Qian / Cao / Su Recent Advances in Computer Science and Information Engineering

Volume 1
1. Auflage 2012
ISBN: 978-3-642-25781-0
Verlag: Springer
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, Band 124, 786 Seiten

Reihe: Lecture Notes in Electrical Engineering

CSIE 2011 is an international scientific Congress for distinguished scholars engaged in scientific, engineering and technological research, dedicated to build a platform for exploring and discussing the future of Computer Science and Information Engineering with existing and potential application scenarios. The congress has been held twice, in Los Angeles, USA for the first and in Changchun, China for the second time, each of which attracted a large number of researchers from all over the world. The congress turns out to develop a spirit of cooperation that leads to new friendship for addressing a wide variety of ongoing problems in this vibrant area of technology and fostering more collaboration over the world. The congress, CSIE 2011, received 2483 full paper and abstract submissions from 27 countries and regions over the world. Through a rigorous peer review process, all submissions were refereed based on their quality of content, level of innovation, significance, originality and legibility. 688 papers have been accepted for the international congress proceedings ultimately.  

Autoren/Hrsg.

Qian, Zhihong

Cao, Lei

Su, Weilian

Wang, Tingkai

Yang, Huamin

Inhaltsverzeichnis

1;Title;1
2;Preface;4
3;Contents;18
4;Data Mining and Data Engineering Track;18
4.1;Statistical Foundations for Data Mining;18
4.1.1;A Novel RS-Based Key Frame Extraction Algorithmfor Video Mining in Compressed-Domain;28
4.1.1.1;Introduction;28
4.1.1.2;Relative Rough Sets Theory in This Paper;29
4.1.1.2.1;Concepts of Rough Sets in Data Analysis;29
4.1.1.2.2;Attributes Reduction in Rough Sets;30
4.1.1.3;The Proposed Algorithm in Compressed Domain;30
4.1.1.3.1;Extract DCT Coefficients;30
4.1.1.3.2;Extract DC Coefficients;31
4.1.1.3.3;Construct Information System;31
4.1.1.3.4;Reduce Information System;31
4.1.1.4;Results and Discussions;32
4.1.1.5;Conclusions;33
4.1.1.6;References;33
4.1.2;A Study on Small Scale Fading Model of AeronauticalRelay Channel;34
4.1.2.1;Introduction;34
4.1.2.2;Physical Model of Aeronautical Relay Channel;34
4.1.2.3;Reference Derivation of Aeronautical Relay Channel Model;35
4.1.2.3.1;Analysis of T2R Channel Model;35
4.1.2.3.2;Analysis of R2R Channel Model;35
4.1.2.3.3;Analysis of Overall Channel Model;36
4.1.2.4;Statistical Characteristics Analysis and Simulationof Aeronautical Relay Channel;38
4.1.2.5;Conclusions;39
4.1.2.6;References;39
4.1.3;An Investigation of the Performanceof Informative Samples Preservation Methods;40
4.1.3.1;Introduction;40
4.1.3.2;Algorithms;41
4.1.3.3;Experimental Results;44
4.1.3.4;Conclusion;45
4.1.3.5;References;45
4.1.4;Analysis of Spatial Intensity Correlation’s Effecton Registration Measures;46
4.1.4.1;Introduction;46
4.1.4.2;Registration Measures;46
4.1.4.3;Analysis of Registration Measures Curve;49
4.1.4.4;Conclusion;52
4.1.4.5;References;53
4.1.5;Application of Data Mining in Analysis of Electrovalence;54
4.1.5.1;Introduction;54
4.1.5.2;Fundamental;54
4.1.5.3;Objective and Methods of Data Mining;55
4.1.5.3.1;Association;55
4.1.5.3.2;Classification and Prediction;56
4.1.5.3.3;Clustering;56
4.1.5.4;Application of Data Mining;56
4.1.5.4.1;Business Understanding;57
4.1.5.4.2;Data Understanding;57
4.1.5.4.3;Data Preparation;57
4.1.5.4.4;Modeling;58
4.1.5.4.5;Evaluation;61
4.1.5.4.6;Deployment;61
4.1.5.5;Conclusion;62
4.1.5.6;References;62
4.1.6;Application of Data Mining in Power Customer Satisfaction Evaluation;63
4.1.6.1;Introduction;63
4.1.6.2;Basic Principle;64
4.1.6.2.1;Clustering;64
4.1.6.2.2;Principal Component Analysis;64
4.1.6.3;Application of Data Mining;65
4.1.6.3.1;Business Understanding;65
4.1.6.3.2;Data Understanding;65
4.1.6.3.3;Data Preparation;66
4.1.6.3.4;Modeling;66
4.1.6.3.5;Model Evaluation;68
4.1.6.3.6;Model Release;69
4.1.6.4;Application Results;69
4.1.6.5;Conclusion;70
4.1.6.6;References;70
4.1.7;CRM Research Based on the Decision Tree Classification Algorithm;71
4.1.7.1;Introduction;71
4.1.7.2;ID3Algorithm Principle;72
4.1.7.3;ID3 Algorithm Improvement;72
4.1.7.4;Design and Realizatino of ID3 Improving Algorithm;74
4.1.7.5;Algorithm Application;74
4.1.7.6;Experimental Result and Comparison;75
4.1.7.7;Conclusion;76
4.1.7.8;References;76
4.1.8;Design of Optimal Controller for Dynamic Positioning System of a Certain Drilling Platform;77
4.1.8.1;Introduction;77
4.1.8.2;The Description of DP System;78
4.1.8.3;Mathematical Modelling;78
4.1.8.4;Design of a Variable Structure Controller;79
4.1.8.5;Simulation Results;81
4.1.8.6;Conclusion;82
4.1.8.7;References;82
4.1.9;KKM Theorems in L-Spaces;83
4.1.9.1;Introduction;83
4.1.9.2;Preliminaries;83
4.1.9.3;KKM Theorems for Noncompact L-Spaces;85
4.1.9.4;References;88
4.1.10;Optimal Output of Hydropower Station Based on Rough Set Theory;89
4.1.10.1;Introduction;89
4.1.10.2;Rough Set Theory and Method;89
4.1.10.2.1;Information System;90
4.1.10.2.2;Lower and Upper Approximations;90
4.1.10.2.3;Knowledge Acquisition;90
4.1.10.2.4;Variable Precision Rough Set(VPRS);91
4.1.10.3;Application Example;91
4.1.10.3.1;Data Preparation;91
4.1.10.3.2;Data Discretization;92
4.1.10.3.3;Attribute Reduction;92
4.1.10.3.4;Role Generation and Filtration;92
4.1.10.3.5;Optimal Output Rules Analysis;93
4.1.10.4;Conclusion;93
4.1.10.5;References;93
4.1.11;Performance Study on Fire Alarm System in Large Space Buildings;94
4.1.11.1;Introduction;94
4.1.11.2;Model Design;95
4.1.11.3;Air Sampling Pipe Network;96
4.1.11.4;Results;97
4.1.11.5;Conclusions;98
4.1.11.6;References;98
4.1.12;Research on Expert System for Ginseng Diseases and Pests Based on CBR;99
4.1.12.1;Introduction;99
4.1.12.2;The Implement of the Expert System;100
4.1.12.2.1;The Abstraction of Properties, Description of Cases, and the Establishment of Cases Database;100
4.1.12.2.2;The Set-Up of Database of Ginseng Diseases and Matching Strategy of the Cases;102
4.1.12.2.3;Input of New Case and the Process of Case Retrieval;103
4.1.12.3;Conclusions Revise and Machine Learning;103
4.1.12.4;System Conclusions and Future Work;104
4.1.12.5;References;104
4.1.13;Research on Reliability Evaluation Based on Vibration Severity for Gearbox;105
4.1.13.1;Introduction;105
4.1.13.2;Concepts of Vibration Severity;106
