Support vector machines : data analysis, machine learning and applications / Brandon H. Boyle, editor.

Format:

Book

Contributor:

Boyle, Brandon H.

Series:

Computer science, technology and applications.

Computer science, technology, and applications

Language:

English

Subjects (All):

Support vector machines.

Physical Description:

1 online resource (214 p.)

Edition:

1st ed.

Place of Publication:

New York : Nova Science Publishers, c2011.

Language Note:

English

Summary:

This book presents topical research in the study of support vector machines. Topics discussed include the support vector machine in medical imaging; monthly air pollution modeling using support vector machine techniques in Spain; support vector machines for image interpolation schemes in image zooming and color array interpolation; using SVM for the prediction of the ultimate capacity of driven piles in cohesionless soils; SVM in medical classification tasks and pattern recognition for machine fault diagnosis using support vector machines.

Contents:

Intro

SUPPORT VECTOR MACHINES: DATA ANALYSIS, MACHINE LEARNING AND APPLICATIONS

CONTENTS

PREFACE

THE SUPPORT VECTOR MACHINE IN MEDICAL IMAGING

ABSTRACT

1. INTRODUCTION

2. THE SUPPORT VECTOR MACHINE

3. THE SUPPORT VECTOR MACHINE'S USE IN MEDICAL IMAGING

3.1. Breast Cancer Imaging

3.2. Brain Imaging

3.3. Skin and Oral Imaging

3.4. Liver Imaging

3.5. Lung Imaging

3.6. Reproductive System Imaging

3.7. Eye Imaging

3.8. Other Imaging Applications

4. CASE STUDY: THE SUPPORT VECTOR MACHINE IN BREAST CANCER DETECTION FROM MAGNETIC RESONANCE IMAGING

4.1. Case Study - Introduction

4.2. Case Study - Methods

Support Vector Machine Classification

Proposed Vector Machine Formulations

Breast MRI Database for Case Study

Image Acquisition and Data Preprocessing

Breast MR Lesion Measurements

Feature Measurement #1: Average Slope

Feature Measurement #2: Average Washout

Feature Measurement #3: Sphericity / Irregularity

Feature Measurement #4: Average Edge Diffuseness

Receiver Operating Characteristic Curve Analysis and Validation

4.3. Case Study - Results

4.4. Case Study - Discussion

4.5. Case Study - Conclusions

CONCLUSION

ACKNOWLEDGMENTS

REFERENCES

A SVM-BASED REGRESSION MODEL TO STUDY THE AIR QUALITY IN THE URBAN AREA OF THE CITY OF OVIEDO (SPAIN)

2. SOURCES AND TYPES OF AIR POLLUTION

2.1. Primary Pollutants

2.2. Secondary Pollutants

2.3. Trends in Air Quality

3. MATHEMATICAL MODEL

3.1. Non-Linear Support Vector Machines

4. EXPERIMENTAL DATA SET

5. METHODOLOGY

6. RESULTS AND DISCUSSION

IMAGE INTERPOLATION USING SUPPORT VECTOR MACHINES

ABSTRACT.

1. INTRODUCTION OF IMAGE INTERPOLATION

1.1. Linear and Cubic Image Interpolation

1.2. Support Vector Regression

2. SUPPORT VECTOR MACHINES BASED IMAGE INTERPOLATION

2.1. Data Fitting Image Interpolation Approach

2.2. Neighbor Pixel Image Interpolation Approach

2.3. Local Spatial Properties Image Interpolation Approach

2.4. Conclusion

3. SUPPORT VECTOR MACHINES BASED INTERPOLATION FOR COLOR FILTER ARRAY

3.1. Introduction to Color Filter Array Interpolation

3.2 Color Filter Array Interpolation Using SVR

3.3. Experiments

ACKNOWLEDGMENT

UTILIZATION OF SUPPORT VECTOR MACHINE (SVM) FOR PREDICTION OF ULTIMATE CAPACITY OF DRIVEN PILES IN COHESIONLESS SOILS

ABSTRACT:

INTRODUCTION

DETAILS OF SVM MODEL

RESULTS AND DISCUSSION

SUPPORT VECTOR MACHINES IN MEDICAL CLASSIFICATION TASKS

1.Introduction

2.SupportVectorMachines

3.Experimentation

3.1.BreastCancerDatabase

3.2.ParkinsonDatabase

3.3.UrologicalDatabase

3.3.1.DimensionalityReduction

3.3.2.ArchitectureoftheSVM

4.Conclusions

Acknowledgment

References

KERNEL LATENT SEMANTIC ANALYSIS USING TERM FUSION KERNELS

Abstract

2.KernelCombinationforTextMiningTasks

3.Application:LatentSemanticClassExtractioninTextMining

3.1.Assigningprobabilitiesoftermstosemanticclasses

4.Experimentalwork

5.Conclusions

Acknowledgments

SVR FOR TIME SERIES PREDICTION

2. RELATED WORK

3. PREDICTION MODELS

3.1 Artificial Neural Networks

3.2 Support Vector Machines

3.3 Support Vector Predictors (SVP)

4. EXPERIMENTS

5. CONCLUSION

APPLICATION OF NEURAL NETWORKS AND SUPPORT VECTOR MACHINES IN CODING THEORY AND PRACTICE

2. RECURRENT NEURAL NETWORK DECODING.

2.1. Theoretical Model of the Encoder

2.2. Theoretical Model of the Decoder

2.3. Application of the Theoretical Model for One and Two-Input Encoders

2.3.1. One Input Encoder

2.3.2. Two Input Encoder

3. Support Vector Machine Decoding

3.1.1. SVM Decoder Analysis

3.1.2. The Training Stage

3.1.3. The Decoding Stage

3.2. Advantages of SVM Decoder

3.3. Complexity of SVM Decoder

3.4. SVM Decoder Design

3.5. Simulation Results

3.5.1 Effect of Training Size on SVM Decoder

3.5.2. Effect of Rayleigh's fading

CONCLUSIONS

PATTERN RECOGNITION FOR MACHINE FAULT DIAGNOSIS USING SUPPORT VECTOR MACHINE

2. PRELIMINARY KNOWLEDGE

2.1. Fault Diagnosis

2.2. Time Domain Analysis

2.3. Frequency Domain Analysis

3. FEATURE-BASED DIAGNOSIS SYSTEM

3.1. Data Preprocessing

3.1.1. Wavelet Transform

3.1.2. Averaging

3.1.3. Enveloping

3.1.4. Cepstrum

3.2. Statistical Feature Representation

3.2.1. Features in Time Domain

3.2.2. Features on Frequency Domain

3.2.3. Auto-regression Coefficient

3.3. Dimensionality Reduction Using Feature Extraction

3.3.1. Principal Component Analysis (PCA)

3.3.2. Independent Component Analysis (ICA)

3.3.3. Kernel PCA

3.3.4. Kernel ICA

4. SUPPORT VECTOR MACHINE (SVM)

4.1. Basic Theory: Binary Classification by SVM

4.2. SVM Solver

4.2.1. Quadratic Programming (QP)

4.2.2. Sequential Minimum Optimization (SMO)

4.3. Multi-class Classification

4.3.1. One-Against-All (OAA)

4.3.2. One-Against-One (OAO)

4.3.3. Direct Acyclic Graph (DAG)

4.4. Wavelet-Support Vector Machine (W-SVM)

5. APPLICATION FOR FAULT DIAGNOSIS OF INDUCTION MOTOR

5.1. Fault Diagnosis Method

5.2. Experiment and Data Acquisition

5.3. Feature Extraction and Reduction

5.4. Classification.

5.5. Results and Discussion

INDEX.

Notes:

Description based upon print version of record.

Includes bibliographical references and index.

Description based on print version record.

ISBN:

1-62257-078-2

OCLC:

839303231