Neural Information Processing

Name: Neural Information Processing | 18th International Conference, ICONIP 2ß11, Shanghai, China, November 13-17, 2011, Proceedings, Part II
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

18th International Conference, ICONIP 2ß11, Shanghai, China, November 13-17, 2011, Proceedings, Part II

Bao-Liang Lu Liqing Zhang James Kwok(Editor)

Springer (Publisher)

Published on 12. November 2011

XXII, 778 pages

E-Book

PDF with digital watermarking

System requirements

978-3-642-24958-7 (ISBN)

€53.49incl. 7% vat

System requirements

for PDF with digital watermarking

E-Book Single Licence

Available for download

Description

Alles über E-Books | Antworten auf Fragen rund um E-Books, Kopierschutz und Dateiformate finden Sie in unserem Info- & Hilfebereich.

Alles über E-Books, Kopierschutz & Dateiformate finden Sie in unserem Info- & Hilfebereich.

The three volume set LNCS 7062, LNCS 7063, and LNCS 7064 constitutes the proceedings of the 18th International Conference on Neural Information Processing, ICONIP 2011, held in Shanghai, China, in November 2011.
The 262 regular session papers presented were carefully reviewed and selected from numerous submissions.
The papers of part I are organized in topical sections on perception, emotion and development, bioinformatics, biologically inspired vision and recognition, bio-medical data analysis, brain signal processing, brain-computer interfaces, brain-like systems, brain-realistic models for learning, memory and embodied cognition, Clifford algebraic neural networks, combining multiple learners, computational advances in bioinformatics, and computational-intelligent human computer interaction.
The second volume is structured in topical sections on cybersecurity and data mining workshop, data mining and knowledge doscovery, evolutionary design and optimisation, graphical models, human-originated data analysis and implementation, information retrieval, integrating multiple nature-inspired approaches, Kernel methods and support vector machines, and learning and memory.
The third volume contains all the contributions connected with multi-agent systems, natural language processing and intelligent Web information processing, neural encoding and decoding, neural network models, neuromorphic hardware and implementations, object recognition, visual perception modelling, and advances in computational intelligence methods based pattern recognition.

More details

Other editions

Content

Intro
Title
Preface
ICONIP 2011 Organization
Table of Contents
Cybersecurity and Data Mining Workshop
Agent Personalized Call Center Traffic Prediction and Call Distribution
Introduction
Call Center Management
Review of Call-Center IT Solutions
Existing Call Prediction Methods
Proposed Call Prediction Method
Experiments and Discussion
Conclusions
References
Mapping from Student Domain into Website Category
Introduction
The Data
The Concepts
The Matching
Discussion and Future Work
References
Entropy Based Discriminators for P2P Teletraffic Characterization
Introduction
Related Work on P2P Network Characterization
Network Level Tracing
Application Level Tracing
Hybrid Virtualization Based Approach
Data Analysis
Host-Level Analysis
Discriminators
Learning Methodology
Experiments
Experiment Settings
Analysis on Time Window Size
Analysis on Sampling Rate
Conclusion
References
Faster Log Analysis and Integration of Security Incidents Using Knuth-Bendix Completion
Introduction
Mechanized Reasoning
Resolution
Set of Support Strategy
Hyperresolution
Subsumption
Knuth-Bendix Completion
Experiment I
Internet Explorer Aurora Attack: MS979352
FTP Server Attack
Results
Experiment II
Malware Log Analysis
Results of Integration
Discussion
Conclusion
References
Fast Protocol Recognition by Network Packet Inspection
Introduction
Background and Relate Work
Motivation
Proposed Method
Evaluation Result
Conclusion
References
Network Flow Classification Based on the Rhythm of Packets
Introduction
Motivation and Related Work
Packet-Level Feature and Bayesian Networks
Network Flow Rhythm
Attributions Selection and Attributions Correlation
Bayesian Networks Parameters Estimation
Our Bayesian Network Structure
Experiment Setup
Data Set
Data Pre-processing
Result and Analysis
Conclusion
References
Data Mining and Knowledge Discovery
Energy-Based Feature Selection and Its Ensemble Version
Introduction
Energy-Based Framework for Feature Ranking
Energy-Based Learning
Framework for Feature Ranking
Feature Ranking Algorithm
Evaluation Function
Algorithm Analysis
Ensemble Feature Selection
Components of Ensemble Feature Selection
Stability Estimation
Experiments
Experimental Results for Single Feature Selection
Experimental Results for Ensemble Feature Selection
Conclusions
References
The Rough Set-Based Algorithm for Two Steps
Introduction
Literature Review and Problem Statement
Rough Set-Based Algorithm for Two-Step
Illustrative Example
Conclusion
References
An Infinite Mixture of Inverted Dirichlet Distributions
Introduction
The Infinite Model
The Finite Inverted Dirichlet Mixture Model
The Infinite Inverted Dirichlet Mixture Model
Priors and Conditional Posteriors
Experimental Results
Conclusion
References
Multi-Label Weighted k-Nearest Neighbor Classifier with Adaptive Weight Estimation
Introduction
A Novel Multi-Label Weighted k-Nearest Neighbor Algorithm
Traditional Multi-class Weighted kNN Method
Multi-Label Weighted kNN Method Based on Bayesian Theorem
Adaptive Weight Estimation Method
Experiments
Five Evaluation Measures and Two Data Sets
Tuning k Value for Four kNN-Based Classifiers
Comparison Study on Two Test Data Sets
Conclusions
References
Emotiono: An Ontology with Rule-Based Reasoning for Emotion Recognition
Introduction
'Emotiono' Ontology Construction
Affective Model Applied in the 'Emotiono' Ontology
The Structure of the 'Emotiono' Ontology
Data Processing Method
Data Collection
Data Preprocessing and EEG Features
Rule-Based Reasoning
The Reason of Generating Rules by C4.5
Emotion Recognition Rules
Reasoning Results
Conclusions and Future Work
References
Parallel Rough Set: Dimensionality Reduction and Feature Discovery of Multi-dimensional Data in Visualization
Introduction
Rough Set Theory Background
Classic Rough Set
Variable Precision Rough Set
Parallel Rough Set System
Dimensionality Reduction via VPRS
Feature Discovery via Rule Induction
Dimension Reorder to Enhance Visual Structure
Case Studies Using PRS
Comparison with Dimensionality Reduction Techniques
Conclusion
References
