Challenges to Decision Support in a Changing World
Papers from the AAAI Spring Symposium
Marek J. Druzdzel and Tze-Yun Leong, Cochairs
Technical Report SS-05-02
136 pp., $30.00
ISBN 978-1-57735-228-0
[Add to Cart] [View Cart]
One of the most daunting challenges faced by decision support systems is a perpetual change in their environment. Existing decision support methodologies, tools, and frameworks are often difficult to scale up and adapt to changing knowledge, workflow, and operational setting. Systems that have to cope with change need to include methodologies that go outside single theories. For example, systems that are based on probabilistic or decision-theoretic principles will be typically unable to cope with change by themselves, as neither probability theory nor decision theory say much about how decision models are constructed, let alone how they should be modified. The general AI concepts of perception, learning, control, abstraction, and personalization must be inherently designed into the methodological, architectural, and operational aspects of adaptive systems, from application design through software and hardware infrastructure support.