ontology editor built on Owlready2

each instance of the ontor class represents an individual ontology and provides support for:

• creating new, loading existing, and saving ontologies
• modifying ontologies:
  • import other ontologies
  • simply extract information such as axioms and class restrictions
  • insert classes, properties, instances, relations, and restrictions
  • insert general class axioms using a workaround for Owlready2
  • delete classes, properties, instances, relations, and restrictions but preserve the ontology's structure by reassigning subclasses and instances appropriately
• reasoning over ontologies and debugging by interactively deleting problematic axioms
• visualizing the entire ontology or selected parts thereof

ontor provides a tuple based syntax with JSON and CSV support for ontology editing to facilitate focusing on the ontology's content

• Python 3.9+
• install ontor using pip
  • from PyPI: pip install ontor
  • from GitHub, in editable mode: pip install -e .
• generate documentation via sphinx using the makefile in docs/: make html

the directory example/ includes a demo application inspired by Protégé's pizza example

in addition to class axioms, General Class Axioms (GCAs) can express more complex statements - the generic axioms are equivalented using helper classes
in the example, a uniform price of 5 is set for all pizzas with seafood toppings without making use of an explicitly defined class for these pizzas:\

[
  ["has_topping",null,"min",1,"seafood_topping",null,null,null,null,null,null,null,true],
  ["has_price",null,"value",null,null,"float",null,null,5,null,null,null,true]
]

this allows a reasoner to infer that the price for all instances of seafood_pizza as well as for the instance Another_pizza is 5

interactively debug an ontology
in the example: ontor3.debug_onto()

visualize selected instances, classes, and properties in a given radius around a focus node; e.g., all nodes in a radius of two relations around the node "John"
in the example: ontor3.visualize(classes=["human", "pizza"], properties=["likes", "diameter_in_cm"], focusnode="John", radius=2)

When creating ontologies from scratch, note that some functions have to be called in a specific order:

1. add_taxo - the taxonomy has to be created first to ensure that all classes are defined, which are required by the properties, axioms, and individuals
2. add_ops, add_dps - properties must be defined before axioms can be specified
3. add_axioms, add_gcas, add_instances - axioms and instances can only be added when all the necessary classes and properties have been defined

GPL v3.0

Felix Ocker - [email protected]