4.1.13.2.1;Definition of Vibration Severity;106
4.1.13.2.2;Classification of Vibration Severity;106
4.1.13.3;Index of Vibrating Signal and Measuring Method;107
4.1.13.3.1;Vibrating Measure;107
4.1.13.3.2;Evaluation Standard of Vibrating;107
4.1.13.4;Research on Vibration Severity;107
4.1.13.5;Reliability Estimation on Gearbox;109
4.1.13.6;Conclusions;110
4.1.13.7;References;110
4.1.14;Study and Development on WEBGIS-Based Emergency Rescue Platform for Mines;111
4.1.14.1;The Goal of the System and Demand Analysis;111
4.1.14.2;The Design of the System Principle and General Structure;112
4.1.14.3;The Structure of Emergency Rescue System;112
4.1.14.4;Conclusion and Prospect;116
4.1.14.5;References;116
4.1.15;Study on Noun Phrase of “B+N” Structure in Search Engine Query Logs;117
4.1.15.1;Introduction;117
4.1.15.2;The Characteristics of Various Elements of "B + N" Structure Noun Phrase in Search Engine Logs;117
4.1.15.2.1;Monosyllable-Morpheme Distinguishing Words;118
4.1.15.2.2;Multi-Morpheme Distinguishing Words;118
4.1.15.3;Usage of "B + N" structure Noun Phrase in Search Engine Logs;120
4.1.15.4;Experimental Results and Analysis;121
4.1.15.5;References;122
4.1.16;Study on Queue Strategy of Gated Polling Multi-access Communication System;123
4.1.16.1;Introduction;123
4.1.16.2;Mathematical Model;123
4.1.16.2.1;System Working Conditions;124
4.1.16.2.2;Generation Function;124
4.1.16.3;Performance Analysis;125
4.1.16.3.1;Mean Queue Length;125
4.1.16.3.2;g(1) (i, j);125
4.1.16.3.3;Mean Waiting Time;127
4.1.16.4;Simulation;127
4.1.16.5;Conclusions;129
4.1.16.6;References;129
4.1.17;The Climatic Characterization of Reference Evapotranspiration of Beijing Meteorological Station;130
4.1.17.1;Introduction;130
4.1.17.2;Methodology;131
4.1.17.2.1;Penman-Monteith Method;131
4.1.17.2.2;Sensitivity Coefficients;131
4.1.17.2.3;The Detrend Method;132
4.1.17.2.4;The Contribution Method;132
4.1.17.3;Study Area and Data;132
4.1.17.4;Results and Discussion;133
4.1.17.4.1;Annual Variation of ETp and Key Climatic Variables;133
4.1.17.4.2;Sensitivity of Key Climatic Variables;133
4.1.17.4.3;Estimate the Influence of Key Climatic Variables on ETp;134
4.1.17.4.4;Intra-annual Variation of Key Climatic Variables;134
4.1.17.4.5;Contribution of Key Climatic Variables Fluctuation to the Intra-annual Variation of Evapotranspiration;135
4.1.17.5;Summary and Conclusions;136
4.1.17.6;References;137
4.1.18;The Digital Design Applied to Consumer Garment Try-On Experience Integrated with Augmented Reality;138
4.1.18.1;Background;138
4.1.18.2;GUI AR Pattern;139
4.1.18.3;Software Construction;140
4.1.18.4;Operates the Procedure;141
4.1.18.5;Conclusion;143
4.1.18.6;References;144
4.1.19;Video Event Detection Based on Temporal Pattern Analysis;146
4.1.19.1;Introduction;146
4.1.19.2;Proposed Framework;147
4.1.19.3;Experiments;150
4.1.19.4;Conclusions;151
4.1.19.5;References;151
4.2;Intelligent Techniques for Data Mining;19
4.2.1;A Memetic Algorithm for Hardware Software Partitioning;152
4.2.1.1;Introduction;152
4.2.1.2;Problem Description;153
4.2.1.3;Memetic Algorithm;154
4.2.1.3.1;Local Search Procedure;154
4.2.1.3.2;Crossover Procedure;155
4.2.1.3.3;The Proposed Algorithm;155
4.2.1.4;Experiment;155
4.2.1.5;Conclusion;156
4.2.1.6;References;156
4.2.2;Dimensionality Reduction for Colon Data;158
4.2.2.1;Introduction;158
4.2.2.2;Dimensionality Reduction Method;158
4.2.2.2.1;Laplacian Eigenmaps;159
4.2.2.2.2;MDS;159
4.2.2.3;Support Vector Machine;160
4.2.2.4;Analysis of Experimental Results;161
4.2.2.5;Conclusion;162
4.2.2.6;References;162
4.2.3;Mechanics Analysis on Set Membership Evaluation of Real-Virtual Sensor-Based on Multi-layer Fusion;163
4.2.3.1;Introduction;163
4.2.3.2;The Principle of Set Membership Evaluation of Real-Virtual Sensor Based on Multi-layer Fusion;164
4.2.3.2.1;The Principle of Real-Virtual Sensor Based on Multi-layer Fusion;164
4.2.3.2.2;The Principle of Set Membership Evaluation of Real-Virtual Sensor Based on Multi-layer Fusion;164
4.2.3.3;Mechanism Analysis;165
4.2.3.3.1;Multi-sensor Adaptive Weighting Fusion Algorithm;165
4.2.3.3.2;The Algorithm of Real-Virtual Sensor Based on Multi-layer Fusion;166
4.2.3.3.3;The Estimation Algorithm of Set Membership of Real-Virtual Sensor Based on Multi-layer Fusion;166
4.2.3.4;Simulation and Analysis;168
4.2.3.4.1;Simulation;168
4.2.3.4.2;Performance Analysis;168
4.2.3.5;Conclusion;169
4.2.3.6;References;169
4.2.4;Web News Pages Extraction Method Based on DOM and Decision Tree;170
4.2.4.1;Introduction;170
4.2.4.2;News Pages Extraction;171
4.2.4.2.1;Page Preprocessing;171
4.2.4.2.2;Problem Definition and Framework of Decision Tree-Based Extraction Method;171
4.2.4.2.3;Attributes Extraction;171
4.2.4.2.4;Learning to Identify News Body Subtrees;172
4.2.4.2.5;Integrating the Results of Extraction;173
4.2.4.3;Experiment and Result;173
4.2.4.3.1;Learning and Building a Decision Tree;173
4.2.4.3.2;Accuracy Testing;174
4.2.4.4;Conclusion and Future Work;175
4.2.4.5;References;175
4.3;Feature Selection and Extraction;20
4.3.1;Study on Feature Selection Based on Fuzzy Clustering Algorithm;176
4.3.1.1;Introduction;176
4.3.1.2;Feature Separability Analysis Based on Fuzzy ISODATA;177
4.3.1.3;Candidate Feature Subset Generating in RFE;178
4.3.1.4;Feature Selection Experiment;179
4.3.1.5;Conclusion;181
4.3.1.6;References;182
4.3.2;Urban Air Quality Forecast System Based on Sample Optimization and Its Application;183
4.3.2.1;Sample Optimization;184
4.3.2.2;BP Neural Network;185
4.3.2.3;Establishment of Urban Air Quality Forecast System;186
4.3.2.3.1;Air Quality Prediction Model;186
4.3.2.3.2;Urban Air Quality Forecast System;187
4.3.2.4;Case Study;187
4.3.2.4.1;Site Description and Data;187
4.3.2.4.2;Results and Discussion;188
4.3.2.5;Conclusions;189
4.3.2.6;References;189
4.4;Clustering and Classification;20
4.4.1;A Fast KNN Categorization Algorithm Based on Feature Space Indexing;190
4.4.1.1;Introduction;190
4.4.1.2;KNN Algorithm and Its Improvements;191
4.4.1.3;A Fast KNN Categorization Algorithm Based on Feature Space Indexing;191
4.4.1.4;Experimental Result and Analysis;193
4.4.1.5;Conclusion;194
4.4.1.6;References;194
4.4.2;An Experimental Comparison of Different Features for Image Retrieval Based on Bayesian Classifier;196
4.4.2.1;Introduction;196
4.4.2.2;Features for CBIR;197
4.4.2.3;Bayesian Estimation for Relevance Feedback;198
4.4.2.4;Experimental Results and Analysis;200
4.4.2.5;Conclusion;202