Feature Extraction via Balanced Average Neighborhood Margin Maximization
Introduction
Average Neighborhood Margin Maximization
Balanced Average Neighborhood Margin Maximization
Side Information
BANMM
Experimental Results
Conclusion
References
The Relationship between the Newborn Rats' Hypoxic-Ischemic Brain Damage and Heart Beat Interval Information
Background
Experiments
Data Collection
Data Analyzing
Multiple Linear Regression Analysis
Successive Multiple Linear Regression Analysis
Results
Conclusions
References
A Robust Approach for Multivariate Binary Vectors Clustering and Feature Selection
Introduction
A Model for Simultaneous Clustering, Feature Selection and Outliers Rejection
The Model
Model Learning
Experimental Results
Handwritten Digit Recognition
Visual Scenes Categorization
Conclusion
References
The Self-Organizing Map Tree (SOMT) for Nonlinear Data Causality Prediction
Introduction
Background
Nonlinear Data Relationship Analysis Using SOM
Nonlinear Data Prediction Analysis Using BPN
Issues of Nonlinear Data Prediction Process
The Self-Organizing Map Tree (SOMT)
Structure of the SOMT and the Prediction Processes
Weight Vector Linking Method for the Prediction Processes
Experimental Results and Discussion
Conclusion
References
Document Classification on Relevance: A Study on Eye Gaze Patterns for Reading
Introduction
Eye Gaze for Reading
The Experiment
Experiment Design
Experimental Setup
Participants
Analysis and Results
Gaze Points to Fixations
Scoring the Participants
Statistical Analysis
Further Analysis by ANN
Discussion
References
Multi-Task Low-Rank Metric Learning Based on Common Subspace
Introduction
Multi-Task Low-Rank Metric Learning
Notation and Problem Definition
Multi-Task Framework for Low-Rank Metric Learning
Optimization
Special Case
Experiments
Illustration on Synthetic Data
Experiment on Real Data
Conclusion
References
Reservoir-Based Evolving Spiking Neural Network for Spatio-temporal Pattern Recognition
Introduction
Spatio-temporal Pattern Recognition with reSNN
Reservoir
Experiments
Data Set
Setup
Results
Parameter and Feature Optimization of reSNN
Conclusion and Future Directions
References
An Adaptive Approach to Chinese Semantic Advertising
Introduction
Related Work
Chinese Semantic Advertising Architecture
Preprocessing Chinese Web Pages and Advertisements
The Ontology
Extracting Related Phrases for Ontology
The Distance Function
Evaluation
Experiment Setup
Experiment Results
Conclusion and Future Work
Reference
A Lightweight Ontology Learning Method for Chinese Government Documents
Introduction
Related Work
Ontology Learning for Chinese Government Documents
Preprocess
Term Extraction
Government Rule Classification
Triple Creation
RDF Generation
Evaluation
Experiment Setup
Results
Conclusion and Future Work
References
Relative Association Rules Based on Rough Set Theory
Introduction
Literature Review and Problem Statement
Incorporation of Rough Set for Classification Processing
Conclusion and Future Works
References
Scalable Data Clustering: A Sammon's Projection Based Technique for Merging GSOMs
Introduction
Background
Self-Organizing Map
Growing Self-Organizing Map
Sammon's Projection
The Parallel GSOM Algorithm
Data Partitioning
Parallel GSOM Training
Merging Process
Refining Process
Results
Accuracy
Performance
Discussion
References
A Generalized Subspace Projection Approach for Sparse Representation Classification
Introduction
Sparse Representation Classification
Subspace Projection for Sparse Representation Classification
Subspace of Each Class
Maximal Linearly Independent Set of Each Class
Experiments
Parameters Setting
Experimental Results
Conclusion and Future Work
References
Evolutionary Design and Optimisation
Macro Features Based Text Categorization
Introduction
Macro Feature Extraction
Clustering Based Method MFCl
Centroid Based Method MFCe
Databases and Experimental Setting
Databases
Experimental Setting
Experimental Results
Performance Comparison of Different Methods
Effectiveness of Labeled Data in MFCl
Effectiveness of Labeled Data in MFCe
Comparison of MFCl and MFCe
Conclusion
References
Univariate Marginal Distribution Algorithm in Combination with Extremal Optimization (EO, GEO)
Introduction
Univariate Marginal Distribution Algorithm
Extremal Optimization Algorithm
Suggested Algorithm
Experiments and Results
Graph Bi-partitioning Problem
Multiprocessor Scheduling Problems
Conclusion
References
Promoting Diversity in Particle Swarm Optimization to Solve Multimodal Problems
Introduction
Preliminaries
Particle Swarm Optimization
Population Diversity Definition
Diversity Promotion
Experimental Study
Benchmark Test Functions and Parameter Setting
Experimental Results
Diversity Analysis and Discussion
Conclusion
References
Analysis of Feature Weighting Methods Based on Feature Ranking Methods for Classification
Introduction
Selection of Rankings
Information Theory Based Feature Rankings
Decision Tree Rankings
Feature Rankings Based on Probability Distribution Distance
Methods of Feature Weighting for Ranking Vectors
Testing Methodology and Results Analysis
Summary
References
Simultaneous Learning of Instantaneous and Time-Delayed Genetic Interactions Using Novel Information Theoretic Scoring Technique
Introduction
The Representational Framework
Our Proposed Scoring Metric, CCIT
Some Properties of CCIT Score
The Search Strategy
Experimental Evaluation
Synthetic Network
Real-Life Biological Data
Conclusion
References
Resource Allocation and Scheduling of Multiple Composite Web Services in Cloud Computing Using Cooperative Coevolution Genetic Algorithm
Introduction
Problem Definition
A Cooperative Coevolutionary Genetic Algorithm
Problem Decomposition
Interaction between Subpopulations
Algorithm Description
Experimental Results
Experiments on the Number of Composite Web Services
Experiments on the Number of Abstract Web Services
Experiments on the Number of Candidate Cloud Services
Conclusion and Future Work
References
Graphical Models
Image Classification Based on Weighted Topics
Introduction
Classification Based on Weighted Topics
Image Representation
pLSA Model for Image Analysis
Learning Weights for Topics
Classifiers with Weights
Experiments
Experimental Setup
Experimental Results
Conclusions
References
A Variational Statistical Framework for Object Detection
Introduction
Model Specification
Variational Learning
Experimental Results: Object Detection
Conclusion
References
Performances Evaluation of GMM-UBM and GMM-SVM for Speaker Recognition in Realistic World
Introduction
Gaussian Mixture Model (GMM)
Support Vector Machines (SVM)
GMM-UBM and GMM-SVM Systems
Results and Discussion
Experimental Protocol and Data Collection
Speaker Recognition in Quiet Environment Using GMM and SVM
Speaker Recognition in Noisy Environments Using GMM and SVM
Speaker Recognition in Quiet Environment Using GMM-UBM and GMM-SVM
Speaker Recognition in Noisy Environments Using GMM-UBM and GMM-SVM
Conclusion
References
SVM and Greedy GMM Applied on Target Identification
Introduction
Classification Scheme
Feature Extraction
Modelisation
Support Vector Machine (SVM)
Classification
Radar System and Data Collection
Results
Conclusion
References
Speaker Identification Using Discriminative Learning of Large Margin GMM
Introduction
Overview on Large Margin GMM with Diagonal Covariances (LM-dGMM)
LM-dGMM Training with k-Best Gaussians
Description of the New LM-dGMM Training Algorithm
Handling of Outliers
The GSL-NAP System
Symmetrical Factor Analysis (SFA)
Experimental Results
Conclusion
References
Sparse Coding Image Denoising Based on Saliency Map Weight
Introduction
Saliency Map
Sparse Coding with Saliency
Image Reconstruction with Saliency
Experiment and Result
Experiment
Result Discussion
Discussion
Reference
Human-Originated Data Analysis and Implementation
Expanding Knowledge Source with Ontology Alignment for Augmented Cognition
Introduction
Related Work
Ontology as Graph
Ontology Alignment
Ontology Alignment with Similarity
Similarity between Ontology Entities
Graph Kernel
Modified Graph Kernel
Composite Kernel
Experiments
Experimental Data and Setting
Experimental Result
Conclusion
References
Nystr¨om Approximations for Scalable Face Recognition: A Comparative Study
Introduction
Methods
KPCA in a Nutshell
Nyström Approximation for KPCA
Nyström KPCA Ensemble
Nyström + Randomized SVD
Numerical Experiments
Random Sampling with Class Label Information
Is Nyström Really Helpful for Face Recognition?
How Many Samples/Principal Components are Needed?
Comparison with Nyström KPCA Ensemble
Nyström vs. rSVD vs. Nyström + rSVD
Experiments on Large-Scale Data
Conclusions
References
A Robust Face Recognition through Statistical Learning of Local Features
Introduction
Representation of Facial Images Using SIFT
Face Recognition through Learning of Local Features
Statistical Learning of Local Features for Facial Images
Weighted Distance Measure for Face Recognition
Experimental Comparisons
Facial Image Database with Occlusions
Experimental Results
Conclusions
References
Development of Visualizing Earphone and Hearing Glasses for Human Augmented Cognition
Introduction
Framework of the Implemented System
Face Detection Based on Skin Color Preferable Selective Attention Model
Incremental Two-Dimensional Two-Directional PCA
Face Selection by Using Eye Movement Detection
Sound Localization and Voice Recognition
Experimental Evaluation
Conclusion and Further Work
References
Facial Image Analysis Using Subspace Segregation Based on Class Information
Introduction
Subspace Segregation
Noise Subspace
Residual Subspace
Experiments
Conclusion
References
An Online Human Activity Recognizer for Mobile Phones with Accelerometer
Introduction
Related Work
Proposed Method
Preprocessing for Direction-Free Analysis
Extracting Features
Singular Value Decomposition
Neural Network
Experiments
Running-Time Assessment
Mother Wavelet Assessment
Conclusion
References
Preprocessing of Independent Vector Analysis Using Feed-Forward Network for Robust Speech Recognition
Introduction
Review on the IVA Using Feed-Forward Separating Filter Network
Missing Feature Techniques for Robust Speech Recognition
Experiments
Concluding Remarks
References
Information Retrieval
Learning to Rank Documents Using Similarity Information between Objects
Introduction
Ranking Function with Topic Based Relationship Information
Constructing Topic Relationship Matrix Based on LDA
Ranking Function with Relationship Information among Objects
Training Algorithm of Ranking Function
Experiments
Conclusions
References
Efficient Semantic Kernel-Based Text Classification Using Matching Pursuit KFDA
Introduction
Brief Review of Kernel Methods
A Novel Semantic Kernel-Based Framework for Efficient TC
VSM Construction Mapping
Semantic Kernel Space Mapping
Approximate Semantic Kernel Subspace Mapping
Experiments
Experimental Settings
Experimental Results and Discussions
Conclusions
References
Introducing a Novel Data Management Approach for Distributed Large Scale Data Processing in Future Computer Clouds
Introduction
Distributed Data Management
Graph Neuron (GN) for Scalable Pattern Recognition
Crosstalk Issue in Graph Neuron
Hierarchical Graph Neuron (GN) for Scalable Pattern Recognition
Distributed Hierarchical Graph Neuron (DHGN)
Tests and Results
Superior Scalability
Recall Accuracy
Conclusion
References
PatentRank: An Ontology-Based Approach to Patent Search
Introduction
Related Work
Methodology
Hypothesis
Ontology-Based Ranking
Reranking Based on Similarity
Evaluation
Conclusion and Future Work
Reference
Fast Growing Self Organizing Map for Text Clustering
Introduction
Related Work
Document Vector Representation
SOM Based Text Clustering Techniques
Document Vector Representation
Fast Growing Self Organizing Map (FastGSOM) Algorithm
Initialization Phase
Training Phase
Smoothing Phase
Experimental Results and Discussion
Comparative Analysis of Accuracy and Efficiency of FastGSOM
Theoretical Analysis of the Runtime Complexity of the Algorithm
Conclusions and Future Research
References
News Thread Extraction Based on Topical N-Gram Model with a Background Distribution
Introduction
Related Work
Our Methods
Motivation
Topical N-Gram Model with Background Distribution
