Keeping documentation consistent across deliverables is frequently a hassle. This program is a method for extracting and utilizing the common documentation annotations and those more speci c to ontology design patterns from OWL les and rendering via LaTeX to provide a human readable, shareable PDF.
Documentation is a critically important component of any system. This especially holds true as the complexity of that system grows. As knowledge graphs (KG) grow both in size and complexity, and commensurately popularity, it is important to meet the demand for in-depth documentation of the KG's schema in order to promote its (re)use. This program is used to generate human readable documentation (in the form of a pdf) from the annotations that are traditionally used for documentation purposes within an OWL le.
Automatic documentation generators are not new (e.g., [
Commons License Attribution 4.0 International (CC BY 4.0)
1 See https://protege.stanford.edu/.
as the OPLaTab [
The current structure is the title page, populated with top level annotations such as dc:creator, a primer chapter talking about schema diagrams and axioms (authored by Pascal Hitzler and can be con gured o ), and then generates the following sections, when the appropriate annotations are present. Where appropriate, it will also attempt to nd gures, such as Schema Diagrams and examples, via the opla-sd:hasSchemaDiagram, opla-sd:hasShortcutDiagram, oplasd:hasExampleDiagram annotations.
{ Overview: is generated using the opla-cp: annotation. The overview section
describes the overall purpose of the ontology (or pattern).
{ Use cases: is generated using the opla-cp:hasScenario annotation. The
program currently expects the title of the use case to be separated from the text
of the use case by a newline character. Use cases are the context of expected
interactions between agents (e.g., users or programs) and the ontology.
{ Competency Questions: is generated using the
opla-cp:hasCompetencyQuestion annotation. The program will create an enumerated (alphabetized)
list of the competency questions for the ontology. Competency questions are
the fundamental interactions that constitute a use case.
{ Formalization: is generated using the opla-core:Axiom opla-core:isNativeTo
owl:Ontology annotation pattern. Currently we support the formalization
and explanation of 17 the axiom patterns that can be found in [
Additional information on the implementation, installation, and use of these tools (including CoModIDE) can be found in our online portal.2 The source 2 See http://daselab.org/content/modular-ontology-engineering-portal. (documentation-generator/), licensing (license.txt), and example execution(documentation/), containing two working patterns, for this demonstration can be found on GitHub.3 3
We expect to provide a live demonstration of the program's functionality in a tutorial-like format. We will provide several additional pre-annotated patterns beyond what is currently included in the online repository for this demonstration in order to show how di erent annotations result in changed documentation. In particular, we will walk potential users through the execution process of this program, customize the nal output, and illustrate integration with the surrounding tool-suite. 4
There are many additional directions that we may take this work. First and
foremost, we intend to incorporate additional insights from [
While this program represents a prototypical demonstration of the ability to generate useful, human-readable documentation from annotations, there are several ways to improve this process: for better code quality and improved, resultant machine-readable (and human-readable) documentation. For the former, by separating static portions of the documentation (e.g., the primer chapter or LATEX speci c formatting) For the latter, by outputting to HTML. Finally, we expect to support additional documentation annotations.
Acknowledgement. The authors acknowledge support by the National Science Foundation under Grant 2033521 A1: KnowWhereGraph: Enriching and Linking Cross-Domain Knowledge Graphs using Spatially-Explicit AI Technologies. 3 See https://github.com/KnowWhereGraph/documentation-generator.