Оглавление
Группа авторов. Machine Learning Algorithms and Applications
Table of Contents
List of Illustrations
List of Tables
Guide
Pages
Machine Learning Algorithms and Applications
Acknowledgments
Preface
1. A Learning-Based Visualization Application for Air Quality Evaluation During COVID-19 Pandemic in Open Data Centric Services
1.1 Introduction. 1.1.1 Open Government Data Initiative
1.1.2 Air Quality
1.1.3 Impact of Lockdown on Air Quality
1.2 Literature Survey
1.3 Implementation Details
1.3.1 Proposed Methodology
1.3.2 System Specifications
1.3.3 Algorithms
1.3.4 Control Flow
1.4 Results and Discussions
1.5 Conclusion
References
2. Automatic Counting and Classification of Silkworm Eggs Using Deep Learning
2.1 Introduction
2.2 Conventional Silkworm Egg Detection Approaches
2.3 Proposed Method
2.3.1 Model Architecture
2.3.2 Foreground-Background Segmentation
2.3.3 Egg Location Predictor
2.3.4 Predicting Egg Class
2.4 Dataset Generation
2.5 Results
2.6 Conclusion
Acknowledgment
References
3. A Wind Speed Prediction System Using Deep Neural Networks
3.1 Introduction
3.2 Methodology
3.2.1 Deep Neural Networks
3.2.2 The Proposed Method
3.2.2.1 Data Acquisition
3.2.2.2 Data Pre-Processing
3.2.2.3 Model Selection and Training
3.2.2.4 Performance Evaluation
3.2.2.5 Visualization
3.3 Results and Discussions. 3.3.1 Selection of Parameters
3.3.2 Comparison of Models
3.4 Conclusion
References
4. Res-SE-Net: Boosting Performance of ResNets by Enhancing Bridge Connections
4.1 Introduction
4.2 Related Work
4.3 Preliminaries. 4.3.1 ResNet
4.3.2 Squeeze-and-Excitation Block
4.4 Proposed Model
4.4.1 Effect of Bridge Connections in ResNet
4.4.2 Res-SE-Net: Proposed Architecture
4.5 Experiments. 4.5.1 Datasets
4.5.2 Experimental Setup
4.6 Results
4.7 Conclusion
References
5. Hitting the Success Notes of Deep Learning
5.1 Genesis
5.2 The Big Picture: Artificial Neural Network
5.3 Delineating the Cornerstones
5.3.1 Artificial Neural Network vs. Machine Learning
5.3.2 Machine Learning vs. Deep Learning
5.3.3 Artificial Neural Network vs. Deep Learning
5.4 Deep Learning Architectures. 5.4.1 Unsupervised Pre-Trained Networks
5.4.2 Convolutional Neural Networks
5.4.3 Recurrent Neural Networks
5.4.4 Recursive Neural Network
5.5 Why is CNN Preferred for Computer Vision Applications?
5.5.1 Convolutional Layer
5.5.2 Nonlinear Layer
5.5.3 Pooling Layer
5.5.4 Fully Connected Layer
5.6 Unravel Deep Learning in Medical Diagnostic Systems
5.7 Challenges and Future Expectations
5.8 Conclusion
References
6. Two-Stage Credit Scoring Model Based on Evolutionary Feature Selection and Ensemble Neural Networks
6.1 Introduction
6.1.1 Motivation
6.2 Literature Survey
6.3 Proposed Model for Credit Scoring
6.3.1 Stage-1: Feature Selection
6.3.2 Proposed Criteria Function
6.3.3 Stage-2: Ensemble Classifier
6.4 Results and Discussion
6.4.1 Experimental Datasets and Performance Measures
6.4.2 Classification Results With Feature Selection
6.5 Conclusion
References
7. Enhanced Block-Based Feature Agglomeration Clustering for Video Summarization
7.1 Introduction
7.2 Related Works
7.3 Feature Agglomeration Clustering
7.4 Proposed Methodology
7.4.1 Pre-Processing
7.4.2 Modified Block Clustering Using Feature Agglomeration Technique
7.4.3 Post-Processing and Summary Generation
7.5 Results and Analysis
7.5.1 Experimental Setup and Data Sets Used
7.5.2 Evaluation Metrics
7.5.3 Evaluation
7.6 Conclusion
References
8. Cardiac Arrhythmia Detection and Classification From ECG Signals Using XGBoost Classifier
8.1 Introduction
8.2 Materials and Methods
8.2.1 MIT-BIH Arrhythmia Database
8.2.2 Signal Pre-Processing
8.2.3 Feature Extraction
8.2.4 Classification
8.2.4.1 XGBoost Classifier
8.2.4.2 AdaBoost Classifier
8.3 Results and Discussion
8.4 Conclusion
References
9. GSA-Based Approach for Gene Selection from Microarray Gene Expression Data
9.1 Introduction
9.2 Related Works
9.3 An Overview of Gravitational Search Algorithm
9.4 Proposed Model
9.4.1 Pre-Processing
9.4.2 Proposed GSA-Based Feature Selection
9.5 Simulation Results