4.4.2.6;References;202
4.4.3;An Improved Document Clustering Algorithm Based on Neural Gas Algorithm;203
4.4.3.1;Introduction;203
4.4.3.2;Our Algorithm;204
4.4.3.3;Experimental Results;206
4.4.3.3.1;Comparison of Quality of Clustering Algorithms;206
4.4.3.3.2;Comparison of Time Spent by Each Clustering Algorithm;207
4.4.3.4;Conclusion;208
4.4.3.5;References;208
4.4.4;An Odor Discrimination Approach Based on Mice Olfactory Neural Network;209
4.4.4.1;Introduction;209
4.4.4.2;Mice Olfactory Neural Network Model;210
4.4.4.3;Establishment of Artificial Olfactory Neural Network;211
4.4.4.4;Experiments and Result Analysis;212
4.4.4.5;Conclusions;213
4.4.4.6;References;213
4.4.5;Application Research of Weighted Fuzzy C Means Clustering about Soil Fertility Evaluation in Nong'an County;215
4.4.5.1;Introduction;215
4.4.5.2;Building FCM Model Based on Entropy Weighted;216
4.4.5.2.1;Determine the Weight Coefficient;216
4.4.5.2.2;Weighted Standardized Data;217
4.4.5.2.3;Clustering Process;217
4.4.5.3;Application of the Algorithm about Soil Fertility Evaluation in Nong'An County;218
4.4.5.3.1;Data Selection;218
4.4.5.3.2;Soil Fertility Evaluation;218
4.4.5.4;Conclusion and Discussion;219
4.4.5.5;References;220
4.4.6;Automatic Classification between Wind and Bowstring Instrumental Music Using Support Vector Machine;222
4.4.6.1;Introduction;222
4.4.6.2;Feature Extraction;223
4.4.6.2.1;Spectral Centroid;223
4.4.6.2.2;Spectral Spread;223
4.4.6.2.3;Low Energy Frame Ratio;223
4.4.6.2.4;MFCCs;224
4.4.6.3;Classification;224
4.4.6.3.1;Support Vector Machine;224
4.4.6.3.2;Parameter Optimization;225
4.4.6.4;Experimental Results;225
4.4.6.5;Conclusions;227
4.4.6.6;References;227
4.4.7;Boundary Point Recognition Algorithm Based on Grid Adjacency Relation;228
4.4.7.1;Introduction;228
4.4.7.2;Concepts;229
4.4.7.3;The Refinement of the Boundary;231
4.4.7.4;Description of the GAB algorithm;231
4.4.7.5;The Analysis of Algorithm Performance;232
4.4.7.6;Experiment Result and Discussion;232
4.4.7.7;Conclusion;234
4.4.7.8;References;235
4.4.8;The Application of Genetic BP Neural Network and D-S Evidence Theory in the Complex System Fault Diagnosis;236
4.4.8.1;Introduction;236
4.4.8.2;Genetic Algorithm Optimizing BP Neural Network;236
4.4.8.3;D-S Evidence Theory [16-24];237
4.4.8.4;The Diagnostic System’s Framework Combined Genetic Neural Network with D-S Evidence Theory;238
4.4.8.5;Example;239
4.4.8.6;Conclusion;241
4.4.8.7;References;241
4.4.9;Research into Fuzzy Clustering with Collaboration between Multi Data Sets;242
4.4.9.1;Introduction;242
4.4.9.2;FCM Clustering Algorithm;243
4.4.9.3;Collaboration Clustering;244
4.4.9.4;Collaboration Clustering Algorithm;245
4.4.9.4.1;Settlement of the Optimization Problem;246
4.4.9.4.2;Algorithm Flow;248
4.4.9.4.3;Evaluation of the Collaboration Effect;249
4.4.9.5;Experimental Analysis;249
4.4.9.5.1;Clustering Results without Collaboration;249
4.4.9.5.2;Optimization of Clustering with Collaboration;250
4.4.9.5.3;Collaboration Effect;250
4.4.9.6;Conclusion;250
4.4.9.7;References;251
4.4.10;Scalable Online Incremental Learning for Web Spam Detection;252
4.4.10.1;Introduction ;252
4.4.10.2;The Proposed Online Incremental Learning Framework ;253
4.4.10.2.1;Framework Overview ;253
4.4.10.2.2;Methodology;254
4.4.10.3;Experimental Evaluation;256
4.4.10.4;Conclusion and Future Work;257
4.4.10.5;References;257
4.4.11;Semi-supervised Clustering Algorithm Based on Shared Nearest Neighbors;259
4.4.11.1;Introduction;259
4.4.11.2;Problems;259
4.4.11.3;Algorithm Descriptions;260
4.4.11.4;Simulations;262
4.4.11.5;Conclusions;263
4.4.11.6;References;264
4.5;Association Rule Mining;21
4.5.1;A Improved MASK Algorithm Based on Privacy Preserving;265
4.5.1.1;Introduction;265
4.5.1.2;Concepts and Algorithms;266
4.5.1.2.1;Association Rules;266
4.5.1.2.2;MASK Algorithm Description;266
4.5.1.3;Improvement on MASK Algorithm;267
4.5.1.3.1;Principle Description;268
4.5.1.3.2;Improved Algorithm Description;268
4.5.1.4;Experiment and Results Analysis;269
4.5.1.5;Conclusions;270
4.5.1.6;References;270
4.5.2;Classification Data Mining for Digital Home Sensor Networks;271
4.5.2.1;Introduction;271
4.5.2.2;Digital Home Sensor Networks;271
4.5.2.2.1;Digital Home;271
4.5.2.2.2;Sensor Network;272
4.5.2.3;Data Mining for Digital Home Sensor Networks;272
4.5.2.3.1;Model of Classification Mining;272
4.5.2.3.2;Algorithms of Classification Mining;273
4.5.2.3.3;Steps of Classification Mining;273
4.5.2.4;Experiment;274
4.5.2.4.1;Pretreatment of Collected Data;274
4.5.2.4.2;Data Sort;274
4.5.2.4.3;Process of Classification Mining;275
4.5.2.5;Conclusion;276
4.5.2.6;References;276
4.6;Data Visualization and Exploration;21
4.6.1;On Ranking Senators by Their Votes;277
4.6.1.1;Introduction;277
4.6.1.2;Background;277
4.6.1.2.1;Ranking Legislators;277
4.6.1.2.2;The Combinatorial Graph Laplacian;278
4.6.1.3;Ranking;278
4.6.1.4;Algorithm;280
4.6.1.4.1;Computing Similarities;280
4.6.1.4.2;Selecting the Labeled fi;281
4.6.1.5;Experimental Results;281
4.6.1.5.1;Using Domain Knowledge;281
4.6.1.5.2;Using Internal Knowledge;281
4.6.1.6;Discussion;284
4.6.1.6.1;Conclusions;284
4.6.1.6.2;Future Work;284
4.6.1.7;References;284
4.6.2;Visibility Graph Analysis on Monthly Macroeconomic Series of U.S.A. from Jan-1959 to Jun-2010 Based on Complex Network Theory;285
4.6.2.1;Introduction;285
4.6.2.2;Methods and Materials;286
4.6.2.3;Results and Discussions;287
4.6.2.3.1;Degree Distribution of Associated Networks;287
4.6.2.3.2;Community Structure of Associated Networks;287
4.6.2.4;Conclusions and Application;290
4.6.2.5;References;290
4.7;Event Detection and Trend Tracking;21
4.7.1;Investigations of Intrusion Detection Based on Data Mining;291
4.7.1.1;Introduction;291
4.7.1.2;Intrusion Detection Techniques and Data Mining Techniques;292
4.7.1.2.1;Intrusion Detection Techniques and Data Mining Techniques;292
4.7.1.2.2;Apriori Algorithm and Clustering Algorithm;292
4.7.1.3;Experimental Test and Analysis;294
4.7.1.4;Conclusions;295
4.7.1.5;References;295
4.8;Time-Series and Sequential Data Mining;21
4.8.1;Research on the Change of HRV in the Process of Gradually Guided Breathing;296
4.8.1.1;Introduction;296
4.8.1.2;Materials and Methods;297
4.8.1.2.1;Experiment Scheme;297
4.8.1.2.2;Data Acquisition;297
4.8.1.2.3;Data Analysis;297
4.8.1.3;Results;299
4.8.1.3.1;The Change of Power Spectrum in Guided Breathing Process;299
4.8.1.3.2;The Results of HRV Parameters in Guided Breathing Process;299
4.8.1.3.3;Statistical Result;300
4.8.1.4;Discussion;300
4.8.1.5;Conclusion;301
4.8.1.6;References;301
4.8.2;The Analysis of Earthquake Precursory Based on Multiscale Technology of Wavelet Transform;302
4.8.2.1;Introduction;302
4.8.2.2;Basic Principles of Wavelet Transform;303
4.8.2.3;Seismic Data Processing Method Based on Frequency Analysis;303
4.8.2.4;Discrimination and Separation between High and Low Frequency Information for Digital Data of Precursors;304
4.8.2.5;Discrimination between Trend and Short-Term Anomalies of Digital Precursory Data;305
4.8.2.6;Conclusions;306
4.8.2.7;References;307
4.9;Image and Multi-media DataMining;21
4.9.1;Analysis Method Based of Digital Image of Wool or Cashmere;308
4.9.1.1;The Principle of Watershed Algorithm;308
4.9.1.2;Mark Watershed Algorithm Used in Cashmere Image Segmentation;309
4.9.1.2.1;Filtering and Enhancement of Cashmere Image;309
4.9.1.2.2;Obtain Morphological Gradient Image;309
4.9.1.2.3;Mark and Modify the Local Maximum Image;310
4.9.1.2.4;Modify the Gradient Image;310
4.9.1.2.5;Do Watershed Algorithm on Modified Gradient Image;311
4.9.1.3;Character Parameter Extract;311
4.9.1.3.1;Using Sub-measurement to Measure the Diameter;311
4.9.1.3.2;Precision Analysis of Measurement Results;312
4.9.1.4;Conclusion;312
4.9.1.5;References;312
4.9.2;Population Density Analysis in Underground Public Space Based on Digital Video and Pattern Recognition;314
4.9.2.1;Introduction;314
4.9.2.2;Design and Analysis for Digital Video Recording System;315
4.9.2.3;Recognition and Classification Mechanism of Population Images;316
4.9.2.3.1;Object Detection with Matlab Toolkit;316
4.9.2.3.2;Matrix Expression for Pre-detection Results;318
4.9.2.3.3;Classification Mechanism Based on Potential Function;318
4.9.2.4;Population Density Analysis in Underground Supermarket;319
4.9.2.4.1;Approach to Population Density Analysis;319
4.9.2.4.2;Algorithm of Population Density Calculation;319
4.9.2.5;Conclusion;319
4.9.2.6;References;319
4.10;Text andWeb Mining;22
4.10.1;A New Preprocessing Phase for LSA-Based Turkish Text Summarization;320
4.10.1.1;Introduction;320
4.10.1.2;Applying LSA;321
4.10.1.2.1;Data Representation;321
4.10.1.2.2;Algortihm I;321
4.10.1.2.3;Algorithm II;322
4.10.1.3;Performance Evaluation Process;322
4.10.1.3.1;Data Corpus and Evaluation Sets;322
4.10.1.3.2;Preprocessing Methods;323
4.10.1.3.3;Test Datasets;323
4.10.1.4;Experimental Results;323
4.10.1.5;Conclusion and Future Works;325
4.10.1.6;References;325
4.10.2;Recognition of Chemical Names in Chinese Texts;326
4.10.2.1;Introduction;326
4.10.2.2;Methods;327
4.10.2.2.1;Overview;327
4.10.2.2.2;Corpus Generation and Word Segmentation;327
4.10.2.2.3;Chinese NER as a Sequence Tagging Task with CRF;328
4.10.2.3;Experiments;330
4.10.2.3.1;Variance under Different Tagging Units;330
4.10.2.3.2;Comparisons of Different Feature Selection;331
4.10.2.3.3;Determine Effective Interval of Feature Values;331
4.10.2.4;Conclusions;332
4.10.2.5;References;332
4.11;Data Mining in Financial Engineering;22
4.11.1;Exploration on Data Recommendation Prototype Based on Data-Sharing-Platform;334
4.11.1.1;Introduction;334
4.11.1.2;System Design;334
4.11.1.2.1;Over all System Framework;334
4.11.1.2.2;Design of User Database;335
4.11.1.2.3;System Work Flow;336
4.11.1.3;System Implementation Technology;337
4.11.1.3.1;System Architecture Technology;337
4.11.1.4;System Application;337
4.11.1.5;Summary and Future Work Prospect;338
4.11.1.6;References;338
4.11.2;The Study of Financial Early-Warning Model in Chinese Listed Companies by Fuzzy Neural Network;340
4.11.2.1;Introduction;340
4.11.2.2;Based on Fuzzy Neural Network Model for Financial Risk Warning;340
4.11.2.2.1;Fuzzy Comprehensive Evaluation Model;340
4.11.2.2.2;Based on BP Neural Network to Build the Financial Early-Warning Model;342
4.11.2.3;Case Analysis and Network Testing;344
4.11.2.3.1;Case Selection;344
4.11.2.3.2;Financial Forecast Results Verify the Analysis;344
4.11.2.4;Paper Preparation;344
4.11.2.5;References;345
4.12;Information Fusion;22
4.12.1;A Multi-source Spatial Data Fusion Method Used for Terrain Simulation;347
4.12.1.1;Introduction;347
4.12.1.2;Multi-source DEM Data Fusion;347
4.12.1.2.1;Representations of DEM;347
4.12.1.2.2;Pretreatment of DEM Data;348
4.12.1.2.3;Fusion Strategy of DEM Data;348
4.12.1.2.4;Treatment of Special Points;350
4.12.1.3;Fusion Results Analysis;351
4.12.1.4;Conclusions and Future Work;352
4.12.1.5;References;352
4.13;Other Topics in Data Mining;22
4.13.1;A New Method of Outlier Detection;353
4.13.1.1;Introduction;353
4.13.1.2;The Method of Outlier Detection Based on Data Whole Difference Degree;353
4.13.1.3;Experiment and Simulation;356
4.13.1.4;Conclusion;358
4.13.1.5;References;358
4.13.2;Acoustic Recognition of Artillery Projectiles by SVM;359
4.13.2.1;Trajectory Shock Wave Signal;359
4.13.2.2;Support Vector Classification Machine;360
4.13.2.2.1;Modelling;360
4.13.2.2.2;Algorithm;361
4.13.2.3;Shock Wave Signal Based Support Vector Target Classification;363
4.13.2.3.1;Extraction of Target Characteristic Variables;363
4.13.2.3.2;Training and Testing;364
4.13.2.4;Conclusion;364
4.13.2.5;References;364
4.13.3;Algorithm for Data Mining Based on Fuzzy Logic;366
4.13.3.1;Introduction;366
4.13.3.2;Fuzzy Relation on Data System;366
4.13.3.3;Effect Characteristic of Intersection Values u;367
4.13.3.4;Data Model of ( X,U,P) and Characteristic;368
4.13.3.5;Studying of Decision Effect of M;369
4.13.3.6;Summary;370
4.13.3.7;References;370
4.13.4;Analysis on Electro-Mechanical Response of Carbon Fiber under Off-Axis Tension in Multivariate Regression Modeling Procedures;371
4.13.4.1;Introduction;371
4.13.4.2;Experimental and Methods;372
4.13.4.2.1;Experimental Design and Specimen Testing;372
4.13.4.2.2;Data Sets;372
4.13.4.3;Theoretical Analysis;374
4.13.4.3.1;Constructing Model;374
4.13.4.3.2;Prediction and Application;374
4.13.4.4;Conclusion;376
4.13.4.5;References;376
4.13.5;Application of Data Mining for Fluid Identification in Complicated Reservoir;377
4.13.5.1;Introduction;377
4.13.5.2;Data Mining;378
4.13.5.3;Practical Example for Data Mining;378
4.13.5.3.1;Date Set;379
4.13.5.3.2;Conventional Cross-Plot in Fluid Identification;379
4.13.5.3.3;The Model and Analysis;380
4.13.5.4;Conclusion;381
4.13.5.5;References;381
4.13.6;Audio Classification for Blackfoot Language Analysis;383
4.13.6.1;Introduction;383
