You are previewing Soft Computing Methods for Practical Environment Solutions: Techniques and Studies.
O'Reilly logo
Soft Computing Methods for Practical Environment Solutions: Techniques and Studies

Book Description

Soft Computing Methods for Practical Environment Solutions: Techniques and Studies presents various practical applications of Soft Computing techniques in real-world situations and problems, aiming to show the enormous potential of such techniques in solving all kinds of problems, and thus, providing the latest advances in these techniques in an extensive state-of-the-art and a vast theoretical study. Ideal for students studying AI and researchers familiarizing themselves with such techniques, so to offer recent and novel applications, helping expand and explore new areas of research.

Table of Contents

  1. Copyright
  2. Editorial Advisory Board
  3. List of Reviewers
  4. Preface
  5. Acknowledgment
  6. 1. Information Processing
    1. 1. A Soft Computing Overview: Artificial Neural Networks and Evolutionary Computation
      1. ABSTRACT
      2. INTRODUCTION
      3. ARTIFICIAL NEURAL NETWORKS
        1. Biological Basis
        2. General Operation
          1. Processing Element
          2. ANN Architecture
          3. Learning
        3. Advantages of Using ANN
      4. EVOLUTIONARY COMPUTATION
        1. General Outline of Performance
          1. Initialization
          2. Selection
          3. Crossover
          4. Mutation
          5. Replacement
          6. Stopping Criterion
      5. CONCLUSION
      6. REFERENCES
      7. ADDITIONAL READING
    2. 2. Artificial Cell Model Used for Information Processing
      1. ABSTRACT
      2. INTRODUCTION
      3. BACKGROUND
      4. BIOLOGICAL INSPIRATION
      5. PROPOSED MODEL
        1. Protein
        2. Cytoplasm
        3. Gene
        4. Operon
      6. DNA
        1. Cell
        2. Environment
      7. COMMUNICATION MODEL
      8. SEARCH METHOD
      9. TESTS
        1. Iris Flower Classification Problem
      10. FUTURE TRENDS
      11. CONCLUSION
      12. FUTURE RESEARCH DIRECTIONS
      13. REFERENCES
      14. ADDITIONAL READING
    3. 3. Soft Computing Techniques for Human-Computer Interaction
      1. ABSTRACT
      2. INTRODUCTION
      3. EYEWEAR SELECTOR
      4. INSTANT MESSAGING CONTROL
      5. CONCLUSION
      6. REFERENCES
    4. 4. LVQ Neural Networks in Color Segmentation
      1. ABSTRACT
      2. INTRODUCTION
        1. Histogram Thresholding and Color Space Clustering
        2. Region Based Approaches
        3. Edge Detection
        4. Probabilistic Methods
        5. Soft-Computing Techniques
        6. Proposed Scheme
      3. BACKGROUND ISSUES
        1. RGB Space Color
        2. Neural Networks
      4. COMPETITIVE NETWORKS
      5. LEARNING VECTOR QUANTIZATION NETWORKS
      6. ARCHITECTURE OF THE COLOR SEGMENTATION SYSTEM
      7. IMPLEMENTATION
      8. RESULTS AND DISCUSSION
      9. CONCLUSION
      10. ACKNOWLEDGMENT
      11. REFERENCES
    5. A. A. APPENDIX
      1. A.1 Creating a Competitive Neural Network in Matlab®
      2. A.2 Training a Competitive Neural Network in Matlab©
      3. A.3 Creating and Training an LVQ Neural Network
      4. A.4 Auxiliary Functions
    6. 5. 3D Modelling and Artificial Intelligence: A Descriptive Overview
      1. ABSTRACT
      2. INTRODUCTION
      3. BACKGROUND
      4. SOME LINKS BETWEEN 3D MODELLING AND AI TECHNIQUES IN ENGINEERING
      5. FUTURE TRENDS
      6. CONCLUSION
      7. ACKNOWLEDGMENT
      8. REFERENCES
    7. 6. User Modeling in Soft Computing Framework
      1. ABSTRACT
      2. INTRODUCTION
