172
A. Gyrard et al.
“Parts of ontologies” refer to Ontology Design Patterns (ODPs) [47] and modular
ontologies [48] research approaches (not covered in this paper).
OQuaRE is a framework for evaluating the quality of ontologies [38] based on
the SQuaRE standard for software quality evaluation. A quality model and quality
metrics (structural, functional adequacy, reliability, operability and maintainability)
have been defined. The framework has been evaluated with units of measurement
ontologies. Future work of this paper highlights the needs of automated ontology
evaluation.
OntoQA [40, 41] assists ontology developers and users to determine the quality of
an ontology. It provides metrics to evaluate ontology design and instances. OntoQA
provides three ontology metrics: (1) Relationship richness: an ontology that contains
many relations other than class-subclass relations is more precious than a taxonomy
with only class-subclass relationships, (2) Attribute richness: the number of attributes
that are defined for each class can indicate both the quality of ontology design and
the amount of information of instance data, and (3) Inheritance richness: a good indication of how well knowledge is grouped into different categories and subcategories
in the ontology. OntoQA defines nine instance metrics: (1) Class richness for KB
Metrics is related to how instances are distributed across classes, (2) Average population for KB Metrics (average distribution of instances across all classes) indicates
the number of instances compared to the number of classes. It can be useful if the
ontology developer is not sure if enough instances were extracted compared to the
number of classes, (3) Cohesion for KB Metrics can be used to indicate what areas
need more instances to connect instances more closely, (4) Importance for class
metrics provides the percentage of instances that belong to classes at the subtree
rooted at the current class for the total number of instances, (5) Fullness for class
metrics is mainly used by an ontology developer interested in knowing how well the
data extraction was with respect to the expected number of instances of each class,
(6) Inheritance richness for class metrics indicates how well knowledge is grouped
into different categories and subcategories under this class, (7) Relationship richness
class metrics: measures how much of the properties in each class in the schema is
being used at the instance level, (8) Connectivity for class metrics explains which
classes play a more central role than other classes, and (9) Readability for class
metrics indicates the existence of human-readable descriptions in the ontology, such
as comments, labels or captions. From our point of view, is also really relevant for
doing automation with user interfaces.
OntoQA provides three kinds of evaluations: (1) evaluation-based validation, (2)
symbolic-based validation, and (3) attribute-based validation. OntoQA has been evaluated with three ontologies: SWETO, TAP and GlycO, and related datasets. As a
future work, the need for a web-based tool to automatically measure the quality of
the ontologies is explained. However, we did not find such tools available online that
we can reuse and integrate with other tools.
A. Gyrard et al.
“Parts of ontologies” refer to Ontology Design Patterns (ODPs) [47] and modular
ontologies [48] research approaches (not covered in this paper).
OQuaRE is a framework for evaluating the quality of ontologies [38] based on
the SQuaRE standard for software quality evaluation. A quality model and quality
metrics (structural, functional adequacy, reliability, operability and maintainability)
have been defined. The framework has been evaluated with units of measurement
ontologies. Future work of this paper highlights the needs of automated ontology
evaluation.
OntoQA [40, 41] assists ontology developers and users to determine the quality of
an ontology. It provides metrics to evaluate ontology design and instances. OntoQA
provides three ontology metrics: (1) Relationship richness: an ontology that contains
many relations other than class-subclass relations is more precious than a taxonomy
with only class-subclass relationships, (2) Attribute richness: the number of attributes
that are defined for each class can indicate both the quality of ontology design and
the amount of information of instance data, and (3) Inheritance richness: a good indication of how well knowledge is grouped into different categories and subcategories
in the ontology. OntoQA defines nine instance metrics: (1) Class richness for KB
Metrics is related to how instances are distributed across classes, (2) Average population for KB Metrics (average distribution of instances across all classes) indicates
the number of instances compared to the number of classes. It can be useful if the
ontology developer is not sure if enough instances were extracted compared to the
number of classes, (3) Cohesion for KB Metrics can be used to indicate what areas
need more instances to connect instances more closely, (4) Importance for class
metrics provides the percentage of instances that belong to classes at the subtree
rooted at the current class for the total number of instances, (5) Fullness for class
metrics is mainly used by an ontology developer interested in knowing how well the
data extraction was with respect to the expected number of instances of each class,
(6) Inheritance richness for class metrics indicates how well knowledge is grouped
into different categories and subcategories under this class, (7) Relationship richness
class metrics: measures how much of the properties in each class in the schema is
being used at the instance level, (8) Connectivity for class metrics explains which
classes play a more central role than other classes, and (9) Readability for class
metrics indicates the existence of human-readable descriptions in the ontology, such
as comments, labels or captions. From our point of view, is also really relevant for
doing automation with user interfaces.
OntoQA provides three kinds of evaluations: (1) evaluation-based validation, (2)
symbolic-based validation, and (3) attribute-based validation. OntoQA has been evaluated with three ontologies: SWETO, TAP and GlycO, and related datasets. As a
future work, the need for a web-based tool to automatically measure the quality of
the ontologies is explained. However, we did not find such tools available online that
we can reuse and integrate with other tools.