SEMANTICS AND SCIENCE: FACILITATING A COMMUNITY CENTRIC VIEW

Scientific information is pouring in from satellites, sensors, instruments, cameras, documents and devices. This raw information is processed into an ever-increasing number of data products for use by the scientific, policy-making, academic and corporate communities. Often, these communities need to share information for collaboration yet the languages of these communities differ. Communities use different terms and relationships to describe and define their information; they trust different sources; they have different experts; and they rely on different and ever-changing models.

Understanding complex systems requires cross-community collaboration and historical perspective. Communities must be connected so that complex systems can be understood, boundary conditions can be examined and sensitivity to potentially impacted or dependent systems can be understood. Forcing a common language for these interconnected communities will not happen, and if it did, it would result in a lack of needed information fidelity and context.

Semantic approaches to cross-community collaboration can facilitate the necessary knowledge sharing and reusability to abstract meaning in support of an overall understanding of complex systems.

This presentation demonstrates how different business, policy-making and scientific communities can collaborate using multiple domain or problem representations (knowledge models). It illustrates the interconnections and information/knowledge sharing between seemingly unrelated communities where serendipitous discovery matters. In addition, the presentation demonstrates how these rich semantic models can be utilized to find similar and related information and discover unexpected connections between seemingly arbitrary pieces of information.

The presentation centers on a map-based interface that allows different communities to visualize changing information and relationships (in a spatial environment) based on their domain or problem of interest. Within this semantically-rich, spatial environment, different groups of users can review metadata, make annotations, create new relationships, view information history and filter their views to focus on the appropriate level of knowledge and information detail to address their questions. New knowledge can be captured and community members can be automatically notified of information of interest.

While semantic models are touched on, the focus of the presentation is on their use in filtering information to allow users to quickly focus on relevant and needed information in their own domains.