Current Topics in Artificial Intelligence: 12th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2007, Salamanca, Spain, ... (Lecture Notes in Computer Science (4788))

This book constitutes the refereed proceedings of the 12th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2007, held in Salamanca, Spain, in November 2007, in conjunction with the 7th Workshop on Artificial Intelligence Technology Transfer, TTIA 2007. The 28 revised full papers presented were carefully selected during two rounds of reviewing and improvement from 134 submissions. The papers address all current issues of artificial intelligence ranging from methodological and foundational aspects to advanced applications in various fields Title Page Preface Organization Table of Contents Fast and Informed Action Selection for Planning with Sensing Introduction The Contingent Planning Problem $P$ The Conformant Translation $K$($P$) The Execution Model $X$($P$) The Classical Part $K^c$($P$) The Sensing Part $K^o$($P$) HeuristicModel $H$($P$) Action Selection and the CLG Planner Preliminary Experimental Results Discussion References Stacking Dynamic Time Warping for the Diagnosis of Dynamic Systems Introduction Consistency-Based Diagnosis Using Possible Conflict Machine Learning Techniques for Fault Identification Integration Proposal Case Study The System to Be Diagnosed Possible Conflicts for the System Experimental Design Results Conclusions References Retrieval of Relevant Concepts from a Text Collection Introduction Methodology Identification of Interesting Co-occurrences Obtaining Relevant Concepts for Each Interesting Co-occurrence Evaluation Conclusions References Interoperable Bayesian Agents for Collaborative Learning Environments Introduction Background Social Agent Reasoning Model Interoperable Knowledge Representation Conclusions References Knowledge Engineering and Planning for the Automated Synthesis of Customized Learning Designs Introduction HTN Planning Foundations HTN Planning Domains Our Approach Domain Extraction in a Simple Case Domain Extraction in Complex General Cases Obtaining a Plan Related Work Concluding Remarks References On the Initialization of Two-Stage Clustering with Class-GTM Introduction Two-Stage Clustering The Class-GTM Model Two-Stage Clustering Based on GTM Experiments Experimental Design and Settings Results and Discussion Conclusion References Three-Dimensional Anisotropic Noise Reduction with Automated Parameter Tuning: Application to Electron Cryotomography Introduction Review of Anisotropic Nonlinear Diffusion Estimation of Local Structure Concept of Diffusion in Image Processing Common Diffusion Approaches Anisotropic Nonlinear Diffusion in cryoET Diffusion Approach Smoothing the Background with Gaussian Filtering Numerical Discretization of the Diffusion Equation The Stopping Criterion: Noise Estimate Variance Automatic Parameter Tuning Scheme of Our Diffusion Approach Experimental Applications Conclusion References A Middle-Ware for the Automated Composition and Invocation of Semantic Web Services Based on Temporal HTN Planning Techniques Introduction SIADEX in a Nutshell Mapping an OWL-S Process Model into a SIADEX Domain Middle-Ware for the Composition and Execution of Web Services Related Work Conclusions References A Multiobjective Approach to Fuzzy Job Shop Problem Using Genetic Algorithms Introduction Uncertain Processing Times and Flexible Constraints The Job Shop Scheduling Problem Description of the Problem A Multiobjective Model Using Genetic Algorithms to Solve FJSP Experimental Results Conclusions and Future Work References CTC: An Alternative to Extract Explanation from Bagging Introduction Two Alternatives to Extract Explanation from Bagging CMM Algorithm CTC Algorithm Experimental Methodology Experimental Results Discriminating Capacity Explaining Capacity Conclusions References Ranking Attributes Using Learning of Preferences by Means of SVM Introduction Computation of Feature Ranking General Framework Support Vector Machines to Learn Preferences Experimental Results Evaluation of Methods for Building Ranking of Attributes Two Comparison Stages Inner Comparisons Outer Comparisons Conclusions References Improving HLRTA*($k$) Introduction Background HLRTA*\sc{LS}($k$) HLRTA*($k, d$) Experimental Results Conclusions References Sliding Mode Control of a Wastewater Plant with Neural Networks and Genetic Algorithms Introduction Plant Description Plant Control Equivalent Control Corrective Control Genetic Adjustment of the Corrective Control Simulation and Results References Efficient Pruning of Operators in Planning Domains Introduction Notation Context Subsumption Pruning Operators Extended Operators Experimental Results Conclusion References Heuristics for Planning with Action Costs Motivation Planning Model and Heuristics The Additive Heuristic