Muutke küpsiste eelistusi

E-raamat: Internet-Scale Pattern Recognition: New Techniques for Voluminous Data Sets and Data Clouds [Taylor & Francis e-raamat]

, ,
  • Formaat: 197 pages
  • Ilmumisaeg: 19-Jun-2019
  • Kirjastus: Chapman & Hall/CRC
  • ISBN-13: 9780429096471
Teised raamatud teemal:
  • Taylor & Francis e-raamat
  • Hind: 180,03 €*
  • * hind, mis tagab piiramatu üheaegsete kasutajate arvuga ligipääsu piiramatuks ajaks
  • Tavahind: 257,19 €
  • Säästad 30%
Internet-Scale Pattern Recognition: New Techniques for Voluminous Data Sets and Data CloudsSuurem pilt
  • Formaat: 197 pages
  • Ilmumisaeg: 19-Jun-2019
  • Kirjastus: Chapman & Hall/CRC
  • ISBN-13: 9780429096471
Teised raamatud teemal:
Taylor & Francis e-raamatudRohkem infot Taylor & Francis e-raamatute kohta
Raamatu kodulehekülg: https://www.taylorfrancis.com/books/9780429096471

For machine intelligence applications to work successfully, machines must perform reliably under variations of data and must be able to keep up with data streams. Internet-Scale Pattern Recognition: New Techniques for Voluminous Data Sets and Data Clouds unveils computational models that address performance and scalability to achieve higher levels of reliability. It explores different ways of implementing pattern recognition using machine intelligence.





Based on the authors’ research from the past 10 years, the text draws on concepts from pattern recognition, parallel processing, distributed systems, and data networks. It describes fundamental research on the scalability and performance of pattern recognition, addressing issues with existing pattern recognition schemes for Internet-scale data deployment. The authors review numerous approaches and introduce possible solutions to the scalability problem.





By presenting the concise body of knowledge required for reliable and scalable pattern recognition, this book shortens the learning curve and gives you valuable insight to make further innovations. It offers an extendable template for Internet-scale pattern recognition applications as well as guidance on the programming of large networks of devices.