Inference
Experiments
Experimental Settings
Evaluation Metrics
Results and Analysis
Conclusion
References
Integrating Multiple Nature-Inspired Approaches
Alleviate the Hypervolume Degeneration Problem of NSGA-II
Introduction
Preliminaries
Dominance Relation and Pareto Optimality
Hypervolume Measure
Alleviate the Hypervolume Degeneration Problem of NSGA-`11II
Crowding Distance Selection of NSGA-II
Hypervolume Degeneration Problem
NSGA-II with Geometric Mean-Based Crowding Distance Selection and Single Point Hypervolume-Based Selection
Experimental Studies
Experimental Settings
Results and Discussions
Conclusions
References
A Hybrid Dynamic Multi-objective Immune Optimization Algorithm Using Prediction Strategy and Improved Differential Evolution Crossover Operator
Introduction
Theoretical Background
The Definition of DMO Problems and Antibody Population
Forecasting Model
Differential Evolution
Proposed Algorithm
Similarity Detection and Prediction Mechanism
The Proposed Dynamic Multi-objective Immune Optimization Algorithm
Improved DE Crossover Operator
Experimental Studies
Benchmark Problems
Experiments on Prediction Scheme and the Improved DE Crossover Operator
Experiment of Comparing HDMIO with Other Three Different Dynamic Multi-objective Optimization Algorithms
Conclusion
References
Optimizing Interval Multi-objective Problems Using IEAs with Preference Direction
Introduction
Proposed Algorithm
Preference Direction
Approximation Metric
Sorting Optimal Solutions
Applications
Preference Function
Parameter Settings
Performance Measures
Results and Analysis
Conclusions
References
Fitness Landscape-Based Parameter Tuning Method for Evolutionary Algorithms for Computing Unique Input Output Sequences
Introduction
Preliminaries
Problem Definition
Evolutionary Algorithm and Its Parameters
Fitness-Probability Cloud
Escape Probability
Fitness-Probability Cloud
Accumulated Escape Probability
Adaptive Selection of EA Parameters
The First Phase: Training Predictor
The Second Phase: Predicting `good' EA Parameters
Experimental Studies
Conclusions
References
Introducing the Mallows Model on Estimation of Distribution Algorithms
Introduction
The Permutation Flowshop Scheduling Problem
The Mallows Model
Kendall- Distance
Learning and Sampling a Mallows Model
Experiments
Analysis of the Spread Parameter
Testing the Mallows EDA on FSP
Conclusions and Future Work
References
Kernel Methods and Support Vector Machines
Support Vector Machines with Weighted Regularization
Introduction
Criterion for Classification
Weighted Regularization
Basic Support Vector Machines
Novel Weighted Regularization
Analytical Calculation of Regularization Matrices
Novel Classifiers
Experiments
Experimental Procedure
Experimental Results
Discussion
Conclusions and Future Work
References
Relational Extensions of Learning Vector Quantization
Introduction
Prototype-Based Clustering and Classification
Dissimilarity Data
Experiments
Conclusions
References
On Low-Rank Regularized Least Squares for Scalable Nonlinear Classification
Introduction
Classification with Regularized Least Squares Classifier
Low-Rank Regularized Least Squares
Low-Rank Approximation for RLS
Time Complexity Analysis
Closed-Form LOOCV Estimation
Experimental Results
Conclusions
References
Multitask Learning Using Regularized Multiple Kernel Learning
Introduction
Related Work
Multitask Learning Using Multiple Kernel Learning
Experiments
Cross-Platform siRNA Efficacy Data Set
MIT Letter Data Set
Cognitive State Inference Data Set
Computational Complexity
Conclusions
References
Solving Support Vector Machines beyond Dual Programming
Introduction
SVMs Supported by Commonwealth Points
Examples and Discussion
Conclusions and Future Work
References
Learning with Box Kernels
Introduction
Learning from Labeled Sets
Box Kernels
Experimental Results
Conclusions
References
A Novel Parameter Refinement Approach to One Class Support Vector Machine
Introduction
Support Vector Data Description (SVDD)
Weighted Support Vector Data Description (WSVDD)
WSVDD Formulation
Refinement Process
