Atjaunināt sīkdatņu piekrišanu

Learning from Data Streams in Dynamic Environments 1st ed. 2016 [Mīkstie vāki]

  • Formāts: Paperback / softback, 75 pages, height x width: 235x155 mm, weight: 1416 g, 43 Illustrations, color; 1 Illustrations, black and white; VIII, 75 p. 44 illus., 43 illus. in color., 1 Paperback / softback
  • Sērija : SpringerBriefs in Applied Sciences and Technology
  • Izdošanas datums: 16-Dec-2015
  • Izdevniecība: Springer International Publishing AG
  • ISBN-10: 3319256653
  • ISBN-13: 9783319256658
  • Mīkstie vāki
  • Cena: 46,91 €*
  • * ši ir gala cena, t.i., netiek piemērotas nekādas papildus atlaides
  • Standarta cena: 55,19 €
  • Ietaupiet 15%
  • Grāmatu piegādes laiks ir 3-4 nedēļas, ja grāmata ir uz vietas izdevniecības noliktavā. Ja izdevējam nepieciešams publicēt jaunu tirāžu, grāmatas piegāde var aizkavēties.
  • Daudzums:
  • Ielikt grozā
  • Piegādes laiks - 4-6 nedēļas
  • Pievienot vēlmju sarakstam
  • Formāts: Paperback / softback, 75 pages, height x width: 235x155 mm, weight: 1416 g, 43 Illustrations, color; 1 Illustrations, black and white; VIII, 75 p. 44 illus., 43 illus. in color., 1 Paperback / softback
  • Sērija : SpringerBriefs in Applied Sciences and Technology
  • Izdošanas datums: 16-Dec-2015
  • Izdevniecība: Springer International Publishing AG
  • ISBN-10: 3319256653
  • ISBN-13: 9783319256658
This book addresses the problems of modeling, prediction, classification, data understanding and processing in non-stationary and unpredictable environments. It presents major and well-known methods and approaches for the design of systems able to learn and to fully adapt its structure and to adjust its parameters according to the changes in their environments. Also presents the problem of learning in non-stationary environments, its interests, its applications and challenges and studies the complementarities and the links between the different methods and techniques of learning in evolving and non-stationary environments.

Introduction to learning.- Learning in dynamic environment.- Handling concept drift.- Summary and final comments.
1 Introduction to Learning
1(10)
1.1 Introduction
1(1)
1.2 Framework and Terminology
2(1)
1.3 Learner Design
3(3)
1.4 Performance Evaluation
6(2)
1.5 From Static Toward Dynamic Environments
8(3)
2 Learning in Dynamic Environments
11(22)
2.1 Introduction
11(1)
2.2 Concept Drift Framework
12(4)
2.2.1 Incremental Learning
12(2)
2.2.2 Adaptive Learning
14(2)
2.3 Causes and Kinds of a Concept Drift
16(7)
2.4 Concept Drift Description
23(7)
2.4.1 Drift Speed
23(2)
2.4.2 Drift Severity
25(1)
2.4.3 Drift Influence Zones
26(2)
2.4.4 Drift Occurrence Frequency and Recurrence
28(2)
2.4.5 Drift Predictability
30(1)
2.5 Drift Concept in Real-World Applications
30(3)
3 Handling Concept Drift
33(28)
3.1 Introduction
33(1)
3.2 General Learning Scheme to Handle Concept Drift
34(1)
3.3 General Classification of Methods to Handle Concept Drift
35(1)
3.4 Informed Methods to Handle Concept Drift
36(5)
3.4.1 Methods Based on Supervised Drift Monitoring Indicators
37(2)
3.4.2 Methods Based on Unsupervised Drift Monitoring Indicators
39(2)
3.5 Drift Handling by Single Learner
41(1)
3.6 Drift Handling by Ensemble Learners
42(8)
3.6.1 Ensemble's Training Set Management
43(3)
3.6.2 Ensemble's Structure Management
46(1)
3.6.3 Ensemble's Final Decision Management
46(4)
3.7 Drift Handling by Sequential Data Processing
50(2)
3.8 Window-Based Processing to Handle Concept Drift
52(3)
3.9 Blind Methods to Handle Concept Drift
55(3)
3.10 Drift Handling Evaluation
58(3)
4 Summary and Final Comments
61(10)
4.1 Summary of Chap. 1
61(2)
4.2 Summary of Chap. 2
63(2)
4.3 Summary of Chap. 3
65(2)
4.4 Future Research Directions
67(4)
References 71
Moamar Sayed-Mouchaweh is a Professor at High National Engineering School of Mines Ecole Nationale Supérieure des Mines de Douai, Computer Science and Automatic Labs, Douai-France