      3. WHAT IS USER MODELING (UM)?
      4. SOFT COMPUTING IN USER MODELING
      5. USER MODELING AS SEQUENCE LEARNING
      6. APPLICATIONS: PROFILING UNIX USERS USING SOFT COMPUTING TECHNIQUES
        1. First User Model: Term Weighting - TFIDF
        2. Second User Model: Hidden Markov Models (HMM)
        3. Third User Model: Graphical Model - Bayesian Networks
        4. Fourth User Model: Artificial Neural Networks
        5. Comparing the Proposed Techniques in this Section
      7. FUTURE RESEARCH DIRECTIONS
      8. CONCLUSION
      9. REFERENCES
      10. ENDNOTE
  7. 2. Industrial Applications
    1. 7. Electromagnetic Optimization Using Genetic Algorithms
      1. ABSTRACT
      2. INTRODUCTION
      3. BACKGROUND TECHNIQUES
        1. Genetic Agorithms (GAs)
        2. Choice of Fitness Function in GA
      4. CASE STUDY APPLICATION
        1. Design of a Matching Network for Microstrip Antenna
        2. The Quarter-Wave Transformer
        3. Parameters for Optimization
        4. Fitness Function
      5. RESULTS AND DISCUSSIONS
      6. FUTURE RESEARCH DIRECTIONS
      7. CONCLUSION
      8. REFERENCES
      9. ADDITIONAL READING
    2. 8. Motor Vehicle Improvement Preference Ranking: A PROMETHEE and Trigonometric Differential Evolution Analysis of their Chemical Emissions
      1. ABSTRACT
      2. INTRODUCTION
      3. BACKGROUND
      4. PROMETHEE
        1. Uncertainty Analysis – Trigonometric Differential Evolution
      5. MOTOR VEHICLE IMPROVEMENT PREFERENCE RANKING
        1. Motor Vehicle Chemical Emissions Data Set
        2. PROMETHEE Analyses of Motor Vehicle Chemical Emissions Data Set
        3. Preference Rank Improvement 'Uncertainty' Based Analysis of Motor Vehicles Using TDE
        4. Progressive Targeted Change of Chemical Emissions Levels for Preference Rank Improvement of a Motor Vehicle
      6. FUTURE TRENDS
      7. CONCLUSION
      8. REFERENCES
      9. KEY TERMS AND DEFINITIONS
    3. 9. A Soft Computing System for Modelling the Manufacture of Steel Components
      1. ABSTRACT
      2. INTRODUCTION
      3. AN INDUSTRIAL PROCESS FOR STEEL COMPONENTS MODELLING
        1. Analyse of the Internal Structure of the Data Set
        2. System Identification and the Knowlegde Based Systems
        3. The Identification Criterion
      4. MODELLING STEEL COMPONENTS: AN INDUSTRIAL TASK
        1. Application of the Two Phases of the Modelling System
      5. FUTURE RESEARCH DIRECTIONS
      6. CONCLUSION
      7. ACKNOWLEDGMENT
      8. REFERENCES
    4. 10. Soft Computing Techniques in Civil Engineering: Time Series Prediction
      1. ABSTRACT
      2. INTRODUCTION
      3. DEVELOPMENT SYSTEM
        1. Creep Phenomenon of Structural Concrete
        2. Predictive Models of Creep
        3. Proposed Method
        4. Schema of the Algorithm
        5. Results
        6. Future Research Directions
      4. CONCLUSION
      5. ACKNOWLEDGMENT
      6. REFERENCES
      7. ADDITIONAL READING
      8. KEY TERMS AND DEFINITIONS
    5. 11. Intrinsic Evolvable Hardware Structures
      1. ABSTRACT
      2. INTRODUCTION
      3. BACKGROUND
        1. Evolutionary Algorithms
        2. Parameter Problem Coding
          1. Fitness Functions
          2. Selection Functions
          3. Operators Used in Evolutionary Algorithms
          4. Terms Used in Genetic Algorithms
        3. Reconfigurable Circuits
        4. Evolvable Hardware
          1. Hardware Implementations of Evolutionary Algorithms
          2. Synthesis of Logic Circuits Using Evolutionary Algorithms
          3. Classic Synthesis Versus Evolvable Synthesis
      4. INTRINSIC EVOLVABLE HARDWARE SYSTEM
