- Developing a vocabulary to support semantic annotations for information interdependencies to capture the rationale for decisions as well as traces to the original information sources used. Our initial version of this language draws from research in various areas, including argumentation and discourse relations, logic deduction, process and plan representations, and event understanding.
- Developing semantic annotations of decisions and tradeoffs through the use of markup languages. The initial version of this vocabulary is now available as a schema/ontology in XMLS, RDFS, and DAML+OIL. TRELLIS allows users to extend this vocabulary and its corresponding schemas.
A nice result of this work is a comparison table of semantic markup languages that helped us the features and tradeoffs of existing markup languages. This table compares the features of XML Schema, RDF Schema, and DAML+OIL and is linked to examples of use of each feature. We also have tutorial slides on semantic markup languages that highlight their features.
- Developing an interface that helps users annotate information analysis and decision rationale. The first version of this interface was released in August 2000 and includes a creation tool, an annotation tool, and search and collaborative capabilities. We have used it to developed several scenarios that demonstrate generality of the approach, including intelligence analysis, genealogy research, and travel planning.
- Annotating the design rationale of knowledge bases, to create a new generation of "Resilient Hyper Knowledge Bases" (RHKBs) that contain pointers to the original sources of knowledge as well as the analysis and decisions of the knowledge engineers that built it.
- Using the annotations to derive an assessment of the source based on the annotations of many individuals.