Ontology Management in an Industrial Environment: The BASF Governance Operational Model for Ontologies (GOMO)

Abstract

Governance on ontology development and maintenance practices within an organization has many advantages over ungoverned, siloed approaches that many organizations exhibit nowadays. This paper presents the BASF Governance Operational Model for Ontologies (GOMO), which addresses all stages of the ontology lifecycle and provides a framework for the development and maintenance of ontologies within BASF. GOMO is comprised of Principles, Standards and Quality Assurance criteria, Best Practices, Training and Outreach and is the result of a collaborative effort between industry and academia in the semantic web field. GOMO Principles, Standards and Best Practices are being applied to all running ontology-based projects in BASF. Through outreach presentations with sections of the BASF community, GOMO has reached a wider audience to foster understanding on the utility and implementation of ontologies. Finally, GOMO stands as a framework that is fit for adoption by other organizations facing similar challenges in ontology governance.

Overview

Governing Principles

• Principle 1: BASF ontologies MUST be FAIR. The FAIR guiding principles for data management and stewardship are a set of technology-agnostic guidelines to make digital assets Findable, Accessible, Interoperable, and Reusable. These guiding principles are applied to semantic resources such as ontologies that MUST also be FAIR. A set of five GOMO FAIR recommendations are applied to BASF ontologies:
  • Use global unique, persistent, and resolvable IRIs.
  • Describe the ontology using a common minimum metadata schema.
  • Represent the ontology using a standard formal language.
  • Reuse ontologies whenever possible.
  • Include a License/Terms-of-Use and consider the licenses of reused ontologies.
• Principle 2. BASF ontologies MUST be available to the BASF Community. The latest ontology release MUST be discoverable through a lookup service and published in an ontology consumption platform where it is available to humans and machines alike, via APIs, direct access to the code, a SPARQL endpoint, or through an ontology visualization interface.
• Principle 3. Access to BASF ontologies MUST be governed by access rights and security policies. Every BASF ontology MUST be accessible according to roles, confidentiality levels, access rights and security policies put in place in an access layer that implements the authorization procedures.  
• Principle 4. The different versions of BASF ontologies MUST persist over time. Ontologies are expected to change over time as they are developed, used, and refined. This will lead to a series of different versions of the ontology that MUST be persistent and accessible through a version control system.
• Principle 5. BASF Ontologies MUST include human-readable documentation. Human-readable documentation MUST be included in BASF ontologies to provide context, descriptions and examples.
• Principle 6. BASF ontologies SHOULD be modularized. BASF ontologies SHOULD be modularized so that interconnected modules addressing specific knowledge areas in the scope of the ontology or having different security requirements, are networked in a usable way.
• Principle 7. BASF ontologies MUST have a community. The ontology development lifecycle is divided into several stages that involve different tasks. Multiple actors within the organization that constitute the ontology community play different roles during the ontology development process and have different degrees of involvement in its development tasks.

Standards

• Standard 1: Naming convention. Provides a generic agreed scheme and syntax for naming terms.
• Standard 2: Valid ontology formats. Establish the supported representation languages and formats for encoding the ontologies.
• Standard 3: Ontology metadata. Indicates the metadata that describes ontologies in general.
• Standard 4: Ontology class and property mandatory metadata. Indicates the mandatory metadata for ontologies that must be included for classes and properties in the ontology.
• Standard 5: Obsoleting ontology entities. Describes how to deprecate ontology entities instead of deleting them in a standardized way.
• Standard 6: IRI structure. Specifies how an IRI must be structured in BASF ontologies.
• Standard 7: Ontology version control repository structure. Indicates the structure of the repository where the ontology is managed to ensure version control.
• Standard 8: Ontology version numbering. Provides an agreed numbering scheme for ontology versions.
• Standard 9: Documentation. Describes the human-readable documentation components necessary for reference and management of the stages in the ontology development lifecycle and for ontology usage as well.
• Standard 10: Ontology diagram notation. Provides an agreed common convention in terms of shapes and elements when creating diagrams that represent the information described in an ontology.