search for terms by name or any attribute (created date, modified user, has scope note, etc).  This allows a small team to maintain the model with minimal training required.