The Relaxed Planning Graph Heuristic The Set-Additive Heuristic Additive and Set-Additive Heuristics Compared The FF($h^s_a$) Planner Experimental Results Summary References Mixed Narrative and Dialog Content Planning Based on BDI Agents Introduction Previous Work Social Systems and BDI Model The Multiagent System: Fantastic Society Natural Language Generation BDI Model in Storytelling Agent Planning Content Planner System Representation of the Reader’s Perception of the Story State Space Search Generation Goals An Example Comparison with Other Systems Conclusions and Future Work References NMUS: Structural Analysis for Improving the Derivation of All MUSes in Overconstrained Numeric CSPs Introduction Motivating Examples $NCSP$ with the Same Real Variables for All Constraints $NCSP$ with Some Different Variables for the Numeric Constraints Definitions and Notations Neighborhood-Based Structural Analysis General Neighborhood Variable-Based Neighborhood Domain-Based Neighborhood NMUS: Numeric MUSes Search Methods NMUS for NCSPs with the Same Variables for All Constraints NMUS for NCSPs with Some Different Variables for the Numeric Constraints Experimental Results Conclusions and Future Work References Interest Point Detectors for Visual SLAM Introduction Related Work Interest Point Detectors Harris Corner Detector Harris-Laplace SIFT SURF SUSAN Evaluation Methods Tracking Evaluation Measurements Experiments Conclusions References TBL Template Selection: An Evolutionary Approach Introduction Techniques Genetic Algorithms Transformation Based Learning Approaches Genetic Coding Fitness Function Cross-Over Operator Mutation Operator Experiments Conclusions References Finiteness Properties of Some Families of GP-Trees Introduction A Combinatorial Machine Model for Some Computer Programs Represented by GP-Trees VC Dimension Bounds Elements of Statistical Learning Theory Some Remarks About the Selection of the Fitness Function in GP Discussion and Related Work References A Workflow for the Networked Ontologies Lifecycle: A Case Study in FAO of the UN Introduction Fisheries Ontologies Lifecycle Users Roles Major Processes Editorial Workflow Use Cases Conclusions References A Logic for Order of Magnitude Reasoning with Negligibility, Non-closeness and Distance Introduction Non-closeness, Distance and Negligibility Syntax and Semantics of the Language ${\cal L}(OM)^{\textsc{NCD}}\,$ Axiom System for ${\cal L}(OM)^{\textsc{NCD}}\,$ Conclusions and Future Work References A Solution to the Rural Postman Problem Based on Artificial Ant Colonies Introduction The Traveling Salesman Problem The Rural Postman Problem Ant-Colony Based Algorithms ProposedSolution Transformation of the RPP into a TSP Ant-Algorithm Applied to the TSP Reconstruction of the RPP Solution Tests Conclusion References Olive Fly Infestation Prediction Using Machine Learning Techniques Motivation Problem Definition Dataset Elemental Problem Findings Applied Techniques Classification Trees Bayesian Networks Experiments Selection of Variables Filtering Variables Experiments Conclusions References Feature Selection Based on Sensitivity Analysis Introduction The Incremental AFN-FS Method The Anova Functional Network (AFN) Learning Method The Incremental AFN-FS Method Results Results for the Basic AFN-FS Method Results for the Incremental AFN-FS Method Conclusions and Future Work References Fitness Function Comparison for GA-Based Feature Construction Introduction MDL-Based Fitness in MFE3/GA Experimental Results Experiments with Synthetic Concepts Experiments with Real-World Data Conclusion References Appendix: Concept Definitions Generation of OWL Ontologies from Concept Maps in Shallow Domains Introduction Conceptual Maps Ontologies and Their Languages Integration of Conceptual Maps and Ontologies Obtaining OWL Ontologies Implementation Related Work Conclusions and Future Work References Effectiveness Study of Lexically Mapping Two Thesauri Introduction Materials The Thesaurus EMTREE The Unified Medical Language System Applications Methods Approaches to Map Terminologies Implementation and Evaluation Results and Discussion Patterns of Inaccuracy Producing the Decrease in Recall Patterns of Inaccuracy Leading to Fall in Precision Conclusions and Future Work References Author Index

This book constitutes the refereed proceedings of the 12th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2007, held in Salamanca, Spain, in November 2007, in conjunction with the 7th Workshop on Artificial Intelligence Technology Transfer, TTIA 2007.

The 28 revised full papers presented were carefully selected during two rounds of reviewing and improvement from 134 submissions. The papers address all current issues of artificial intelligence ranging from methodological and foundational aspects to advanced applications in various fields.