Rational of the Proposed Refinement Process
Proposition of Empirical Error
Experimental Results
Conclusion
References
Multi-Sphere Support Vector Clustering
Introduction
Support Vector Data Description (SVDD)
Support Vector Clustering (SVC)
Multi-Sphere Support Vector Clustering (MSSVC)
Problem Formulation
Calculating Radii and Centres
Calculating Matrix U
MSSVC Algorithm
Clustering Assignment
Experimental Results
Clustering Examples for SVC and MSSVC
Conclusion
References
Testing Predictive Properties of Efficient Coding Models with Synthetic Signals Modulated in Frequency
Introduction
Efficient Coding as a Sparse Code Neural Network
Synthetic Signals Modulated in Frequency
Estimating Sparse Codes from Synthetic Signals
Results
Discussion and Conclusion
References
Learning and Memory
A Novel Neural Network for Solving Singular Nonlinear Convex Optimization Problems
Introduction
Problem Formulation and Neural Design
Stability Analysis
Numerical Example
Concluding Remarks
References
An Extended TopoART Network for the Stable On-line Learning of Regression Functions
Introduction
Related Work
TopoART
Using TopoART for Regression Analysis
Training TopoART-R
Predicting with TopoART-R
Results
Conclusion
References
Introducing Reordering Algorithms to Classic Well-Known Ensembles to Improve Their Performance
Introduction
Theoretical Background
Original Training Algorithm
Ensemble Methods
Reordering Algorithms
Static Reordering
Dynamic Reordering
Experimental Setup
Experiments
Description of the Databases
Results
Analysis of the Results
Conclusions
References
Improving Boosting Methods by Generating Specific Training and Validation Sets
Introduction
Theoretical Background
Learning Process and Stopping Criteria
Description of Boosting Methods
Generating New Specific Sets for Boosting Methods
Specific Sets and Boosting
Bagging as Set Generator
Cross Validation Committee as Set Generator
Experimental Setup
Experiments
Description of the Databases
Results and Discussion
General Measurements
General Results
Analysis of the Results
Conclusions
References
Using Bagging and Cross-Validation to Improve Ensembles Based on Penalty Terms
Introduction
Theoretical Background
Learning Process and Stopping Criteria
Description of Ensemble Methodologies
Creating New Specific Sets for Penalty Based Ensembles
Combining Penalties and Specific Sets
Bagging as Set Generator
Cross Validation Committee as Set Generator
Experimental Setup
Experiments
Description of the Databases
Results and Discussion
General Measurements
General Results
Conclusions
References
A New Algorithm for Learning Mahalanobis Discriminant Functions by a Neural Network
Introduction
Mahalanobis and Bayesian Discriminant Functions
Preliminaries
Conversion of a Bayesian Discriminant Function to a Mahalanobis Discriminant Function
Construction of the Neural Network
Training of the Neural Network
Simulations
One-Dimensional Case
Two-Dimensional Case
Discussions
References
Learning of Dynamic BNN toward Storing-and-Stabilizing Periodic Patterns
Introduction
Dynamic Binary Neural Networks
GA-Based Learning Algorithm
Numerical Experiments
Conclusions
References
Self-organizing Digital Spike Interval Maps
Introduction
The Digital Spike Interval Map and Learning
Numerical Experiments
Conclusions
References
Shape Space Estimation by SOM2
Introduction
Framework
The Generative Model and Goal of the Task
Training Data
Theory and Algorithm
Distance between Shapes
Shape Space Estimation by SOM2
Observation Invariant Algorithm for SOM2
Simulations and Results
Affine Invariant Shape Space Estimation
Skyline Shape Map of Omnidirectional Images
Conclusion
References
Neocognitron Trained by Winner-Kill-Loser with Triple Threshold
Introduction
Outline of the Network
Competitive Learning with Winner-Kill-Loser
Winner-Kill-Loser with Dual Threshold
Use of Triple Threshold for Winner-Kill-Loser
Discussions
References
Nonlinear Nearest Subspace Classifier
Introduction
Nonlinear Nearest Subspace Classifier
Kernel Empirical Mapping
Nonlinear Nearest Subspace Classifier
Numerical Experiments
