Update cookies preferences

E-book: Hybrid Classifiers: Methods of Data, Knowledge, and Classifier Combination

4.00/5 (2 ratings by Goodreads)
Other books in subject:
  • Format - PDF+DRM
  • Price: 110,53 €*
  • * the price is final i.e. no additional discount will apply
  • Add to basket
  • Add to Wishlist
  • This ebook is for personal use only. E-Books are non-refundable.
Hybrid Classifiers: Methods of Data, Knowledge, and Classifier CombinationLarger Image
Other books in subject:

DRM restrictions

  • Copying (copy/paste):

    not allowed

  • Printing:

    not allowed

  • Usage:

    Digital Rights Management (DRM)
    The publisher has supplied this book in encrypted form, which means that you need to install free software in order to unlock and read it.  To read this e-book you have to create Adobe ID More info here. Ebook can be read and downloaded up to 6 devices (single user with the same Adobe ID).

    Required software
    To read this ebook on a mobile device (phone or tablet) you'll need to install this free app: PocketBook Reader (iOS / Android)

    To download and read this eBook on a PC or Mac you need Adobe Digital Editions (This is a free app specially developed for eBooks. It's not the same as Adobe Reader, which you probably already have on your computer.)

    You can't read this ebook with Amazon Kindle

This book delivers a definite and compact knowledge on how hybridization can help improving the quality of computer classification systems. In order to make readers clearly realize the knowledge of hybridization, this book primarily focuses on introducing the different levels of hybridization and illuminating what problems we will face with as dealing with such projects. In the first instance the data and knowledge incorporated in hybridization were the action points, and then a still growing up area of classifier systems known as combined classifiers was considered. This book comprises the aforementioned state-of-the-art topics and the latest research results of the author and his team from Department of Systems and Computer Networks, Wroclaw University of Technology, including as classifier based on feature space splitting, one-class classification, imbalance data, and data stream classification.



This book details how hybridization can help improve the quality of computer classification systems. It introduces the different levels of hybridization and illuminates common problems faced when dealing with such projects.

Reviews

From the book reviews:

The author presents an up-to-date review of recent advances in this area. this is a very interesting, complete, and up-to-date book about various aspects of machine learning and decision making using hybrid classifiers. Although the author makes this book accessible to students and practitioners, it is probably more oriented to advanced undergraduate or graduate courses focused on improving machine learning methods and applications. (Fernando Osorio, Computing Reviews, July, 2014)

1 Introduction
1(58)
1.1 Why Do We Want Machines to Learn
1(4)
1.2 Pattern Classification
5(15)
1.2.1 Pattern Recognition Stages
5(2)
1.2.2 Classification
7(3)
1.2.3 Canonical Model of Classifier
10(1)
1.2.4 Learning Information
11(3)
1.2.5 Uncertainty Representation
14(6)
1.3 Supervised Learning
20(7)
1.3.1 Learning Task
20(1)
1.3.2 Modes
21(2)
1.3.3 Overfitting
23(2)
1.3.4 Bias and Variance
25(2)
1.4 Chosen Classifiers
27(18)
1.4.1 Bayesian Classifiers
27(4)
1.4.2 Minimal Distance Classifiers
31(2)
1.4.3 Rule-Based Classifiers
33(7)
1.4.4 Neural Networks
40(4)
1.4.5 Support Vector Machines
44(1)
1.5 Methods of Classifier Evaluation
45(10)
1.5.1 Initial Remarks
45(2)
1.5.2 Preparing Datasets
47(1)
1.5.3 Evaluation of Binary Classifiers
48(2)
1.5.4 ROC Curve
50(1)
1.5.5 Assessing Classifier Error
51(1)
1.5.6 Comparing Classifiers
52(3)
1.6 Hybrid Classifiers
55(4)
2 Data and Knowledge Hybridization
59(36)
2.1 Motivation
59(1)
2.2 Data and Knowledge Quality
60(2)
2.3 Consistency
62(2)
2.3.1 Knowledge Consistency
62(2)
2.3.2 Data Consistency
64(1)
2.3.3 Data and Knowledge Consistency
64(1)
2.4 Unification
64(11)
2.4.1 Unification a Learning Set into a Rules
65(2)
2.4.2 Unification Rules into a Learning Set
67(1)
2.4.3 Unification Using Hyperrectangles
68(7)
2.5 Cost-Sensitive Classifier
75(5)
2.5.1 Cost Sensitive Decision Tree Induction
76(1)
2.5.2 Cost Sensitive Decision Tree Induction with Cost Limit
76(2)
2.5.3 Experiments on the Cost-Sensitive Decision Tree
78(2)
2.6 Data Privacy
80(15)
2.6.1 Privacy Taxonomy
82(2)
2.6.2 Dataset Partitioning
84(2)
2.6.3 Privacy Preserving Minimal Distance Algorithms
86(2)
2.6.4 Experiments on Privacy Preserving Minimal Distance Algorithms
88(7)
3 Classifier Hybridization
95(46)
3.1 Motivation
95(3)
3.2 Topology
98(1)
3.3 Classifier Ensemble
99(13)
3.3.1 Diversity Measure
100(4)
3.3.2 Diversity Assurance
104(6)
3.3.3 Ensemble Pruning
110(2)
3.4 Combination Rule
112(21)
3.4.1 Taxonomy
113(1)
3.4.2 Fuser Based on Classifier Responses
113(8)
3.4.3 Fusers Based on Discriminants
121(12)
3.5 Hybrid Classifier Learning for Parametric Case
133(8)
4 Chosen Applications of Hybrid Classifiers
141(38)
4.1 Feature Space Splitting
141(14)
4.1.1 Classifier Model
143(1)
4.1.2 Clustering and Selection Algorithm
143(2)
4.1.3 Adaptive Splitting and Selection
145(10)
4.2 Hybrid One-Class Classification
155(11)
4.2.1 One-Class Classification
156(2)
4.2.2 Combining One-Class Classifiers
158(6)
4.2.3 Assuring Diversity of One-Class Classifier Ensembles
164(2)
4.3 Imbalanced Classification
166(2)
4.4 Hybrid Classifiers for Non-stationary Environment
168(11)
4.4.1 Concept Drift
169(1)
4.4.2 Online Learners
170(1)
4.4.3 Sliding Windows
170(1)
4.4.4 Ensemble Approach
171(1)
4.4.5 Drift Detection
172(2)
4.4.6 Weighted Aging Classifier Ensemble
174(5)
5 Conclusions
179(2)
A Appendix
181(10)
A.1 Hypothesis Testing
181(3)
A.2 Dataset Description
184(7)
References 191(24)
Index 215
More information about ebooks More information about ebooks
Permanent link: https://www.kriso.ee/db/97836424099742e.html
Keywords: