Hybrid Artificial Intelligent Systems

Name: Hybrid Artificial Intelligent Systems | 7th International Conference, HAIS 2012, Salamanca, Spain, March 28-30th, 2012, Proceedings, Part II
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

7th International Conference, HAIS 2012, Salamanca, Spain, March 28-30th, 2012, Proceedings, Part II

Emilio S. Corchado Rodriguez Vaclav Snasel Ajith Abraham Michal Wozniak Manuel Grana Sung-bae Cho(Editor)

Springer (Publisher)

Published on 15. March 2012

XXXII, 606 pages

E-Book

PDF with digital watermarking

System requirements

978-3-642-28931-6 (ISBN)

€53.49incl. 7% vat

System requirements

for PDF with digital watermarking

E-Book Single Licence

Available for download

Description

More details

Other editions

Content

Title
Preface
Organization
Table of Contents
Special Sessions
Methods of Classifier Fusion
Hybrid Decision Tree Architecture Utilizing Local SVMs for Multi-Label Classification
Introduction
Related Work
The Landscape of MLC Approaches
Combining Decision Trees and SVMs
Integration of Decision Trees and SVMs
Experiments
Datasets and Experimental Setup
Results
Conclusions
References
Ensemble Pruning Using Harmony Search
Introduction
Ensemble of Classifiers
Generating Models
Pruning the Ensemble
Combining the Models
Harmony Search
Harmony Search for Ensemble Pruning
Defining the Optimization Problem
Initialisation Step
New Harmony Improvisation and Memory Update
Proposed Model
Experimental Analysis
Conclusion and Future Work
References
A First Study on Decomposition Strategies with Data with Class Noise Using Decision Trees
Introduction
Classification with Noisy Data
Addressing Multi-class Classification by Decomposition
Decomposition Strategies for Multi-class Problems
One-vs-One Decomposition Scheme
Experimental Framework
Base Datasets
Inducing Noise in Datasets
Analysis Methodology
Analysis of the One-vs-One Decomposition Strategy with Data with Class Noise
Concluding Remarks
References
Combining the Advantages of Neural Networks and Decision Trees for Regression Problems in a Steel Temperature Prediction System
Introduction
Neural Network-Based Rule Extraction for Regression Problems
Temperature Control in the Electric Arc Furnace
Methodology
Data Preprocessing
Network Construction
Network Training
Network Test
Experimental Evaluation
Experimental Methodology
Datasets
Experimental Results
Logical Rules
Conclusions
References
Transfer Learning Approach to Debt Portfolio Appraisal
Introduction
Related Work
Debt Portfolio Value Prediction
Learning Based on Similarity and Transfer Learning Techniques for Debt Portfolio Appraisal
Experiments and Results
Conclusions
References
Generalized Weighted Majority Voting with an Application to Algorithms Having Spatial Output
Introduction
Majority Voting
The Generalized Majority Voting
Modifications on the Decision Rule
Weighted Voting System
Generalized Weighted Voting System
Weighted Majority Voting in OD Detection
Experimental Results
Conclusion and Future Plans
References
HAIS for Computer Security (HAISFCS)
Towards the Reduction of Data Used for the Classification of Network Flows
Introduction
Flow Features-Based Spam Detection
The Spam Data Set
Data Preprocessing
Model Development and Data Reduction
Key Objectives
The Evaluation Method
Experimental Results
Summary
References
Encrypting Digital Images Using Cellular Automata
Introduction
Mathematical Preliminaries
Chaotic Discrete Maps
Cellular Automata
The Encryption Scheme
The Decryption Protocol
Computational Complexity
The Security Analysis
Brutte-Force Attacks and Statistical Analysis
Sensitivity to Initial Conditions
Differential Attack
Other Cryptanalytic Attacks
Conclusions and Further Work
References
Self-Organizing Maps versus Growing Neural Gas in Detecting Data Outliers for Security Applications
Introduction
Previous Work
Problem Definition
Feature Extraction and Formation of Model
Attack Detection
Recovery from Attacks
Developed Techniques
Experimental Evaluation
Conclusions
References
Cryptographic Applications of 3x3 Block Upper Triangular Matrices
Introduction
Description
Block Upper Triangular Matrices
Set Cardinality
Examples
Key Exchange Scheme
Pseudorandom Number Generator
Conclusions
References
Digital Chaotic Noise Using Tent Map without Scaling and Discretization Process
Introduction
Piecewise Lineal Chaotic Maps: Tent Map
Tent Map Digital Implementation
Map Tent without the Processes of Scaling and Discretization in Digital Applications
Conclusions
References
Data Mining: Data Preparation and Analysis
Hubness-Aware Shared Neighbor Distances for High-Dimensional k-Nearest Neighbor Classification
Introduction
