Advances in Artificial Intelligence

AI2000 was the 13th in the series of biennial Arti cial Intelligence conf- ences sponsored by the Canadian Society for Computational Studies of Int- ligence/Soci et e canadienne pour l etude de l intelligence par ordinateur. For the rst time, the conference was held in conjunction with four other conferences. TwooftheseweretheannualconferencesoftwootherCanadiansocieties,Gra- icsInterface(GI2000)andVisionInterface(VI2000),withwhichthisconference has been associated in recent years. The other two conferences were the Int- national Symposium on Robotics (ISR2000) and the Institute for Robotics and Intelligent Systems conference (IRIS2000). It is our hope that the overall ex- rience will be enriched by this conjunction of conferences. The Canadian AI conference has a 25 year tradition of attracting Canadian and international papers of high quality from a variety of AI research areas. All papers submitted to the conference received three independent reviews. - proximately one third were accepted for plenary presentation at the conference. A journal version of the best paper of the conference will be invited to appear in Computational Intelligence. The conference attracted submissions from six continents, and this diversity is represented in these proceedings. The overall framework is similar to that of the last conference, AI 98. Plenary presentations were given of 25 papers, organized into sessions based on topics. Poster pres- tations were given for an additional 13 papers. A highlight of the conference continues to be the invited speakers. Three speakers, Eugene Charniak, Eric _ Horvitz, and Jan Zytkow, were our guests this year.

Includes supplementary material: sn.pub/extras

Inhalt
Games / Constraint Satisfaction.- Unifying Single-Agent and Two-Player Search.- Are Bees Better than Fruitflies?.- A Constraint Directed Model for Partial Constraint Satisfaction Problems.- Natural Language I.- Using Noun Phrase Heads to Extract Document Keyphrases.- Expanding the Type Hierarchy with Nonlexical Concepts.- Using Object Influence Areas to Quantitatively Deal with Neighborhood and Perception in Route Descriptions.- An Extendable Natural Language Interface to a Consumer Service Database.- Knowledge Representation.- Identifying and Eliminating Irrelevant Instances Using Information Theory.- Keep It Simple: A Case-Base Maintenance Policy Based on Clustering and Information Theory.- On the Integration of Recursive -Theories.- Natural Language II.- Collocation Discovery for Optimal Bilingual Lexicon Development.- The Power of the TSNLP: Lessons from a Diagnostic Evaluation of a Broad-Coverage Parser.- A Parallel Approach to Unified Cognitive Modeling of Language Processing within a Visual Context.- AI Applications.- Interact: A Staged Approach to Customer Service Automation.- Towards Very Large Terminological Knowledge Bases: A Case Study from Medicine.- The Use of Ontologies and Meta-knowledge to Facilitate the Sharing of Knowledge in a Multi-agent Personal Communication System.- Machine Learning / Data Mining.- ASERC - A Genetic Sequencing Operator for Asymmetric Permutation Problems.- CViz: An Interactive Visualization System for Rule Induction.- Learning Pseudo-independent Models: Analytical and Experimental Results.- Planning / Theorem Proving / Artificial Life.- Learning Rewrite Rules versus Search Control Rules to Improve Plan Quality.- Scheduling Methods for Parallel Automated Theorem Proving.- Simulating Competing Alife Organisms by Constructive Compound Neural Networks.- Neural Networks.- A Recognition-Based Alternative to Discrimination-Based Multi-layer Perceptrons.- Accelerated Backpropagation Learning: Extended Dynamic Parallel Tangent Optimization Algorithm.- Neural ARX Models and PAC Learning.- Posters.- Qualitative Descriptors and Action Perception.- A Comparison of Association Rule Discovery and Bayesian Network Causal Inference Algorithms to Discover Relationships in Discrete Data.- Towards an Automated Citation Classifier.- Typical Example Selection for Learning Classifiers.- Comparative Study of Neural Network Controllers for Nonlinear Dynamic Systems.- The Iterative Multi-agent Method for Solving Complex Search Problems.- Relational Learning with Transfer of Knowledge Between Domains.- Surviving in a Hostile Multi-agent Environment: How Simple Affective States Can Aid in the Competition for Resources.- Task-Structure Based Mediation: The Travel-Planning Assistant Example.- Considerations on Compositional Update Operators.- The Degeneration of Relevance in Uncertain Temporal Domains: An Empirical Study.- The Learnability of Naive Bayes.- Invited Presentations.- Parsing to Meaning, Statistically.- Automated Discovery: A Fusion of Multidisciplinary Principles.