Processing Ontology for Tracing, Training and Extraction of Results (POTTER)
language en

Processing Ontology for Tracing, Training and Extraction of Results (POTTER)

Release: 2026-06-01 00:00

Latest version:
https://github.com/predict-idlab/potter/
Revision:
1.0.0
Authors:
Colin Soete
Bram Steenwinckel
Publisher:
Colin Soete
Imported Ontologies:
sosa
ssn
fno
Download serialization:
JSON-LD RDF/XML N-Triples TTL
License:
https://creativecommons.org/licenses/by/4.0/
Visualization:
Visualize with WebVowl
Cite as:
@unpublished{soete2026_potter,
  title={Enabling Ontologies for Semi-Automated Feature Extraction and Traceability of Quality Issues in Production Line Monitoring},
  author={Soete, Colin and Steenwinckel, Bram and Mussche, Joris and Ongenae, Femke and Van Hoecke, Sofie},
    note = {Under review at: AI in Advanced Manufacturing, special issue of International Journal of Intelligent Systems},
}
Ontology Specification Draft

Abstract

POTTER is an ontology for semantic machine learning pipeline traceability in Industry 4.0 settings. It integrates and extends existing ontologies (MLS, EEP, SSN, and SPARKS) to embed production lines into semantic workflows. All previous versions of the ontology and its documentation are available on [GitHub](https://github.com/predict-idlab/potter).

Introduction back to ToC

This is a place holder text for the introduction. The introduction should briefly describe the ontology, its motivation, state of the art and goals.

Namespace declarations

Table 1: Namespaces used in the document
bibo<http://purl.org/ontology/bibo/>
dc<http://purl.org/dc/elements/1.1/>
fno<https://w3id.org/function/vocabulary/composition#>
foaf<http://xmlns.com/foaf/0.1/>
ns<http://www.w3.org/2003/06/sw-vocab-status/ns#>
ontology<https://w3id.org/function/ontology#>
owl<http://www.w3.org/2002/07/owl#>
potter<https://github.com/predict-idlab/potter/>
rdf<http://www.w3.org/1999/02/22-rdf-syntax-ns#>
rdfs<http://www.w3.org/2000/01/rdf-schema#>
schema<http://schema.org/>
skos<http://www.w3.org/2004/02/skos/core#>
ssn<http://www.w3.org/ns/ssn/>
swrl<http://www.w3.org/2003/11/swrl#>
swrla<http://swrl.stanford.edu/ontologies/3.3/swrla.owl#>
swrlb<http://www.w3.org/2003/11/swrlb#>
terms<http://purl.org/dc/terms/>
vann<http://purl.org/vocab/vann/>
xml<http://www.w3.org/XML/1998/namespace>
xsd<http://www.w3.org/2001/XMLSchema#>

Overview figure

Figure 1: Overview of the potter ontology, defining and combining new concepts and relationships of the Execution-Executor-Procedure paradigm with the Smart industries sparks, MLSchema, Semantic sensor network, and Function ontology.

Processing Ontology for Tracing, Training and Extraction of Results (POTTER): Overview back to ToC

This ontology has the following classes and properties.

Classes

Object Properties

Data Properties

Processing Ontology for Tracing, Training and Extraction of Results (POTTER): Description back to ToC

POTTER is an ontology for semantic machine learning pipeline traceability in Industry 4.0 settings. It integrates and extends existing ontologies (MLS, EEP, SSN, and SPARKS) to embed production lines into semantic workflows.

Cross-reference for Processing Ontology for Tracing, Training and Extraction of Results (POTTER) classes, object properties and data properties back to ToC

This section provides details for each class and property defined by Processing Ontology for Tracing, Training and Extraction of Results (POTTER).

Classes

DataLoaderc back to ToC or Class ToC

IRI: https://github.com/predict-idlab/potter/DataLoader

The DataLoader concept was made to specify data as an input for other concepts.
has super-classes
InputFunction c

DataProcessorc back to ToC or Class ToC

IRI: https://github.com/predict-idlab/potter/DataProcessor

An InputOutputFunction subclass concept which represents a machine learning data processor.
has super-classes
InputOutputFunction c

FeatureProcessorc back to ToC or Class ToC

IRI: https://github.com/predict-idlab/potter/FeatureProcessor

An InputOutputFunction subclass concept which represents a machine learning feature processor.
has super-classes
InputOutputFunction c

InputFunctionc back to ToC or Class ToC

IRI: https://github.com/predict-idlab/potter/InputFunction

The InputFunction concept provides a way to define the input operations of our machine learning pipeline.
has super-classes
Function c
has sub-classes
DataLoader c
is in range of
usesInput op

InputOutputFunctionc back to ToC or Class ToC

IRI: https://github.com/predict-idlab/potter/InputOutputFunction

The InputOutputFunction concept converts input values to new output values.
has super-classes
Function c
has sub-classes
DataProcessor c, FeatureProcessor c
is in range of
usesIO op

ModelBuilderc back to ToC or Class ToC

IRI: https://github.com/predict-idlab/potter/ModelBuilder

A ModelBuilder is a function to output a machine learning model that further on can be used to make predictions.
has super-classes
OutputFunction c

OutputFunctionc back to ToC or Class ToC

IRI: https://github.com/predict-idlab/potter/OutputFunction

The OutputFunction concepts defines how different inputs can be used to create one output concept.
has super-classes
Function c
has sub-classes
ModelBuilder c

Object Properties

isInputop back to ToC or Object Property ToC

IRI: https://github.com/predict-idlab/potter/isInput

Helper property used to define the InputFunction.

isIOop back to ToC or Object Property ToC

IRI: https://github.com/predict-idlab/potter/isIO

Helper property used to define the InputOutputFunction.

isOutputop back to ToC or Object Property ToC

IRI: https://github.com/predict-idlab/potter/isOutput

Helper property used to define the OutputFunction.

pertainop back to ToC or Object Property ToC

IRI: https://github.com/predict-idlab/potter/pertain

The pertain concept is used to indicate when a function pertains to a particular system component.
has domain
Function c
has range
System c

usesInputop back to ToC or Object Property ToC

IRI: https://github.com/predict-idlab/potter/usesInput

Helper property used to define the relation of the InputFunction concept using property chaining.
has range
InputFunction c
has sub-property chains
isIO op o expects op o o isInput op
isOutput op o expects op o o isInput op

usesIOop back to ToC or Object Property ToC

IRI: https://github.com/predict-idlab/potter/usesIO

Helper property used to define the relation of the InputOutputFunction concept using property chaining.
has range
InputOutputFunction c
has sub-property chains
isIO op o expects op o o isIO op
isOutput op o expects op o o isIO op

Data Properties

pythonDefinitiondp back to ToC or Data Property ToC

IRI: https://github.com/predict-idlab/potter/pythonDefinition

Data property used to link the semantic concepts into the specific implementation using Python decorators.

Legend back to ToC

c: Classes
op: Object Properties
dp: Data Properties

Acknowledgments back to ToC

The authors would like to thank Silvio Peroni for developing LODE, a Live OWL Documentation Environment, which is used for representing the Cross Referencing Section of this document and Daniel Garijo for developing Widoco, the program used to create the template used in this documentation.