This book presents the refereed proceedings of the 4th Congress of the Italian Association for Artificial Intelligence, AI*IA '95, held in Florence, Italy, in October 1995.The 31 revised full papers and the 12 short presentations contained in the volume were selected from a total of 101 submissions on the basis of a careful reviewing process.

Natural Language Generation as constraint-based configuration.- Issues of multilinguality in the automatic generation of administrative instructional texts.- Extending Q-learning to Fuzzy Classifier Systems.- Fuzzy cognitive maps in multi-agent environments.- Multiple predicate learning with RTL.- Learning while -Solving problems in single agent search: Preliminary results.- Automatic construction of navigable concept networks characterizing text databases.- Temporal prediction: Dealing with change and interactions within a causal framework.- Non-first-order features in concept languages.- PDL-based framework for reasoning about actions.- Dependency graphs in natural language processing.- Integrating shallow and linguistic techniques for Information extraction from text.- Recognizing preliminary sentences in dialogue interpretation.- Contextuality and non-extensional identity: the inescapable symbiosis in NLP.- Priorities in Default Logic revisited.- Boolean approach for representing and solving constraint-satisfaction problems.- Composing decision procedures: the approach and a case study.- Knowledge representation, exemplification, and the Gupta-Belnap theory of circular definitions.- $$\mathcal{T}\mathcal{R}\mathcal{D}\mathcal{L}$$ : A language for conceptual modelling in Information Systems Engineering.- The different roles of abstraction in abductive reasoning.- A cognitive model of causal reasoning about the physical world.- A generalized approach to consistency based belief revision.- A framework for dealing with belief-goal dynamics.- Evolving non-trivial behaviors on real robots: An autonomous robot that picks up objects.- A formal domain description language for a temporal planner.- A method for solving multiple autonomous robots collisions problem usingspace and time representation.- Mapping symbolic knowledge into locally receptive field networks.- Knowledge representation for robotic vision based on conceptual spaces and attentive mechanisms.- A weakest precondition semantics for conditional planning.- A cognitive hybrid model for autonomous navigation.- Using a chemical metaphor to implement autonomous systems.- Modeling process diagnostic knowledge through causal networks.- Formalizing reasoning about change: A temporal diagnosis approach.- Preventive diagnosis: Definition and logical model.- Learning programs in different paradigms using Genetic Programming.- Revision of logical theories.- Reformulation of examples in concept learning of structural descriptions.- Integrated Model — A Proposal to handle noise.- Exceptions-based synthesis of Boolean functions as a core mechanism to perform concept learning.- Seeing is believing.- Agents as reasoners, observers or believers.- Modelling interactions in agent system.- Agent coordination and control through logic theories.