Related Work
Shared Neighbor Distances
Hubs: Very Frequent Nearest Neighbors
Hubness-Aware Classification Methods
Hubness-Aware Shared-Neighbor Distances
Experiments
Overview of the Data
Image Classification with Nearest-Neighbor Methods
Conclusions and Future Work
References
Comparison of Competitive Learning for SOM Used in Classification of Partial Discharge
Introduction
Partial Discharge: Concepts
Self Organizing Map (SOM)
Winner Takes All
The Frequency Sensitive Competitive Algorithm
The Rival Penalized Competitive Learning Algorithm
Analysis of PD Data
Conclusion
References
Identification of Different Types of Minority Class Examples in Imbalanced Data
Introduction
Distribution of Examples in the Minority Class
Assessing Types of Examples
Analysing Real-World Datasets
Datasets
Labelling Results and Categorization of Datasets
Impact of Different Data Categories on Classifiers
Conclusions
References
Non-Disjoint Discretization for Aggregating One-Dependence Estimator Classifiers
Introduction
AODE
HAODE
Disjoint vs. Non-Disjoint Discretization
Equal Frequency Discretization
Non-Disjoint Discretization
NDD Adapted to AODE and HAODE
Experiments
Conclusions
References
An Adaptive Hybrid and Cluster-Based Model for Speeding Up the k-NN Classifier
Introduction
Related Work
The Proposed Classification Model
Speed-Up Data Structure Construction
The Fast Hybrid and Adaptive Classification Algorithm
Discussion
Performance Evaluation
Experimental Setup
Comparisons
Conclusion and Future Work
References
A Co-evolutionary Framework for Nearest Neighbor Enhancement: Combining Instance and Feature Weighting with Instance Selection
Introduction
Background
Co-evolution
Instance Selection
Weighting Schemes
Proposed Model
CIW-NN Subcomponents
Fitness Function
Co-evolutionary Model
Experimental Study
Data Sets
Algorithms and Parameters
Results Obtained
Statistical Study
Conclusions and Future Work
References
Improving Multi-label Classifiers via Label Reduction with Association Rules
Introduction
Multi-label Classification
The Data Transformation Approach
The Method Adaptation Approach
New Measures and Evaluation Metrics
LRwAR: Label Reduction with Association Rules
The Problem of Dimensionality in the Label Space
LRwAR: A Method for Reducing Dimensionality in the Label Space
Method Description.
Method Implementation.
Experimental Framework, Results and Analysis
Experimental Framework
Results and Analysis
Conclusions and Future Work
References
A GA-Based Wrapper Feature Selection for Animal Breeding Data Mining
Introduction
Data and Methods
Data
Classifiers
GA-Based Wrapper Feature Selection
Experimental Settings
Experimental Results
Conclusions
References
A Simple Noise-Tolerant Abstraction Algorithm for Fast k-NN Classification
Introduction
State-of-the-Art Abstraction and Filtering Algorithms
Condensing Nearest Neighbor Rule
Reduction by Space Partitioning
Reduction through k-Means Clustering
Performance Evaluation
Experimental Setup
Comparisons
Conclusion
References
Hybrid Artificial Intelligence Systems inManagement of Production Systems
Adaptive Inventory Control in Production Systems
Introduction
Inventory Control
The ANFIS Controller
Simulation Setup
Results
Conclusion
References
Hybrid Artificial Intelligence System in Constraint Based Scheduling of Integrated Manufacturing ERP Systems
Introduction
Description of Constraint Based Scheduling
Methodology of Creation of Models for Selection and Forecasting of Tools
Describe of Simple and Hybrid Neural Networks
Neural Networks for Selection of Tools for Constraint Based Scheduling
Data Preparation.
Experiments with Classification Models.
Neural Networks as Forecasting Models of Tool Use in Different Intervals of the Time
Data Preparation.
Assessment of Forecasting Models.
Experiments with Forecasting Models.
Conclusion
References
Intelligent Data Processing in Recycling of Household Appliances
Introduction
Refrigerator Recycling
Scanning of a Refrigerator Housing
Data Processing
Summary
References
Assessment of Risk in a Production System with the Use of the FMEA Analysis and Linguistic Variables
Introduction
Identification and Assessment of Risk in Production Systems
Determination of the Risk Priority Number (RPN) in the FMEA Method
Determination of Risk in the Process of Haulage of Excavated Material by Belt Conveyors, Using Linguistic Variables
Conclusion
References
Hybrid Methods Aiding Organisational and Technological Production Preparation Using Simulation Models of Nonlinear Production S
Introduction
Characteristics of Production Processes
Forecasted Sales Volumes
Production and Organisational Parameters
Layout Planning
Determining the Required Number of Machine Tools
