Knowledge representation artifacts for use in sensemaking support systems

Abstract

The development of sensemaking support systems requires that one cares about knowledge representation. Motivated by the fact that no single representation method is ideally suited by itself for all tasks, the authors propose a collection of knowledge representation artifacts appropriate for processing in computer-based support systems for situation analysis. The approach described makes it possible to combine the advantages of different representational forms. Each representation paradigm can be matched to an aspect of sensemaking that is a natural fit with this aspect. For example, representing information as propositions is suitable for automated reasoning, while encoding this information using a graph representation enables knowledge discovery through network analytics techniques. The spatial features are a good fit with geospatial reasoning, while situation cases evidently fit well with the case-based reasoning paradigm. These representation artifacts (and a few others) are briefly described in the paper, and some directions for future work are discussed.