Keywords
component-based modelling, semantic annotation, python, pcraster
Start Date
1-7-2012 12:00 AM
Abstract
Integrated models are valuable tools for research and decision support as they allow a comprehensive analysis of environmental systems. Component–based software frameworks aid in their development by a more straightforward construction and coupling of generic components. Formal descriptions of model components and their relationships by means of ontologies ease the clarification of the embedded knowledge. However, existing approaches do not adequately cover the development and assessment phases of integrated models. Here, we define invariant characteristics of process–based spatio– temporal systems and introduce an ontology to semantically describe model components and their interactions. We address the role of the ontology during the different model development phases. During model initialisation, the framework uses the information to check the validity of linked model components in order to prevent the exchange of mismatching information such as different data types or spatial extents. For model execution, the framework combines the semantic information of the individual model components to an integrated description. The ontology is integrated in a declarative modelling framework allowing an environmental scientist to add semantic information for component properties such as parameter values and state variables. The semantic enrichment of the model components consolidates their application in integrated systems and is a step towards improved interoperability with other modelling frameworks.
Towards integrated model building with semantically annotated components
Integrated models are valuable tools for research and decision support as they allow a comprehensive analysis of environmental systems. Component–based software frameworks aid in their development by a more straightforward construction and coupling of generic components. Formal descriptions of model components and their relationships by means of ontologies ease the clarification of the embedded knowledge. However, existing approaches do not adequately cover the development and assessment phases of integrated models. Here, we define invariant characteristics of process–based spatio– temporal systems and introduce an ontology to semantically describe model components and their interactions. We address the role of the ontology during the different model development phases. During model initialisation, the framework uses the information to check the validity of linked model components in order to prevent the exchange of mismatching information such as different data types or spatial extents. For model execution, the framework combines the semantic information of the individual model components to an integrated description. The ontology is integrated in a declarative modelling framework allowing an environmental scientist to add semantic information for component properties such as parameter values and state variables. The semantic enrichment of the model components consolidates their application in integrated systems and is a step towards improved interoperability with other modelling frameworks.