        1. Hardware Genetic Algorithm
          1. Combinational Networks Used for Hardware Implementation of Basic Functions
            1. Sorting Networks
            2. Selection Module Implementation Using Sorting Networks
          2. Hardware Implementation of the Genetic Algorithm
            1. Coding Schema
        2. Dynamic Reconfigurable Circuit
        3. Reconfigurable Hardware and HGA System. Example of Applications
          1. Application 1: Boolean Function Evolvable Synthesis
          2. Application 2: Fault Tolerance Evolvable Circuit
          3. Application 3: Real Time Command for a Power Supply Regulator
      5. FUTURE RESEARCH DIRECTIONS
      6. CONCLUSION
      7. REFERENCES
      8. ADDITIONAL READING
    6. 12. Connectionist Systems and Signal Processing Techniques Applied to the Parameterization of Stellar Spectra
      1. ABSTRACT
      2. INTRODUCTION
      3. SPECTRALIB: A LIBRARY OF SYNTHETIC SPECTRA FoR GAIA RVS
      4. INPUT DOMAINS
      5. EXPERIMENT DESIGN
      6. EQUIPMENT AND TOOLS
      7. NATURE OF THE SPECTRAL SIGNAL AND PREDICTION OF THE NOISE LEVEL
      8. PARAMETERIZATION ALGORITHM
      9. RESULTS
      10. CONCLUSION
      11. ACKNOWLEDGMENT
      12. REFERENCES
  8. 3. Biomedical Approaches
    1. 13. Automatic Arrhythmia Detection
      1. ABSTRACT
      2. INTRODUCTION
      3. BACKGROUND
      4. ARRHYTHMIA DETECTION SYSTEM
        1. ECG Data
          1. Types of Arrhythmia to Detect
        2. Feature Extraction
        3. Classification System
          1. Support Vector Machine (SVM)
      5. FUTURE RESEARCHS DIRECTIONS
      6. CONCLUSION
      7. REFERENCES
      8. ADDITIONAL READING
    2. 14. GA-Based Data Mining Applied to Genetic Data for the Diagnosis of Complex Diseases
      1. ABSTRACT
      2. INTRODUCTION
      3. BACKGROUND
        1. Statistical Methods
        2. Data Mining Methods
          1. Classification Methods
          2. Clustering Methods
        3. Soft Computing
          1. Usage of Genetic Algorithms for Rule Extraction
      4. METHODS
        1. Issues, Controversies, Problems
        2. Solutions and Recommendations
          1. Definitions
            1. Terms Used
            2. Structure of the Association Rules
          2. Genetic Algorithm
            1. GA Individual
            2. Fitness Function
            3. Operators
            4. Parameters
          3. Iterative Algorithm
      5. RESULTS
        1. Test Bed
          1. Real Clinical Data
          2. Generation of Artificial Input Data
          3. UCI Machine Learning Repository
        2. Results
          1. Data Mining Techniques
          2. Results
      6. FUTURE RESEARCH DIRECTIONS
      7. CONCLUSION
      8. ACKNOWLEDGMENT
      9. REFERENCES
      10. ADDITIONAL READING
    3. 15. Improving Ontology Alignment through Genetic Algorithms
      1. ABSTRACT
      2. INTRODUCTION
      3. BACKGROUND
        1. Ontology Alignment
        2. Genetic Algorithms
        3. Related Work
      4. A GENETIC ALGORITHM BASED APPROACH TO OPTIMIZE THE AGGREGATION OF MULTIPLE SEMANTIC SIMILARITY MEASURES
        1. Description of the Problem
        2. Encoding Mechanism and Initialization
        3. Reproduction Methods
        4. Fitness or Objective Function
        5. Stop Criterion
        6. Example
      5. FUTURE RESEARCH DIRECTIONS
      6. CONCLUSION
      7. ACKNOWLEDGMENT
      8. REFERENCES
      9. ADDITIONAL READING
      10. ENDNOTES
  9. 4. Natural Environment Applications
    1. 16. Characterization and Modelization of Surface Net Radiation through Neural Networks
      1. ABSTRACT
