ECAI-2002 Conference Paper

[full paper]

Modelling Contextual Meta-Knowledge in Temporal Model Based Diagnosis

J. Palma, R. Marín

Applying Model Based Diagnosis (MBD) techniques in medical domains reveal the need to use deep causal knowledge modelling frameworks as well as temporal management techniques to capture the dynamic component of disease evolution (with the latter being very important in other domains). Despite the intense research activity in the field of Temporal MDB, there are three issues that not been analysed in depth: (a) modelling complex interaction of contextual information, (b) evaluation of hypotheses possibility degrees and (c) the structure of explanations. Our aim is to present a general framework for Temporal MDB which approaches these problems and to demonstrate the suitability of the Fuzzy Temporal Constraints Networks formalism (FTCN) for representing the domain temporal dimension.

Keywords: Diagnosis, Model Based Reasoning, Temporal Reasoning, Causal Reasoning

Citation: J. Palma, R. Marín: Modelling Contextual Meta-Knowledge in Temporal Model Based Diagnosis. In F. van Harmelen (ed.): ECAI2002, Proceedings of the 15th European Conference on Artificial Intelligence, IOS Press, Amsterdam, 2002, pp.407-411.