Why the development of environmental indicators needs ontologies

Investment in ontology design for the OEPI project is not to be taken literally as “just ontology”. It is one important and necessary part of the Semantic Web approach and thus this investment is actually an important step in setting the direction for the whole implementation approach in OEPI. Therefore, the actual question is why OEPI needs to go for application of Semantic Web technologies with Ontology being a part thereof.

It is about this alternative:

Is OEPI technology old-fashioned and obsolete before its implementation even has begun and fails to meet its goals? – Or is OEPI technology future-oriented and able to leverage the capabilities of “the Web” even beyond its current exploitation and succeeds in fulfilling its mission?

The mission of OEPI is to deliver more than what is possible today regarding Environmental Performance Indicators and their widespread use.
Semantic Web technologies – including ontology as one important piece – are necessary to ensure the fulfilment of this mission, in more detail:

  • OEPI wants to use and to combine as much existing (web) resources as possible and bring them close to the user. The ability to exchange the semantics of such resources is of paramount importance for the practical implementation.
  • OEPI wants to automate these processes in an easy and user-friendly way.
  • OEPI wants to enable flexible and ad-hoc use of resources.

A very big ontology
Conventional technologies alone, even including modern concepts as, for example, Model Driven Architecture or Service Oriented Architecture, are helpful but not sufficient to meet these requirements because they lack explicit semantics of the modelled domain as also of existing data and services. Currently, conventional Systems and Software Engineering is able to solve configuration tasks in advance and according to predefined decision patterns but not
by inferring decisions from semantically rich descriptions, constraints, and restrictions at runtime. Semantic Web technologies add capabilities for provision of such features to the conventional technologies. One first approach to bridge the gap from ontology to conventional MDA, which is quite common already, could be the definition of a domain-specific UML profile
representing ontology.

OEPI needs Semantic Web technologies (in combination with conventional technologies) to describe, to categorize, to find and to utilize existing resources and to formalize spontaneous requirements of users for supporting their daily work flexibly.

In this approach, the ontology has the purpose to unambiguously and explicitly describe all the “OEPI things” with their properties, relationships, and semantics in a formalized way that can be evaluated by humans and by software. It constitutes the common reference and resource for all data models, service descriptions etc. that have to be developed for the implementation of the platform as well as for different independent services or solutions. It is
needed to enable and ensure semantic interoperability among existing, new, and future services and solutions.

OEPI will not be able to fulfil all these requirements completely during the limited course of the project but it is responsibility of OEPI to provide a sound foundation that can be built on and further extended and enhanced by industry as well as research beyond the end of the
project. OEPI Ontology will be one part of this foundation.


“Ontology Driven Architectures and Potential Uses of the Semantic Web in Systems and Software Engineering”, Working Group Note of W3C Semantic Web Best Practices & Deployment Working Group, http://www.w3.org/2001/sw/BestPractices/SE/ODA/

An overview of top level ontologies

Ontologies are mostly created for a specific purpose as for OEPI that is the creation of an ontology for environmental performance indicators. Apart from these specific ontologies there are also more general ones that are called top level or upper ontologies. They serve as a kind of superstructure above the specific domain, task and application ontologies. Its purpose is to provide the possibility to integrate specialized ontologies into a generic semantic context that is levelled above them. Examples for such ontologies are those that provide proper integration of concepts into the earth system, common time and space knowledge or just proper mapping to a multilingual dictionary or thesaurus.

logo knot
wonder web
onto med

During the development of  OEPI’s ontology for environmental performance indicators, several top level ontologies have been reviewed. This is part of the ontology design process as by that it may be avoided to create an own comprehensive top level ontology which is a complex and lengthy task. The basis for the research done was a comprehensive review done by Mascardi, V., Cordì, V. and Rosso, P. (2007):  A comparison of upper ontologies, Proceedings of Conf. on Agenti e industria: Applicazioni tecnologiche degli agenti software, WOA07, Genova, Italy, pp. 24-25. They developed a set of criteria to evaluate several ontologies. By that we could have a better understanding of the most important top level ontologies that exist.

The evaluated ontologies are in detail:

During our review we found that most top level ontologies have common agreements of top level issues like objects, world, properties, events or processes. Nevertheless many of the evaluated ontologies are already geared toward specific purposes like for example the medical domain, natural sciences or language processing. So it will make more sense to have the OEPI ontology mapped to existing abstract concepts of one or several of the ontologies from above.

If you know of additional top level ontologies that may be added to this list, please let us know and we might include it here.