      2. INTRODUCTION
      3. BACKGROUND
        1. The Multilayer Perceptron
        2. Self-Organizing Map
      4. ANALYSIS OF NET RADIATION WITH ARTIFICIAL NEURAL NETWORKS
        1. 1. Issues, Controversies, Problems
        2. 2. solutions and Recommendations
      5. FUTURE RESEARCH DIRECTIONS
      6. CONCLUSION
      7. REFERENCES
      8. ADDITIONAL READING
    2. 17. Application of Machine Learning Techniques in the Study of the Relevance of Environmental Factors in Prediction of Tropospheric Ozone
      1. ABSTRACT
      2. INTRODUCTION
      3. LINEAR MODELS EMPLOYED IN ENVIRONMENTAL SCIENCE
      4. ARTIFICIAL NEURAL NETWORKS
      5. TROPOSPHERIC OZONE PREDICTION (1 YEAR) AND THE STUDY OF THE RELEVANCE OF INPUT VARIABLES
      6. TROPOSPHERIC OZONE PREDICTION (3 YEARS) AND THE RELEVANCE OF INPUT VARIABLES ON AN HOURLY BASIS
      7. FUTURE TRENDS
      8. CONCLUSION
      9. REFERENCES
    3. 18. Evolutionary Lagrangian Inverse Modeling for PM10Pollutant Dispersion
      1. ABSTRACT
      2. INTRODUCTION
      3. BACKGROUND
      4. ISSUES, CONTROVERSIES, PROBLEMS
      5. DEVELOPMENT OF THE MODEL.
        1. Estimated Emissions Submodel
        2. Fitness Function – Genetic Estimation substructure
        3. Validation of the Submodel of Estimation
        4. Submodel of Integrated Interpolation Representation, Densified EKFN Model
        5. Genetic Substructure Interpolation Representation
        6. Fitness Function – Genetic Substructure of Estimation
        7. Analysis of the Spatial Behavior of the Concentration of PM10 within the Study Area (Case of Cell Densification in the Output Layer)
        8. Asynchronous Mechanism of Evolution
          1. Mechanism of Evolution - Submodel for Estimating Emissions
          2. Mechanism of Evolution - Submodel of Interpolation Representation
          3. Space Sensibility Mechanism
      6. CONCLUSION
      7. FUTURE RESEARCH DIRECTIONS
      8. REFERENCES
    4. 19. Artificial Intelligence Applied to Natural Resources Management
      1. ABSTRACT
      2. INTRODUCTION
      3. RELATED WORKS
      4. CELLULAR AUTOMATA
        1. Some Examples of Categorizations
      5. MULTI-AGENT-BASED SIMULATION
        1. GMABS Methodology
        2. ViP-JogoMan: A case Study in Natural Resources Management
          1. Game Description
          2. Virtual Players
          3. Tests and Preliminary Results
      6. FUTURE DIRECTIONS AND CONCLUSIONS
      7. REFERENCES
    5. 20. Applications of Self-Organizing Maps to Address Environmental Studies
      1. ABSTRACT
      2. INTRODUCTION
        1. Theory of Self-Organizing Maps, Counter-Propagation ANN and MOLMAP
          1. Notation
          2. Self-Organizing Maps
          3. Counter-Propagation Artificial Neural Networks
          4. MOLMAP Approach
          5. Software
      3. CASES OF STUDY
        1. Case Study 1: Analysis of Soil Pollution
          1. Pattern Recognition by Kohonen SOMs
          2. Classification by CP-ANNs
        2. CASE Study 2: Identification and Monitoring of Oil Spillages
          1. Pattern Recognition by Kohonen SOM
          2. Pattern Recognition by MOLMAP Approach
      4. REFERENCES
    6. 21. Neural Models for Rainfall Forecasting
      1. ABSTRACT
      2. INTRODUCTION
      3. MATERIALS AND METHODS
        1. Data sets
        2. Modeling Techniques
          1. Linear Models
          2. Artificial Neural Networks
          3. Experimental Design
        3. Results
      4. FUTURE RESEARCH
      5. CONCLUSION
      6. ACKNOWLEDGMENT
      7. REFERENCES
      8. ADDITIONAL READING
  10. Compilation of References
  11. About the Contributors