Building a Simulation Model of the Planned Production System
The Experiment
Analysis of the Results
References
The Concept of Intelligent System for Horizontal Transport in a Copper Ore Mine
Characteristics of the Transport Infrastructure in Copper Ore Mines
A concept of Implementing the Intelligent Suspension System for Roller Sets in Band Conveyors
Telematics Transport System in Ore Mine
Summary
References
Hybrid Artificial Intelligent Systems for Ordinal Regression
Integration Production Planning and Scheduling Systems for Determination of Transitional Phases in Repetitive Production
Introduction
Problem Formulation
Construction of Schedule for the Transition Phase
Practical Example
Summary
References
The Hybrid Method of Knowledge Representation in a CAPP Knowledge Based System
Introduction
The Variant Method
The Generative Method
The Object Design Representation
The Manufacturing Knowledge Representation
Meta Inference Rules
The Manufacturing Process Structure Element Selection Rules
Summary
References
An Experimental Study of Different Ordinal Regression Methods and Measures
Introduction
Methods
Proportional Odd Model (POM)
Gaussian Processes for Ordinal Regression (GPOR)
Support Vector Machines
Support Vector Machines for Binary Classification.
Support Vector Machines for Ordinal Regression (SVOR).
Extended Binary Classification (EBC)
Measures
Experiments
Conclusions
References
Neural Network Ensembles to Determine Growth Multi-classes in Predictive Microbiology
Introduction
Data Preprocessing
Negative Correlation Learning for Ordinal Regression
Experimental Setup
Dataset: Staphylococcus Aureus
Machine Learning Methods Used for Comparison Purposes and Experimental Design
Ordinal Classification Evaluation Metrics
Experimental Results
Conclusions
References
Ordinal Classification Using Hybrid Artificial Neural Networks with Projection and Kernel Basis Functions
Introduction
Model
Algorithm
Experiments
Conclusions
References
Hybrid Metaheuristics for Combinatorial Optimization and Modelling Complex Systems
A Genetic Programming Approach for Solving the Linear Ordering Problem
Introduction
The Genetic Program for Solving the LOP
Genetic Representation
Initial Population
The Fitness Value
Genetic Operators
Selection
Genetic Parameters
Computational Results
Conclusions
References
Comparison of Fuzzy Functions for Low Quality Data GAP Algorithms
Introduction
Representation of the Vagueness in GAP Models
Representation of the GAP Individual
Genetic Operators
Fitness Fuction
Experiments and Numerical Results
Experiments
Standard Data Sets
Numerical Results
Conclusions and Future Work
References
A Simple Artificial Chemistry Model for Nash Equilibria Detection in Large Cournot Games
Introduction
Artificial Chemistries
Nash Equlibria - Definition and Generative Relation
Nash Ascendancy Relation
Artificial Chemistry for Nash Equilibria
The Addition Reaction
The Dissociation Reaction
Numerical Experiments
Cournot Oligopoly
Parameter Settings
Results
Conclusions and Further Work
References
Dynamics of Networks Evolved for Cellular Automata Computation
Introduction
Cellular Automata Computation: Density Classification Task
Network Based CAs
Evolution and Dynamics of Network Topologies for DCT
Evolving Network Topologies for DCT
Network Dynamics Analysis
Conclusions
References
From Likelihood Uncertainty to Fuzziness: A Possibility-Based Approach for Building Clinical DSSs
Introduction
Related Concepts
Properties of Probability-Possibility Transformation
Probability-Possibility Transformations
Fuzzy Sets Characterization
The Proposed Approach
Comparison of Likelihood - Fuzzy Set Transformations
Conclusions
References
Combining Metaheuristic Algorithms to Solve a Scheduling Problem
Introduction
Previous Work
Analysis of the Problem to Solve
Information Requirements
Driver Scheduling Problem Modelling
Proposed Method
Analysis of Results
Conclusions
References
Hybrid Computational Intelligence and Lattice Computing for Image and Signal Processing
Image Analysis Pipeline for Automatic Karyotyping
Introduction
State of the Art
Feature Extraction
Classification
Material and Methods
Segmentation
Feature Extraction
Classification
Results
Single Step Classification
Two Steps Classification
Conclusions and Future Work
References
A Hybrid Gradient for n-Dimensional Images through Hyperspherical Coordinates
Introduction
Hyperspherical Coordinates
From Euclidean to Hyperspherical Coordinates
Gradient
Gradient Operators
Hybrid Gradient
Experimental Results
Conclusions
References
A Hybrid Segmentation of Abdominal CT Images
Introduction
Methods
Active Learning
Random Forest Classifiers