Setting of Algorithms and Their Parameters
Synthetic Data Set
Face Data Sets
Conclusion
References
A Novel Framework Based on Trace Norm Minimization for Audio Event Detection
Introduction
Low-Rank Matrix Representation Features
Matrix Classification and AED
Experimental Validation
Conclusions
References
A Modified Multiplicative Update Algorithm for Euclidean Distance-Based Nonnegative Matrix Factorization and Its Global Convergence
Introduction
Multiplicative Update by Lee and Seung
Modifications to Multiplicative Update Rule and Optimization Problem
Global Convergence of Modified Algorithm
Conclusion
References
A Two Stage Algorithm for K-Mode Convolutive Nonnegative Tucker Decomposition
Introduction
Two Stage Algorithm for K-Mode Convolutive Nonnegative Tucker Decomposition
ALS Convolutive NMF with Sparse Constraint
Multilinear Algebra and ALS Nonnegative Tucker Decomposition
Algorithm for K-CNTD
Simulation
Synthetic Data
Speaker Recognition in Noisy Condition
Conclusion
References
Making Image to Class Distance Comparable
Introduction
Review of Image to Class Distance Based on Naive Bayes Assumption
Probabilistic Random Sampling Image to Class Distance (PRSI2CD)
Weighted PRSI2CD
Experiments
Graz-01 Dataset
Caltech101 Dataset
Conclusions
References
Margin Preserving Projection for Image Set Based Face Recognition
Introduction
Preliminaries
Convex Model
SVM Approximation
Margin Preserving Projection
The Proposed Algorithm
Intuition of MPP
Experiments
Experimental Results and Discussions
Conclusions
References
An Incremental Class Boundary Preserving Hypersphere Classifier
Introduction
Incremental Hypersphere Classifier Algorithm
Experimental Results
Conclusions and Future Work
References
Co-clustering for Binary Data with Maximum Modularity
Introduction
Generalized Modularity Measure
Modularity and Graphs
Modularity Measure for Binary Data
Maximization of Normalized Generalized Modularity
Spectral Connection
Spectral Co-clustering Algorithm
Numerical Experiments
Evaluation of SpecCo
Real Data Sets
Conclusion
References
Co-clustering under Nonnegative Matrix Tri-Factorization
Introduction
Double kmeans
NMF Framework for Co-clustering
DNMF Formulation
ODNMF Formulation
Numerical Experiments
Synthetic Data
Real Datasets
Conclusion
References
SPAN: A Neuron for Precise-Time Spike Pattern Association
Introduction
Learning Method
Neural and Synaptic Model
Learning
Experiments
Precise-Time Spike Pattern Association
Memory Capacity
Conclusion and Future Directions
References
Induction of the Common-Sense Hierarchies in Lexical Data
Introduction
The Data
PCA Directions
Creating Hierarchical Partitioning
Hierarchical Agglomerative Partitioning
Hierarchical Partitioning with Principal Components
Discussion and Future Directions
References
A Novel Synthetic Minority Oversampling Technique for Imbalanced Data Set Learning
Introduction
Motivation
Proposed CBSO Algorithm
Synthetic Data Generation Mechanism of CBSO
Clustering Minority Class
Experimental Study
Conclusion
References
A New Simultaneous Two-Levels Coclustering Algorithm for Behavioural Data-Mining
Introduction
Self-Organizing Map Adaptation for Disjunctive Data
A New Two-Levels Coclustering Algorithm: S2L-KDisj
Principle of the Algorithm
Algorithm
Application
Conclusion
References
An Evolutionary Fuzzy Clustering with Minkowski Distances
Introduction
Evolutionary Fuzzy Clustering with Minkowski Distances
Membership Function
Chromosome Representation
Population Initialization
Computation of Fitness Function
Selection
Experimental Results and Analysis
Iris Data Problem
Wine Data Problem
SPECTF Heart Problem
Conclusions
References
A Dynamic Unsupervised Laterally Connected Neural Network Architecture for Integrative Pattern Discovery
Introduction
Method
Model Overview and Architecture
Algorithm
Experimental Design and Results
Discussion and Conclusion
References
Author Index

Content (PDF)

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

Neural Information Processing

Description

More details

Other editions

Additional editions

Content

System requirements