Preface xi
Acknowledgments xv
About the Authors xvii
I Recognition: A New Perspective
1(32)
1 Introduction
3(12)
1.1 As We See, We Learn
3(1)
1.2 Recognition at a Large Scale
4(4)
1.3 Computational Intelligence Approach for Pattern Recognition
8(3)
1.4 Scalability in Pattern Recognition
11(4)
1.4.1 Common Barriers
11(1)
1.4.2 Possible Solutions
12(1)
1.4.3 Distributed Computing Solution for Scalability of PR Schemes
13(2)
2 Distributed Approach for Pattern Recognition
15(18)
2.1 Scalability of Neural Network Approaches
16(4)
2.1.1 Pattern Storage Capacity
16(1)
2.1.2 Inter-Neuron Communication Frequency
17(3)
2.2 Key Components of DPR
20(2)
2.2.1 Learning Mechanism
20(1)
2.2.2 Processing Approach
21(1)
2.2.3 Training Procedure
21(1)
2.3 System Approaches
22(3)
2.4 Pattern Distribution Techniques
25(2)
2.4.1 Subpattern Distribution
25(1)
2.4.2 Pattern Set Distribution
26(1)
2.5 Current DPR Schemes
27(3)
2.5.1 Graph Neuron
27(2)
2.5.2 Hierarchical Graph Neuron
29(1)
2.5.3 Distributed Hierarchical Graph Neuron
30(1)
2.6 Resource Considerations for DPR Implementations
30(3)
2.6.1 Resource-Aware Approach
31(1)
2.6.2 Message-Passing Model in DPR
31(2)
II Evolution of Internet-Scale Recognition
33(58)
3 One-Shot Learning Considerations
35(14)
3.1 One-Shot Learning Graph Neuron (GN) Scheme
36(5)
3.1.1 Pattern Representation
37(1)
3.1.2 Recognition Procedure
38(3)
3.2 One-Shot Learning Model
41(3)
3.2.1 Bias Array Design for Pattern Memorization
42(1)
3.2.2 Collaborative-Comparison Learning Technique
42(2)
3.3 GN Complexity Estimation
44(2)
3.4 Graph Neuron Limitations
46(2)
3.5 Significance of One-Shot Learning
48(1)
4 Hierarchical Model for Pattern Recognition
49(24)
4.1 Evolution of One-Shot Learning: The Hierarchical Approach
49(8)
4.1.1 Solution to Crosstalk Problem
51(1)
4.1.2 Computational Design for a Hierarchical One-Shot Learning DPR Scheme
52(3)
4.1.3 HGN Recognition Procedure
55(2)
4.2 Complexity and Scalability of Hierarchical DPR Scheme
57(3)
4.2.1 Complexity Estimation
57(3)
4.2.2 Scalability in HGN Approach
60(1)
4.3 Reducing Hierarchical Complexity: A Distributed Approach
60(5)
4.3.1 Distributed Neurons of HGN Network
61(2)
4.3.2 Distributed HGN Approach
63(2)
4.4 Design Evaluation for Distributed DPR Approach
65(8)
4.4.1 Non-Uniform Distribution
65(4)
4.4.2 Uniform Distribution
69(4)
5 Recognition via Divide-and-Distribute Approach
73(18)
5.1 Divide-and-Distribute Approach for One-Shot Learning IS-PR Scheme
73(14)
5.1.1 Associative Memory (AM) Concept in Pattern Recognition
74(1)
5.1.2 DHGN Computational Design
75(5)
5.1.3 Dual-Phase Recognition Procedure
80(7)
5.2 Dimensionality Reduction in Pattern Pre-Processing
87(2)
5.2.1 Structural Reduction
87(1)
5.2.2 Content Reduction
88(1)
5.3 Remarks on DHGN DPR Scheme
89(2)
III Systems and Tools
91(16)
6 Internet-Scale Applications Development
93(14)
6.1 Distributed Computing Models for IS-PR
93(7)
6.1.1 Commodity Grid (CoG)
94(1)
6.1.2 Cloud Computing
94(4)
6.1.3 Peer-to-Peer (P2P) Computing
98(2)
6.2 Parallel Programming Techniques
100(4)
6.2.1 Message-Passing Scheme
100(3)
6.2.2 GPU Programming
103(1)
6.3 From Coding to Applications
104(3)
IV Implementations and Applications
107(52)
7 Multi-Feature Classifications for Complex Data
109(12)
7.1 Data Features for Pattern Recognition
110(1)
7.2 Distributed Multi-Feature Recognition
111(5)
7.2.1 Conceptual Design and Implementation
112(1)
7.2.2 Complexity Estimation
113(3)
7.3 Handwritten Object Classification with Multiple Features
116(4)
7.3.1 Handwritten Object
117(1)
7.3.2 Classification Procedures
118(2)
7.4 Distributed Multi-Feature Recognition Perspective
120(1)
8 Pattern Recognition within Coarse-Grained Networks
121(18)
8.1 Network Granularity Considerations
121(7)
8.1.1 DHGN Configurations for Adaptive Granularity
122(2)
8.1.2 DHGN Commodity Grid Framework
124(4)
8.2 Face Recognition Using the Multi-Feature DPR Approach
128(4)
8.2.1 Color and Spatio-Structural Features Consideration
129(3)
8.3 Distributed Data Management within Cloud Computing
132(6)
8.3.1 Cloud Data Access Scheme
133(2)
8.3.2 DHGN Approach for Cloud Data Access
135(3)
8.4 Adaptive Recognition: A Different Perspective
138(1)
9 Event Detection within Fine-Grained Networks
139(20)
9.1 Distributed Event Detection Scheme for Wireless Sensor Networks
139(9)
9.1.1 WSN Event Detection
140(1)
9.1.2 DHGN-WSN Event Detection Configuration
141(2)
9.1.3 Dimensionality Reduction in Sensory Data
143(1)
9.1.4 Event Classification
144(1)
9.1.5 Performance Metrics: Memory Utilization
145(1)
9.1.6 Spatio-Temporal Analysis of Event Data
146(2)
9.2 Integrated Grid-Sensor Scheme for Structural Analysis
148(8)
9.2.1 Integrated Grid-Sensor Network Framework for Structural Engineering
150(1)
9.2.2 Structural Analysis, Design, and Monitoring Applications
151(5)
9.3 Distributed Event Detection: A Lightweight Approach
156(3)
V The Way Forward
159(8)
10 Recognition: The Future and Beyond
161(6)
10.1 Medium of Change
161(1)
10.2 Future of Internet-Scale PR
162(1)
10.3 Making a Case
163(4)
10.3.1 Changing the Fundamentals
164(1)
10.3.2 Recognition as Commodity
165(2)
Bibliography 167(10)
Index 177
Anang Hudaya Muhamad Amin is a senior lecturer in the Faculty of Information Science and Technology at Multimedia University in Malaysia. He received a BTech (Hons.) in information technology from Universiti Teknologi PETRONAS and a masters in network computing and PhD from Monash University. His research interests include artificial intelligence with specialization in distributed pattern recognition and bio-inspired computational intelligence, wireless sensor networks, and distributed computing.





Asad I. Khan is a senior lecturer in the Faculty of Information Technology at Monash University. Dr. Khan is an Australian Research Council assessor and has published over 80 refereed papers. His research areas include parallel computation, neural networks, and distributed pattern recognition as well as the development of e-research systems and intelligent sensor networks.





Benny Nasution is with the Department of Computer Engineering at Politeknik Negeri Medan. Dr. Nasution was awarded the IBM Award from Tokyo Research Lab and the Mollie Holman Medal from Monash University.
Püsilink: https://www.kriso.ee/db/9780429096471_pe.html