IPMU'92 - Advanced Methods in Artificial Intelligence

The papers in this volume are extended versions of presentations at the fourth International Conference on Information Processing and Management of Uncertainty in Knowledge-based Systems (IPMU), held in Palma de Mallorca, July 6-10, 1992. The conference focused on issues related to the acquisition, representation, management, and transmission of information in knowledge-based and decision-making systems. The book is focused on methodologies related to artificial intelligence, including both theoretical and applied papers. The first section of the book is devoted to non-monotonic reasoning and the managementof default knowledge. The second section deals with methods using non-classical logics to deal with imperfect knowledge and to represent its spatial and temporal components. The third section presents various methods for the acquisition of uncertain and imprecise knowledge. The fourthsection is concerned with the use of qualitative, uncertain, temporal, and ambiguous pieces of information in knowledge-based systems, expert systems, and process controllers. The last section contains papers using artificial neural network methodologies.

Inhalt
Possibilistic abduction.- Management of preferences in assumption-based reasoning.- Default exclusion in a KL-ONE-like terminological component.- Unifying various approaches to default logic.- Using maximum entropy in a defeasible logic with probabilistic semantics.- Legality in inheritance networks.- A note on information systems associated to termal algebras.- A backward chaining resolution process involving non-monotonic operators.- On fuzzy conditionals generalising the material conditional.- Integrating resolutionlike procedures with Lukasiewicz implication.- The development of a Logic of Argumentation.- From and to or.- Representing spatial and temporal uncertainty.- An analysis of the temporal relations of intervals in relativistic space-time.- Accumulation and inference over finite-generated algebras for mapping approximations.- Similarity measures for case-based reasoning systems.- Statistical methods in learning.- Learning from erroneous examples using fuzzy logic and textbook knowledge.- Incremental learning of roughly represented concepts.- Self-organizing qualitative multimodel control.- MoHA, an hybrid learning model.- A new perspective in the inductive acquisition of knowledge from examples.- Knowledge representation through object in the development of expert system chemical synthesis and reaction.- Hierarchical representation of fuzzy if-then rules.- Approximate reasoning in expert systems: Inference and combination tools.- Modes of interval-based plausible reasoning viewed via the checklist paradigm.- Rule-based systems with unreliable conditions.- Fuzzy semantics in expert process control.- Qualitative operators and process engineer semantics of uncertainty.- Facing uncertainty in the management of large irrigation systems:Qualitative approach.- Semantic ambiguity in expert systems: The case of deterministic systems.- A deduction rule for the approximated knowledge of a mapping.- On knowledge base redundancy under uncertain reasoning.- A fuzzy logic approach for sensor validation in real time expert systems.- Application of Neuro-Fuzzy Networks to the identification and control of nonlinear dynamical systems.- Comparison between artificial neural networks and classical statistical methods in pattern recognition.- Learning methods for odor recognition modeling.