Feature Set Construction through Active Learning
Results
Conclusion and Future Works
References
Hybrid Computational Methods for Hyperspectral Image Analysis
Introduction
The Hyperspectral Image Analysis as a Hybrid Process
Dimensionality Reduction
SSI by Space Transformation
SSI by Band Selection
Spectral Unmixing
Endmember Induction
Linear and Non-linear Spectral Unmixing
Hybrid Approaches
Conclusions
References
Image Security and Biometrics: A Review
Introduction
Image Security
Watermarking
Image Cryptography
Information Hiding on Images
Biometrics and Image Security
Imaging and Biometrics
Biometric Image Security
Conclusion
References
Cocaine Dependent Classification Using Brain Magnetic Resonance Imaging
Introduction
Methods and Materials
Support Vector Machines
Results
Conclusion
References
A Non-parametric Approach for Accurate Contextual Classification of LIDAR and Imagery Data Fusion
Introduction
Method
Data Description
Feature Set Extraction
Training Set Selection
SVMNNS
Experiments
Results
Discussion
Conclusion
References
Spherical CIELab QAMs: Associative Memories Based on the CIELab System and Quantales for the Storage of Color Images
Introduction
The Mathematical Framework: Quantales
The Quantale-Based Auto-associative Memories
The Spherical CIELab QAM
The Spherical CIELab Quantale
The Spherical CIELab QAM for the Storage and Recall of RGB Patterns
References
Fuzzy Associative Memories Based on Subsethood and Similarity Measures with Applications to Speaker Identification
Introduction
Fuzzy Subsethood and Similarity Measures
Subsethood and Similarity Measure FAMs
Application to the Problem of Speaker Identification
Concluding Remarks
References
A Novel Lattice Associative Memory Based on Dendritic Computing
Introduction
The Dendritic Lattice Based Model of ANNs
Dendritic Lattice Associative Memories
Experiments with Noisy and Corrupted Inputs
Conclusions
References
Vascular Section Estimation in Medical Images Using Combined Feature Detection and Evolutionary Optimization
Introduction
Review of Vascular Detection and Extraction
Vascular Feature Detection with Evolutionary Optimization
Experiments and Results
Conclusions and Future Work
References
Workshop
Nonstationary Models of Pattern Recognition and Classifier Combinations
Modifications of Classification Strategies in Rule Set Based Bagging for Imbalanced Data
Introduction
Related Works
Rule Induction and Classification Strategies
MODLEM Algorithm
Classification Strategies
Abstaining in Bagging
Experiments
Conclusions
References
Semi-supervised Ensemble Learning of Data Streams in the Presence of Concept Drift
Introduction
Related Works
The SSEL Algorithm
Experimental Results
Data Sets
Implementation and Parameter Setting
Results and Discussion
Conclusion and Future Works
References
Continuous User Feedback Learning for Data Capture from Business Documents
Introduction
Related Work
Overview
Incorporating User Feedback
Document Analysis
Model Adaptation
Model Reduction
Evaluation Results
Dataset
Experimental Setup
Representative Performance
Average Performance
Conclusion
References
Evolutionary Adapted Ensemble for Reoccurring Context
Introduction
Evolutionary Adopted Ensemble for Concept Drift
Experiments
Conclusions
References
Drift Detection and Model Selection Algorithms: Concept and Experimental Evaluation
Introduction
Problem Statement
Algorithms
Drift Detector
Compare all Models" Method
``Random Searching'' Method
Experiment
Set-Up
Ressults
Remarks
Final Remarks
References
Decomposition of Classification Task with Selection of Classifiers on the Medical Diagnosis Example
Introduction
Hierarchical Classifier
Experiments
Experimental Investigation
Discussion
Conclusion
References
Ensemble of Tensor Classifiers Based on the Higher-Order Singular Value Decomposition
Introduction
N-Mode Principal Component Analysis for Pattern Recognition
Tensor Algebra Concepts
Computation of the Higher-Order Singular Value Decomposition
Pattern Recognition in the Tensor Spanned Spaces
Construction of the Ensemble of HOSVD Classifiers
Experimental Results
Conclusions
References
Combining Diverse One-Class Classifiers
Introduction
Model of Pattern Recognition Task
One-Class Classification
One-Class Support Vector Machine
Proposed Approach
A Pool of Individual One-Class Classifiers
Classifier Selection
Diversity Assurance
Classifiers Fusion
Experimental Evaluation
Experimental Set-Up
Results
Results Discussion
Conclusions and Future Work
References
Author Index

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

Hybrid Artificial Intelligent Systems

Description

More details

Other editions

Additional editions

Content

System requirements