9.5.1 Biological Analysis
9.6 Conclusion
References
10. On Fusion of NIR and VW Information for Cross-Spectral Iris Matching
10.1 Introduction
10.1.1 Related Works
10.2 Preliminary Details
10.2.1 Fusion
10.3 Experiments and Results
10.3.1 Databases
10.3.2 Experimental Results
10.3.2.1 Same Spectral Matchings
10.3.2.2 Cross Spectral Matchings
10.3.3 Feature-Level Fusion
10.3.4 Score-Level Fusion
10.4 Conclusions
References
11. Fake Social Media Profile Detection
11.1 Introduction
11.2 Related Work
11.3 Methodology
11.3.1 Dataset
11.3.2 Pre-Processing
11.3.3 Artificial Neural Network
Hyperparameters
11.3.4 Random Forest
11.3.5 Extreme Gradient Boost
11.3.6 Long Short-Term Memory
11.4 Experimental Results
11.5 Conclusion and Future Work
Acknowledgment
References
12. Extraction of the Features of Fingerprints Using Conventional Methods and Convolutional Neural Networks
12.1 Introduction
12.2 Related Work
12.3 Methods and Materials. 12.3.1 Feature Extraction Using SURF
12.3.2 Feature Extraction Using Conventional Methods
12.3.2.1 Local Orientation Estimation
12.3.2.2 Singular Region Detection
12.3.3 Proposed CNN Architecture
12.3.4 Dataset
12.3.5 Computational Environment
12.4 Results
12.4.1 Feature Extraction and Visualization
12.5 Conclusion
Acknowledgements
References
13. Facial Expression Recognition Using Fusion of Deep Learning and Multiple Features
13.1 Introduction
13.2 Related Work
13.3 Proposed Method
13.3.1 Convolutional Neural Network
13.3.1.1 Convolution Layer
13.3.1.2 Pooling Layer
13.3.1.3 ReLU Layer
13.3.1.4 Fully Connected Layer
13.3.2 Histogram of Gradient
13.3.3 Facial Landmark Detection
13.3.4 Support Vector Machine
13.3.5 Model Merging and Learning
13.4 Experimental Results. 13.4.1 Datasets
13.5 Conclusion
Acknowledgement
References
14. AnimNet: An Animal Classification Network using Deep Learning
14.1 Introduction
14.1.1 Feature Extraction
14.1.2 Artificial Neural Network
14.1.3 Transfer Learning
14.2 Related Work
14.3 Proposed Methodology
14.3.1 Dataset Preparation
14.3.2 Training the Model
14.4 Results
14.4.1 Using Pre-Trained Networks
14.4.2 Using AnimNet
14.4.3 Test Analysis
14.5 Conclusion
References
15. A Hybrid Approach for Feature Extraction From Reviews to Perform Sentiment Analysis
15.1 Introduction
15.2 Related Work
15.3 The Proposed System
15.3.1 Feedback Collector
15.3.2 Feedback Pre-Processor
15.3.3 Feature Selector
15.3.4 Feature Validator
15.3.4.1 Removal of Terms From Tentative List of Features on the Basis of Syntactic Knowledge
15.3.4.2 Removal of Least Significant Terms on the Basis of Contextual Knowledge
15.3.4.3 Removal of Less Significant Terms on the Basis of Association With Sentiment Words
15.3.4.4 Removal of Terms Having Similar Sense
15.3.4.5 Removal of Terms Having Same Root
15.3.4.6 Identification of Multi-Term Features
15.3.4.7 Identification of Less Frequent Feature
15.3.5 Feature Concluder
15.4 Result Analysis
15.5 Conclusion
References
16. Spark-Enhanced Deep Neural Network Framework for Medical Phrase Embedding
16.1 Introduction
16.2 Related Work
16.3 Proposed Approach
16.3.1 Phrase Extraction
16.3.2 Corpus Annotation
16.3.3 Phrase Embedding
16.4 Experimental Setup
16.4.1 Dataset Preparation
16.4.2 Parameter Setting
16.5 Results
16.5.1 Phrase Extraction
16.5.2 Phrase Embedding
16.6 Conclusion
References
17. Image Anonymization Using Deep Convolutional Generative Adversarial Network
17.1 Introduction
17.2 Background Information
17.2.1 Black Box and White Box Attacks
17.2.2 Model Inversion Attack
17.2.3 Differential Privacy
17.2.3.1 Definition
17.2.4 Generative Adversarial Network
17.2.5 Earth-Mover (EM) Distance/Wasserstein Metric
17.2.6 Wasserstein GAN
17.2.7 Improved Wasserstein GAN (WGAN-GP)
17.2.8 KL Divergence and JS Divergence
17.2.9 DCGAN
17.3 Image Anonymization to Prevent Model Inversion Attack
17.3.1 Algorithm
17.3.2 Training
17.3.3 Noise Amplifier
17.3.4 Dataset
17.3.5 Model Architecture
17.3.6 Working
17.3.7 Privacy Gain
17.4 Results and Analysis
17.5 Conclusion
References
Index
WILEY END USER LICENSE AGREEMENT