4.13.6.2;The Proposed Framework;384
4.13.6.2.1;Audio Syntactic Analysis;384
4.13.6.2.2;Subspace-Based Modeling;384
4.13.6.2.3;SS Classification and Decision Fusion;386
4.13.6.3;Conclusions;388
4.13.6.4;References;388
4.13.7;Neural-Based Decision Trees Classification Techniques: A Case Study in Water Resources Management;389
4.13.7.1;Introduction;389
4.13.7.2;Neural Decision Tree;390
4.13.7.2.1;Artificial Neural Networks;390
4.13.7.2.2;Neural Decision-Tree Construction;390
4.13.7.3;Experiment;391
4.13.7.3.1;Study Site;391
4.13.7.3.2;Data and Modelling;391
4.13.7.3.3;Results;392
4.13.7.3.4;Simulation and Discussions;393
4.13.7.4;Conclusions;393
4.13.7.5;References;394
4.13.8;Personalized Recommendation Based on Desktop Context;395
4.13.8.1;Introduction;395
4.13.8.2;Extraction of Desktop Context;396
4.13.8.2.1;Desktop Resources Extraction;397
4.13.8.2.2;Work Scenario Context Extraction;398
4.13.8.3;Experiment of Personalized Recommendation;399
4.13.8.4;Conclusions;399
4.13.8.5;References;400
4.13.9;The Classification Recognition of Projectiles Wear Mark Based on Support Vector Machine Method;401
4.13.9.1;Introduction;401
4.13.9.2;The Basic Theory of Support Vector Machines;402
4.13.9.3;The Classification of Striation Wear Mark with SVM Method;403
4.13.9.4;Conclusion;405
4.13.9.5;References;406
4.14;Data Integration, Interoperability, and Metadata;23
4.14.1;An Integrated Spatial Topology Model of Urban Public Transport Networks Based on Walking Paths Oriented to Passenger’s Views;407
4.14.1.1;Introduction;407
4.14.1.2;Spatial Topology Modeling of Urban Transport Networks;408
4.14.1.2.1;Urban Road Network Spatial Topology Modeling;408
4.14.1.2.2;Public Transport Network Spatial Topology Modeling;409
4.14.1.3;Integrated Spatial Topology Model of Urban Transport Networks Based on Walking Paths;410
4.14.1.3.1;Waking Path Network Spatial Topology Modeling;410
4.14.1.3.2;Integrated Multiple Transport Networks Based on Walking Paths;410
4.14.1.4;Conclusion;411
4.14.1.5;References;412
4.15;Data Structures and DataManagement Algorithms;23
4.15.1;The Design of ATmega161 and USB Connection Data Acquisition System;413
4.15.1.1;Introduction;413
4.15.1.2;The Data Acquisitions System Outlines;414
4.15.1.2.1;Data Acquisition System's Composition;414
4.15.1.2.2;USB Introduces;414
4.15.1.3;The System Hardware Designs;414
4.15.1.3.1;Based on USB Data Acquisition System Structure Drawing;414
4.15.1.3.2;Chipset Options;415
4.15.1.3.3;Hardware Connection Diagram;415
4.15.1.4;System Software Design;416
4.15.1.4.1;The Composition of USB Device;416
4.15.1.4.2;Initialize USB;416
4.15.1.5;The Conclusion;417
4.15.1.6;References;417
4.16;Data Privacy and Security;23
4.16.1;Order Exchange Key in Data Encryption;418
4.16.1.1;Introduction;418
4.16.1.2;Fundamentals;419
4.16.1.3;Detailed Application Method and Particular Techniques;422
4.16.1.4;Conclusion and Further Research;423
4.16.1.5;References;423
4.16.2;Weighted Social Networks Anonymizing Publication;424
4.16.2.1;Introduction;424
4.16.2.2;Weighted Social Network Privacy Model;425
4.16.2.2.1;Weighted Social Networks Graphs;425
4.16.2.2.2;Bipartite Graphs Model;425
4.16.2.3;Anonymizing Weighted Social Networks;427
4.16.2.3.1;K-Automorphism Publication;427
4.16.2.3.2;KAP Algorithm;427
4.16.2.3.3;Information Loss;429
4.16.2.3.4;Experimental Results;430
4.16.2.4;Conclusion;431
4.16.2.5;References;431
4.17;XML Data Processing and Algorithms;23
4.17.1;Efficient Structural XML Index for Multiple Queries;433
4.17.1.1;Introduction;433
4.17.1.2;Index for Multiple Queries;434
4.17.1.2.1;Architecture of Index System;434
4.17.1.2.2;Document Parser;434
4.17.1.2.3;Query Engine;434
4.17.1.3;Further Optimization;436
4.17.1.3.1;Supporting Complex Query;436
4.17.1.3.2;Optimization for Queries;437
4.17.1.3.3;Optimization for Indexing Documents;437
4.17.1.3.4;Evaluation for Optimized Index;438
4.17.1.4;Experiments;438
4.17.1.4.1;Experimental Parameters;439
4.17.1.4.2;Experimental Results;439
4.17.1.5;Conclusion;440
4.17.1.6;References;440
4.18;Distributed, Parallel, Peer-to-Peer Databases;23
4.18.1;An Efficient KNN Query Processing Approach on High-Dimensional Data Objects in P2P Systems;442
4.18.1.1;Introduction;442
4.18.1.2;Construction of SCBO Network;443
4.18.1.2.1;Constructing Strategy and Clustering Scheme;443
4.18.1.2.2;Clustering Encoding Strategy;444
4.18.1.2.3;Neighborhood Establishment among Clusters;444
4.18.1.3;KNN Query Processing Algorithm;445
4.18.1.4;Performance Evaluation;446
4.18.1.4.1;Experiment Setup;446
4.18.1.4.2;Experiment Results;447
4.18.1.5;Conclusion and Future Works;448
4.18.1.6;References;448
4.18.2;Research on Parallelizing Trie-Based Apriori Algorithm;450
4.18.2.1;Introduction;450
4.18.2.2;Trie-Based Apriori Algorithm;450
4.18.2.3;Parallelizing Programs;452
4.18.2.4;Experiments and Analysis;453
4.18.2.5;Conclusions;455
4.18.2.6;References;455
4.19;Web Search;23
4.19.1;A Web Log Mining System Based on MDP Algorithm;456
4.19.1.1;Introduction;456
4.19.1.2;MDP Algorithm;457
4.19.1.2.1;Basic Idea of MDP Algorithm;457
4.19.1.2.2;Design and Construction of MDP-Tree;457
4.19.1.3;Web Log Mining System Based on MDP Algorithm;458
4.19.1.3.1;Analysis of Data Source;458
4.19.1.3.2;Data Preprocessing;459
4.19.1.3.3;Mining Association Rules;459
4.19.1.4;Conclusion;461
4.19.1.5;References;461
4.20;Databases Applications;24
4.20.1;Design and Implementation of Test Paper Management System;463
4.20.1.1;Introduction;463
4.20.1.2;System Design;463
4.20.1.2.1;Goals;463
4.20.1.2.2;System Structure Design;464
4.20.1.2.3;Database Design;464
4.20.1.3;System Implementation;465
4.20.1.3.1;Knowledge Point Editor;465
4.20.1.3.2;Test Question Browsing and Editing;466
4.20.1.4;Summary;467
4.20.1.5;References;467
4.20.2;Embedded Real-Time Database System Concurrency Control Protocol A-Based FDA;468
4.20.2.1;Introduction;468
4.20.2.2;Alternative and Concurrency Control;469
4.20.2.2.1;Alternative Features;469
4.20.2.2.2;Effects of Alternative on Concurrency Control;469
4.20.2.3;Concurrency Control Protocol Based on Alternative A-BASED FDA;470
4.20.2.3.1;A-Based FDA's Basic Strategy;470
4.20.2.3.2;A-Based FDA Implementation;472
4.20.2.4;A-Based FDA Advantages;472
4.20.2.5;Summary and Outlook;473
4.20.2.6;References;473
4.21;Data Streams;24
4.21.1;A Data Mining Algorithm of Frequent Pattern for Data Flow Based on Landmark Window;474
4.21.1.1;Introduction;474
4.21.1.2;Two-Tuple Transaction Algorithm Principle;475
4.21.1.3;Two-Tuple Transaction Representation;475
4.21.1.4;The Data Mining Algorithms Description;476
4.21.1.5;Experiment;477
4.21.1.6;Conclusion;478
4.21.1.7;References;479
4.22;Temporal and Multimedia Databases;24
4.22.1;A Recovery Approach for Real-Time Database Based on Transaction Fusion;480
4.22.1.1;Introduction;480
4.22.1.2;Related Works;481
4.22.1.3;Real-Time Database Recovery Algorithm with Transaction Fusion;481
4.22.1.3.1;Definitions;482
4.22.1.3.2;Recovery Algorithm;482
4.22.1.4;Performance Analysis and Test;484
4.22.1.5;Conclusions;485
4.22.1.6;References;485
4.23;Systems, Platforms, and Middleware;24
4.23.1;A Data Collection Framework with Extensible Protocol Based on XML;487
4.23.1.1;Introduction;487
4.23.1.2;DCFEP XML Core Technology and Implementation;488
4.23.1.2.1;Uniform Web Service Interface(UWSI);489
4.23.1.2.2;XML Protocol Converter (XMLPC);489
4.23.1.2.3;Data Matching Processor (DMP);490
4.23.1.2.4;Hash Cache(HC);491
4.23.1.2.5;Data Check and Backup Device (DCBD);491
4.23.1.2.6;Data Check and Backup Device (DCBD);491
4.23.1.2.7;Thread Scheduling (TS);492
4.23.1.2.8;Protocol Selector (PS);492
4.23.1.3;Application of Building Energy Data Collection for DCFEP XML;492
4.23.1.3.1;The Uniform Code and XML file;492
4.23.1.3.2;Energy Data Collection Experiment;493
4.23.1.4;Conclusion;494
4.23.1.5;References;494
4.24;Data Storage;24
4.24.1;Optimizing Oracle System Based on SGA;496
4.24.1.1;Introduction;496
4.24.1.2;SGA Overview;496
4.24.1.3;Optimization;497
4.24.1.3.1;Shared Pool Optimization;497
4.24.1.3.2;Buffer Cache Optimization;498
4.24.1.3.3;Redo Log Buffer Optimization;499
4.24.1.3.4;Large Pool Optimization;499
4.24.1.3.5; Java Pool Optimization;500
4.24.1.4;Conclusion;500
4.24.1.5;References;500
4.24.2;Road Crosses High Locality Sorting for Navigation Route Planning;502
4.24.2.1;Introduction;502
4.24.2.2;Gap-Free Quantity Binary Sort;503
4.24.2.2.1;Spatial Locality of Continuous Blocking;503
4.24.2.2.2;Cyclic Recursive Sorting;504
4.24.2.3;Continuous Blocking of Sorted Crosses;505
4.24.2.4;Application on Road Network;506
4.24.2.5;References;507
5;Intelligent Systems Track I;24
5.1;Case-Based and Temporal Reasoning;24
5.1.1;Emergency Case Supporting System for EngineeringAccidents by CBR;508
5.1.1.1;Introduction;508
5.1.1.2;Implement Framework of Two-Stage ECSS for Engineering Accidents;509
5.1.1.3;Basic Cycles of CBR in ECSS for Engineering Accidents;509
5.1.1.3.1;Case Presentation;509
5.1.1.3.2;Case Initializing and Case Retrieving;512
5.1.1.3.3;Case Adaptation;513
5.1.1.3.4;Case Maintenance;513
5.1.1.4;Running Result and Analysis;513
5.1.1.5;Conclusion;514
5.1.1.6;References;514
5.2;Decision Support Systems;24
5.2.1;Fast Face Detection Based on Enhanced AdaBoost;516
5.2.1.1;Introduction;516
5.2.1.1.1;Related Work;517
5.2.1.2;Training and Computing Weak Clasthreshold;518
5.2.1.3;Experiment and Analysis;519
5.2.1.4;Conclusions;521
5.2.1.5;References;521
5.3;Fuzzy Systems;25
5.3.1;A Compound Fuzzy Control Studying on Main Road in City;523
5.3.1.1;Introduction;523
5.3.1.2;Intersection Model and Analysis;524
5.3.1.3;Design of Fuzzy Controller in Artery Road;524
5.3.1.3.1;Fuzzy Controller Design Ideas;524
5.3.1.3.2;Fuzzy Controller Design Steps;525
5.3.1.4;Simulation Computer and Analysis;527
5.3.1.5;Conclusion;527
5.3.1.6;References;528
5.3.2;A Semi-supervised Fuzzy SVM Clustering Framework;529
5.3.2.1;Introduction;529
5.3.2.2;Semi-supervised Fuzzy SVM Clustering Framework;530
5.3.2.2.1;SVM;530
5.3.2.2.2;Semi-surprised Fuzzy SVM Clustering Framework;530
5.3.2.2.3;Optimization of Objective Function;531
5.3.2.3;Experimental Results and Discussions;532
5.3.2.4;Concluding Remarks;533
5.3.2.5;References;534
5.3.3;An Algorithm of Building Fuzzy Petri Nets Based on the Simplified Fuzzy Productions;535
5.3.3.1;Introduction;535
5.3.3.2;The Relevant Concepts of FPN;536
5.3.3.2.1;The Formal Definition of FPN;536
5.3.3.2.2;A Series of Basic Concepts of FPN;536
5.3.3.3;Three Types of Basic Fuzzy Production;536
5.3.3.3.1;Formal Definition of Fuzzy Production;537
5.3.3.3.2;Three Types of Fuzzy Production;537
5.3.3.3.3;Simplification Algorithm of “or” Fuzzy Production;538
5.3.3.4;An Algorithm of Building FPN Based on Fuzzy Productions;538
5.3.3.4.1;One by One Relationship between Fuzzy Productions and FPN;538
5.3.3.4.2;Main Body of This Algorithm;539
5.3.3.5;A Case in These Algorithms;540
5.3.3.5.1;Fuzzy Production Base as “Ethernet Has Problem”;540
5.3.3.5.2;To Further Process the ‘or’ Fuzzy Productions by Algorithm1;541
5.3.3.5.3;The Final FPN Model;541
5.3.3.6;Conclusion;542
5.3.3.7;References;542
5.3.4;An Effective Approach to Detect Hard Exudates in Color Retinal Image;544
5.3.4.1;Introduction;544
5.3.4.2;Proposed Alogorithm;545
5.3.4.2.1;Preprocessing;545
5.3.4.2.2;Histogram Thresholding;546
5.3.4.2.3;Fuzzy C-Mean Segmentation;547
5.3.4.3;Experiment and Discussion;547
5.3.4.4;Conclusion;549
5.3.4.5;References;549
5.3.5;Design and Simulation of Fuzzy Control Algorithm in OCC Processing;550
5.3.5.1;Introduction;550
5.3.5.2;Fuzzy Control Algorithm Design;551
5.3.5.2.1;System Model;551
5.3.5.2.2;Fuzzy Rules;552
5.3.5.3;Fuzzy Control Process Simulation;553
5.3.5.3.1;Simulation Model;553
5.3.5.3.2;Simulation Results and Discussion;553
5.3.5.4;Conclusions;556
5.3.5.5;References;556
5.4;Genetic and Evolutionary Algorithms;25
5.4.1;Optimal Hierarchical Remote Sensing Image Clustering Using Imperialist Competitive Algorithm;557
5.4.1.1;Introduction;557
5.4.1.2;Imperialist Competitive Algorithm;558
5.4.1.3;Hierarchical Image Clustering Using ICA;560
5.4.1.4;Implementation Results;561
5.4.1.5;Conclusion;563
5.4.1.6;References;563
5.4.2;Optimization Estimation of Muskingum Model Parameter Based on Genetic Algorithm;564
5.4.2.1;Introduction;564
5.4.2.2;Principle of Muskingum Model;565
5.4.2.3;Principle and Process of Genetic Algorithm;566
5.4.2.4;The Application of Genetic Algorithm for Muskingum Model Parameter Estimation;567
5.4.2.5;Conclusion;569
5.4.2.6;References;570
5.4.3;Recursive Genetic Algorithm for Robot Manipulator Motion Planning in the Existence of Obstacles;571
5.4.3.1;Introduction;571
5.4.3.2;Problem Definition;572
5.4.3.3;Methodology;573
5.4.3.3.1;Representation;573
5.4.3.3.2;Trajectory Generation;573
5.4.3.3.3;Genetic Operators;573
5.4.3.3.4;Fitness Function;574
5.4.3.3.5;The Recursive Genetic Mechanism;575
5.4.3.4;Results and Discussions;576
5.4.3.4.1;Problem 1;576
5.4.3.4.2;Problem 2;577
5.4.3.4.3;Problem 3;578
5.4.3.4.4;Problem 4;578
5.4.3.5;Conclusions and Future Work;579
5.4.3.6;References;580
5.4.4;Research on Convergence of Self-adaptive Mutation Algorithm;582
5.4.4.1;Introduction;582
5.4.4.2;Algorithm;582
5.4.4.3;Preparation;583
5.4.4.4;Convergence;585
5.4.4.5;Conclusion;587
5.4.4.6;References;587
5.5;Human-Machine Interaction;25
5.5.1;Head Movements and Animating Facial Expressions Based on Multi-curve Spectrum;588
5.5.1.1;Introduction;588
5.5.1.2;3D Virtual Face Model and Head Movements;589
5.5.1.2.1;The External Modelling of Virtual Face;589
5.5.1.2.2;The Internal Modelling of Virtual Eye;590
5.5.1.2.3;The 3D Movement of Virtual Head;590
5.5.1.3;The Driven of 3D Head Movement Using HFMS;590
5.5.1.3.1;The Expression of HFMS;590
5.5.1.3.2;The Control Mechanism of HFMS;590
5.5.1.4;Experiment Result;591
5.5.1.5;Conclusion;593
5.5.1.6;References;593
5.5.2;Implementation of Eye Tracking with Webcam Images;594
5.5.2.1;Introduction;594
5.5.2.2;Controlling the Flow of the Images;595
5.5.2.3;Detecting Eye Pupil Center;596
5.5.2.4;Conclusions and Future Work;598
5.5.2.5;References;599
5.5.3;Research and Development of the Head Unit of Interactive Humanoid Robot;600
5.5.3.1;Introduction;600
5.5.3.2;Mechanical System of Humanoid Head;601
5.5.3.3;Face Detection and Face Recognition;601
5.5.3.3.1;Face Detection Based on Adaboost;601
5.5.3.3.2;Face Recognition Based on 2DPCA;602
5.5.3.4;Control Unit of Humanoid Robot Head;603
5.5.3.5;The Realization of Human-Computer Interaction Function;604
5.5.3.6;Conclusion;604
5.5.3.7;References;605
5.5.4;The Chinese Describing of GIS Path Based on Continuous Polar Coordinate;606
5.5.4.1;Introduction;606
5.5.4.2;Describing Knowledge Base of Path;607
5.5.4.2.1;Describing Words Base of Path;607
5.5.4.2.2;Landmark Features’ Name Base;607
5.5.4.2.3;Describing Rules Base of Path;608
5.5.4.3;Algorithm of PNLD Based on Continuous Polar Coordinate;608
5.5.4.3.1;Structure of Road Network Topology;608
5.5.4.3.2;Preprocessing of Road Nodes;609
5.5.4.3.3;Calculating Polar Coordinate of Nodes;610
5.5.4.3.4;Constructing of Sentence of PNLD;611
5.5.4.4;Prototype System;612
5.5.4.5;Conclusions;613
5.5.4.6;References;613
5.5.5;Usability Evaluation and Redesign of E-Government: Users’ Centred Approach;614
5.5.5.1;Introduction;614
5.5.5.2;Literature Review;615
5.5.5.3;Methodology;616
5.5.5.3.1;Task Design;616
5.5.5.3.2;Usability Questionnaire;617
5.5.5.3.3;E-Government Website Selection;617
5.5.5.3.4;Participants;617
5.5.5.3.5;Experimental Evaluation Procedure;618
5.5.5.4;Results of Experiment 1;618
5.5.5.4.1;User’ Perception;618
5.5.5.4.2;Users’ Performance;619
5.5.5.5;Design of the Proposed Solutions;620
5.5.5.6;Results of Experiment 2;621
5.5.5.6.1;Users’ Perception;621
5.5.5.6.2;Users’ Performance;622
5.5.5.7;Conclusion;623
5.5.5.8;References;623
5.6;Neural Networks;26
5.6.1;A Research Based on a Modified Genetic Algorithm for the Overfitting of Resonance Maching Network;625
5.6.1.1;Introduction;625
5.6.1.2;A Learning Strategy of Network Improvement Bases on the Genetic Algorithm;626
5.6.1.2.1;A Kind of Modified Adaptive Genetic Algorithm;626
5.6.1.2.2;A Network Learning Strategy Based on the Modified Genetic Algorithm;628
5.6.1.3;Experiments;629
5.6.1.3.1;Data Description;629
5.6.1.3.2;Initialization;629
5.6.1.3.3;Experiments and Results;629
5.6.1.4;Conclusion;630
5.6.1.5;References;630
5.6.2;Application of BP Neural Network in History Match and Productivity Prediction of Coalbed Methane;631
5.6.2.1;Introduction;631
5.6.2.2;Basic Principles of BP Neural Network;632
5.6.2.3;Predicting Technique of BP Network;632
5.6.2.4;Application of BP Network in Productivity Matching and Predicting for CBM Wells;633
5.6.2.4.1;Learning Example;633
5.6.2.4.2;Training Example;633
5.6.2.4.3;Match and Predictive Results;634
5.6.2.5;Conclusions;635
5.6.2.6;References;636
5.6.3;Application of RBF Neural Network and ANFIS on the Prediction of Corrosion Rate of Pipeline Steel in Soil;637
5.6.3.1;Introduction;637
5.6.3.2;Introduction of RBF-NN and ANFIS;638
5.6.3.3;Soil Corrosion Test;638
5.6.3.4;Corrosion Rate Predictions;639
5.6.3.5;Results Analysis;640
5.6.3.6;Conclusions;642
5.6.3.7;References;642
5.6.4;Dynamic Network Clustering with Affinity Propagation;643
5.6.4.1;Introduction;643
5.6.4.2;Affinity Propagation;644
5.6.4.3;Dynamic Network Clustering Algorithm;645
5.6.4.4;Experiments;646
5.6.4.5;Summary;648
5.6.4.6;References;648
5.6.5;Electric Load Forecasting Based on Improved Grey Neural Network;649
5.6.5.1;Introduction;649
5.6.5.2;Power Load Forecasting Model;649
5.6.5.2.1;Grey System Theory;649
5.6.5.2.2;BP Neural Network;650
5.6.5.2.3;Improved Gray Neural Network;650
5.6.5.3;Algorithm Applied;652
5.6.5.4;Conclusion;653
5.6.5.5;References;653
5.6.6;Forest Growth Simulation Based on Artificial Neural Network;654
5.6.6.1;Introduction;654
5.6.6.2;Materials and Methods;655
5.6.6.3;Results and Analysis;656
5.6.6.3.1;Model Development;656
5.6.6.3.2;Training Results of the Model;657
5.6.6.3.3;Performance Analysis of the Mode;657
5.6.6.4;Conclusions and Discussion;658
5.6.6.5;References;659
5.6.7;Fuzzy Adaptive Back Propagation Model Based on Genetic Algorithm;661
5.6.7.1;Introduction;661
5.6.7.2;TBP-Algorithm;661
5.6.7.3;CFABP Algorithm;662
5.6.7.4;Extracting Fuzzy Rules for CFABP Based on Genetic Algorithm;663
5.6.7.4.1;Encode Fuzzy Rules;664
5.6.7.4.2;Evaluate Individuals;664
5.6.7.5;Simulation Results;665
5.6.7.5.1;Simulation 1: Function Approximation;665
5.6.7.5.2;Simulation 2: Character Recognition;666
5.6.7.6;Conclusion;666
5.6.7.7;References;666
5.6.8;Lagurre Orthogonal Polynomial Basis Functions Neural Network;667
5.6.8.1;Introduction;667
5.6.8.2;Non-linearity Compensation Principle of Thermocouple Sensors;668
5.6.8.3;Lagurre ANN;669
5.6.8.4;Experiments and Results;670
5.6.8.5;Conclusions;671
5.6.8.6;References;671
5.6.9;Study on Fabric Classification with Support Vector Machines Based on Half-Circle Skirts Shape Simulated in 3D Virtual Try-On;673
5.6.9.1;Introduction;673
5.6.9.2;Experiment;674
5.6.9.2.1;Fabric Test;674
5.6.9.2.2;3D Virtual Try-On;674
5.6.9.2.3;Parameters Selection for Objective Evaluation of Half-Circle Skirt Sharp Style;675
5.6.9.2.4;Simulate Validation;675
5.6.9.2.5;Cluster Analysis of Half-Circle Skirt Shape Style;676
5.6.9.3;Model for Fabric Classification on SVM According to Skirt Shape Style;677
5.6.9.3.1;Construction of SVM Classification Model;677
5.6.9.3.2;C-Support Vector Classification Machine Classifier;678
5.6.9.3.3;Experiments of the Fabric Classification Model on SVM;678
5.6.9.4;Conclusion;679
5.6.9.5;References;680
5.6.10;The Classification of Children’s Occupational Therapy Problems Using Neural Network;682
5.6.10.1;Introduction;682
5.6.10.2;Methods;683
5.6.10.3;Results and Discussion;685
5.6.10.4;Conclusion;686
5.6.10.5;References;687
5.7;Pattern Recognition;26
5.7.1;Enhancement of Template-Based Face Detection by Belief Propagation in Ordered Component Search;688
5.7.1.1;Introduction;688
5.7.1.2;The Face Detection Algorithm;690
5.7.1.3;Experimental Results and Discussion;692
5.7.1.4;Conclusion;693
5.7.1.5;References;693
5.7.2;Face Recognition Based on DWT and Improved Linear Discriminant Analysis;694
5.7.2.1;Introduction;694
5.7.2.2;Wavelet Analysis;695
5.7.2.3;Linear Discriminant Analysis;696
5.7.2.4;Experiment;698
5.7.2.5;Conclusion;699
5.7.2.6;References;699
5.7.3;Incomplete Barcode Reading Mechanism with Remote Database Access;700
5.7.3.1;Introduction;700
5.7.3.2;Barcode Systems;700
5.7.3.3;Hardware and Software Architecture;701
5.7.3.4;Recognition;701
5.7.3.4.1;Image Pre-processing;702
5.7.3.4.2;Barcode Localization and Division;702
5.7.3.4.3;Bars Recognition;702
5.7.3.4.4;Code Recognition;703
5.7.3.4.5;Comparing the Results of Bars and Digits Recognition;703
5.7.3.4.6;Data Base Access;703
5.7.3.5;Experimental Results;704
5.7.3.6;Conclusion;705
5.7.3.7;References;705
5.7.4;Local Feature Image Fusion Algorithm Based on Wavelet Transform;706
5.7.4.1;Introduction;706
5.7.4.2;Fusion Algorithms;706
5.7.4.2.1;Fusion Rules;707
5.7.4.2.2;Fusion Procedures;707
5.7.4.3;Fusion Evaluations;708
5.7.4.4;Experiments;709
5.7.4.5;Conclusions;710
5.7.4.6;References;710
5.7.5;One Infrared Face Recognition Method with Matrixized Model;712
5.7.5.1;Instruction;712
5.7.5.2;The Framework of MatLSSVM;713
5.7.5.3;Experiment;715
5.7.5.4;Conclusion;716
5.7.5.5;References;716
5.7.6;One-Dimensional Feature Extraction Method for Iris Image;718
5.7.6.1;Introductions;718
5.7.6.2;Local Texture Image of Iris;718
5.7.6.2.1;The Proposition of Local Texture Image;718
5.7.6.2.2;LTI Method;719
5.7.6.2.3;Analysis of Algorithm LTI;719
5.7.6.3;One-Dimensional Iris Feature Extracting and Encoding;721
5.7.6.4;Registration of Iris Image;722
5.7.6.5;Conclusions;722
5.7.6.6;References;723
5.7.7;Rademacher Complexity Analysis for Matrixized and Vectorized Classifier;724
5.7.7.1;Introduction;724
5.7.7.2;Rademacher Complexity Analysis;725
5.7.7.2.1;MHKS and MatMHKS;725
5.7.7.2.2;Rademacher Complexity Analysis;726
5.7.7.3;Experiments;727
5.7.7.4;Conclusion and Future Work;728
5.7.7.5;References;729
5.7.8;Recognition of Plate Shape Defect Based on RBF-BP Neural Network;730
5.7.8.1;Introduction;730
5.7.8.2;Analysis of the Shape Defect Pattern;731
5.7.8.2.1;Determination of the Basic Plate Shape Patterns;731
5.7.8.2.2;Normalization Processing of Standard Shape;731
5.7.8.3;The Plate Shape Defects Recognition Model of Combinational RBF-BP Neural Network;732
5.7.8.3.1;Determination of Input Neurons;732
5.7.8.3.2;Determination of Output Neurons;732
5.7.8.4;Effectiveness Analysis of Shape Pattern Recognition;733
5.7.8.4.1;Training of BP, RBF-BP Network;733
5.7.8.4.2;The Results of Simulation Analysis;734
5.7.8.4.3;The Analysis of Actual Shape Pattern Recognition;734
5.7.8.5;Conclusion;735
5.7.8.6;References;735
5.7.9;The Texture Feature Extraction for Multi-objective Image in Complex Traffic Environment;736
5.7.9.1;Introduction;736
5.7.9.2;Gray Level Concurrence Matrix;737
5.7.9.2.1;The Principle of Gray Level Concurrence Matrix;737
5.7.9.2.2;Parameters of the CLCM;737
5.7.9.3;Two-Dimensional Wavelet Transform;738
5.7.9.4;Texture Feature Extraction Algorithm of Multi-target Image of Intricate Traffic Environment;738
5.7.9.4.1;Texture Feature Selection of Multi-target Image;738
5.7.9.4.2;Multi-target Image Texture Feature Extraction;739
5.7.9.5;Experiment Results and Analysis;739
5.7.9.6;Conclusion;740
5.7.9.7;References;741
5.8;Machine Learning;27
5.8.1;A Novel Fuzzy Sarsa Learning Incorporated with Ant Colony Optimization;742
5.8.1.1;Introduction;742
5.8.1.2;Fuzzy Sarsa Learning and Ant Colony Optimization;743
5.8.1.2.1;Fuzzy Sarsa Learning;743
5.8.1.2.2;Ant Colony Optimization;744
5.8.1.3;A NACO-FSL Method;745
5.8.1.3.1;Configuration of NACO-FSL;745
5.8.1.3.2;Simulation;746
5.8.1.3.3;Conclusion;746
5.8.1.4;References;746
5.8.2;A Study Based on Distributed Supervised Machine Learning System for Text Classification;748
5.8.2.1;Introduction;748
5.8.2.2;Consistency Analyses;749
5.8.2.3;Classifier Performance Evaluations;749
5.8.2.4;Dempster-Shafer Theory (Evidence Theory) [11];750
5.8.2.5;Distributed Supervised Machine Learning Algorithm;750
5.8.2.5.1;Concepts;750
5.8.2.5.2;Algorithms;751
5.8.2.6;DSMLS Model;752
5.8.2.7;Experiments;752
5.8.2.8;Results;753
5.8.2.9;Conclusions and Future Work;754
5.8.2.10;References;754
5.8.3;Content-Based Remote Sensing Image Retrieval Using Image Multi-feature Combination and SVM-Based Relevance Feedback;756
5.8.3.1;Introduction;756
5.8.3.2;Features Extraction in Remote Sensing Images;757
5.8.3.2.1;Color and Texture Features;757
5.8.3.2.2;Spectral Feature;757
5.8.3.3;Support Vector Machine Based Relevance Feedback;757
5.8.3.3.1;Support Vector Machine;757
5.8.3.3.2;SVM-Based Relevance Feedback;758
5.8.3.4;Experiments;759
5.8.3.4.1;Experimental Environment;759
5.8.3.4.2;Parameter Selection;759
5.8.3.4.3;Experimental Results;759
5.8.3.5;Conclusions and Future Work;761
5.8.3.6;References;761
5.8.4;Ranking Learning Model Applying Visual Features for Image Search;763
5.8.4.1;Introduction;763
5.8.4.2;Relative Works;763
5.8.4.2.1;Ranking Learning;763
5.8.4.2.2;Reranking;764
5.8.4.3;Ranking Learning Model for Image Retrieval;764
5.8.4.3.1;Modeling;764
5.8.4.3.2;The Optimization of Factor w for Ranking Learning;765
5.8.4.4;Model Solution;766
5.8.4.4.1;Solution for Ranking Parameter w;766
5.8.4.4.2;Ranking Prediction;766
5.8.4.5;Experiment Analysis;766
5.8.4.5.1;Experiment Setting;766
5.8.4.5.2;Experimental Results;767
5.8.4.6;Conclusions;767
5.8.4.7;References;768
5.8.5;SVM Classification for Large Data Sets by Support Vector Estimating and Selecting;769
5.8.5.1;Introduction;769
5.8.5.2;Support Vector Machine;770
5.8.5.3;Support Vectors Estimating and Selecting;771
5.8.5.4;Experiments;772
5.8.5.5;Summary;774
5.8.5.6;References;774
6;Author Index;776