Proving formulas through reduction to decidable classes -- Building and executing proof strategies in a formal metatheory -- Computing 3-valued stable models by using the ATMS -- Abstract properties for the choice provability relation in nonmonotonic logics -- Characterizing prime implicants as projective spaces -- EFH-Soar: Modeling education in highly interactive microworlds -- Foundations for interaction: The dependence theory -- Letter spirit: An architecture for creativity in a microdomain -- New systems for extracting 3-D shape information from images -- Projecting sub-symbolic onto symbolic representations in artificial neural networks -- Integrating the symbolic and the sub-symbolic level in sonar-based navigation -- Randomness, imitation or reason explain agents' behaviour into an artificial stock market? -- Neural networks for constraint satisfaction -- Reasoning with individuals in concept languages -- A family of temporal terminological logics -- Logic programming and autoepistemic logics: New relations and complexity results -- Inferring in lego-land: an architecture for the integration of heterogeneous inference modules -- MAP — a language for the modelling of Multi-Agent systems -- Developing co-operating legal knowledge based systems -- Negation as a specializing operator -- Constructing refinement operators by decomposing logical implication -- Learning relations: Basing top-down methods on inverse resolution -- Complexity of the CFP, a method for Classification based on Feature Partitioning -- Genetic algorithms elitist probabilistic of degree 1, a generalization of simulated annealing -- Learning relations using genetic algorithms -- Evolutionary learning for relaxation labeling processes -- Increasing cohesion in automatically generated natural language texts -- Production of cooperative answers on the basis of partial knowledge in information-seeking dialogues -- Coping with modifiers in a restricted domain -- Explanation strategies in a tutoring system -- Maintaining consistency in quantitative temporal constraint networks for planning and scheduling -- Making an autonomous robot plan temporally constrained maintenance operations -- A generative constraint formalism for configuration problems -- Selecting observation time in the monitoring and interpretation of time-varying data -- Spatial reasoning in a holey world.

This book contains 22 long papers and 13 short ones selected for the Scientific Track of the Third Congress of the Italian Association for Artificial Intelligence. The long papers report completed work whereas the short papers are mainly devoted to ongoing research. The papers report significant work carried out in the different subfields of artificial intelligence not only in Italy but also elsewhere: 8 of the papers come from outside Italy, with 2 from the United States and 1 eachfrom Australia, Austria, Germany, The Netherlands, Spain, and Turkey. The papers in the book are grouped into parts on: automated reasoning; cognitive models; connectionist models and subsymbolic approaches; knowledge representation and reasoning; languages, architectures and tools for AI; machine learning; natural language; planning and robotics; and reasoning about physical systems and artifacts.