Kym J. Pohl

A translation engine in support of context-level interoperability

The need to support context-level interoperability is increasingly gaining importance in todays arena of semantic-oriented, decision-support systems. Unlike data-oriented exchange, such semantic interoperability must venture beyond the elementary communication of discrete data values and endeavor to translate between significantly more expressive, context-rich representations. Further, support of this level of interoperability must not require contamination of native perspective embedded within each participants representation. The solution offered in this paper presents a service-oriented framework supporting an extensible set of translation paradigms to effectively connect expressive, ontology-based environments. Fundamental to this solution is the notion of a remote service request. Employing this metaphor as the basis for participant interaction allows each system within a universe of potentially diverse representations to interoperate as collections of invocatible services. Further, by transparently marshalling such requests between client and service representations, such translation engine offers each member of this multi-lingual reality the means to interoperate within the familiar confines of their native representations. Finally, the discussion concludes with an evaluation of this capability in terms of the Levels of Conceptual Interoperability Model LCIM for assessing degrees of interoperability.

Semantically-enhanced information extraction

Information Extraction using Natural Language Processing (NLP) produces entities along with some of the relationships that may exist among them. To be semantically useful, however, such discrete extractions must be put into context through some form of intelligent analysis. This paper1,2 offers a two-part architecture that employs the statistical methods of traditional NLP to extract discrete information elements in a relatively domain-agnostic manner, which are then injected into an inference-enabled environment where they can be semantically analyzed. Within this semantic environment, extractions are woven into the contextual fabric of a user-provided, domain-centric ontology where users together with user-provided logic can analyze these extractions within a more contextually complete picture. Our demonstration system infers the possibility of a terrorist plot by extracting key events and relationships from a collection of news articles and intelligence reports.

Intelligent software for ecological building design

Building design is a complex process because of the number of elements and issues involved and the number of relationships that exist among them. Adding sustainability issues to the list increases the complexity of design by an order of magnitude. There is a need for computer assistance to manage the increased complexity of design and to provide intelligent collaboration in formulating acceptable design solutions. Software development technology today offers opportunities to design and build an intelligent software system environment that can serve as a reliable intelligent partner to the human designer. In this paper the authors discuss the requirements for an intelligent software design environment, explain the major challenges in designing this environment, propose an architecture for an intelligent design support system for sustainable design and present the existing technologies that can be used to implement that architecture.