Semantic Web Technologies for a Common Language

With SC3 on its way to create a common language for total collaboration between humans and machines and all partners involved, here is a description of project partner TIB’s motivation and approach in contributing to an overall success. The Technische Informationsbibliothek Hannover is an information service provider for major international corporations and it builds on its expertise in ontology development, maintenance and publication.

Collaborative Ontology Development Framework

Addressing the machine-readable conceptualisation of the semiconductor domain and related supply chains, it takes standardised ontologies of high quality and a common agreement of conceptualisations to fully apply the advantages of Semantic Web technologies into the domain. Ontology development typically involves joint efforts of domain experts and knowledge engineers. While the latter tend to lack the depth knowledge of a specific domain to create ontologies of sufficient quality, the domain experts in turn find it often hard to follow logical notations in OWL (Web Ontology Language) representation.

To more directly involve them in ontology modelling and foster communication with knowledge engineers, approaches require mechanisms to reduce OWL formalisation complexity for domain experts. Still, they have to maintain the full OWL modelling features for knowledge engineers. This is why TIB is developing a user-centred approach for collaborative ontology development, addressing different user groups, their needs and backgrounds.

Bridging the gap between knowledge engineers and domain experts

A core aspect of the Semantic Web is to create and communicate conceptualisations of information and data in certain domains, with taxonomies, vocabularies, and ontologies providing a formal, machine-readable representation of the domain knowledge. Yet, ontology modelling is a difficult task involving the conceptualisations of processes and their relations. Moreover, the specifications of a knowledge domain have to be fulfilled serving the needs of different stakeholders.

To do so and to ensure a successful usage in specific use cases, an agreement of the conceptualisation must be established between various stakeholders. Visualisations provide an abstraction layer for the underlying data and a good starting point for exploration and understanding. Thus, they reduce the entrance barrier for domain experts to get more directly involved in ontology modelling. However, suitable visualisations are highly dependent on individual use cases and targeted user groups. Flexible and customisable approaches are required to allow users to adjust visual representations to their needs.

Hybrid modes of operation

TIB identified that the most crucial requirement for the approach is to serve users' needs from various audiences with diverse backgrounds and in different contexts. Because web-based approaches are ready-to-use without the need to install additional software they reduce entrance barriers and engage different user groups more directly.

Having defined the above-mentioned domain experts (non-experts as to ontology modelling) and the knowledge engineers (expert users) as the two relevant user groups in the project, TIB describes the individual requirements as follows:

“For domain experts the approach should be easy-to-learn/easy-to-use with a least complex OWL modelling, providing guidance and best-practice suggestions during the modelling. For expert users in turn the approach should support all modelling features of the commonly used open source ontology editor Protégé; and visual modelling paradigms in the form of node-link diagrams can address the requirements easy-to-use and low complexity.

However, they have to provide additional customisations for the visual representation to facilitate understanding. The requirements guidance and best practices can be addressed using auto-complete functionalities to align the created nodes and links with existing terms of ontologies, reducing manual labour to transform high-level conceptualisations to OWL elements.

To support all OWL modelling features, we envision the use and extension of the OWL-API (Java Application Programming Interface), the backbone of Protégé. Additionally, to engage diverse groups in the modelling process, we envision three modes of modelling operation: textual (for expert users), widget-based (for intermediate users) and visual (for novel users).

For a successful adoption and broad use, we define additional requirements for a collaborative ontology development framework which allows to create and to share different views, thus enabling and fostering communication on all abstraction levels and between various stakeholders.

Since the created ontologies and their conceptualisations thrive from discussions and joint agreement upon their definitions and their usage, blockchain technologies providing the means for versioning and history of changes should be part of the solution. Integrating them with ontology development will foster trust and validation of ontologies and enable long term support for downward compatible systems using old ontology versions”.