Ontology based dependency analysis: Understanding the impacts of decisions in a collaborative environment

Central to any collaboration problem involving planning, scheduling, analysis, simulation, etc. is the ability to model and understand how the effects of decisions and actions of one system propagate through the collective understanding, bringing to light insights and information. This raises the important research question of how information in different systems can be brought together in an understandable way but without the need to re-engineer either their interfaces or the data needs. The approach to ldquoloose systemsrdquo coupling explored in this paper is to organize information so that the dependency between information elements (representing people/groups, locations, resources and concepts and provided by disparate sources) can be identified, the consequences of the impacts on these dependencies analyzed and any consequences made explicit. The proposed generic information mapping capability is provided by means of two high level ontologies: the node ontology and the event/action (EA) ontology, details of which are provided in this paper. These high level ontologies allow existing ontologies developed by other groups (CIA Fact book, ldquoFriend of a Friendrdquo, etc.) to be easily integrated via the Protege ontology editor. This significantly reduces the effort required to develop and maintain ontologies and tools such as OntoLT allow candidate ontologies to be extracted automatically from structured and unstructured sources and linked via Protege The collaboration architecture is agnostic to the type of application and the type of dependencies being analyzed and can be easily applied to crisis management, cyber analysis or military applications. Details are provided of the components of the proposed architecture together with a working example that shows the type of collaboration process it supports and the scope of problems it is able to support and solve. The proposed collaboration architecture is available for demonstration and evaluation and is provided- with an example set of model and information sources.