Advanced Deep Learning for Engineers and Scientists: A Practical Approach

Preface
Acknowledgments
About the Book
Contents
About the Editors
Introduction to Deep Learning
	1 Introduction
	2 Neurons
	3 History of Deep Learning
	4 Feed-Forward Neural Networks
		4.1 Backpropagation
	5 Types of Deep Learning Networks
	6 Deep Learning Architecture
		6.1 Supervised Learning
			6.1.1 Multilayer Perceptron (MLP)
			6.1.2 Recurrent Neural Network (RNN)
			6.1.3 Convolutional Neural Network (CNN)
		6.2 Unsupervised Learning
			6.2.1 Autoencoder (AE)
			6.2.2 Restricted Boltzmann Machine (RBM)
	7 Platforms for Deep Learning/Deep Learning Frameworks
		7.1 TensorFlow
		7.2 Microsoft Cognitive Toolkit
		7.3 Caffe
		7.4 DeepLearning4j
		7.5 Keras
		7.6 Neural Designer
		7.7 Torch
	8 Deep Learning Application
		8.1 Speech Recognition
		8.2 Deep Learning in HealthCare
		8.3 Deep Learning in Natural Language Processing
	9 Conclusion
	References
Deep Learning Applications with Python
	1 Introduction
	2 Deep Learning for Face Recognition
		2.1 Brief Introduction
		2.2 Datasets
		2.3 Practical Example
	3 Deep Learning for Fingerprint Recognition
		3.1 Brief Introduction
		3.2 Datasets
		3.3 Practical Example
	4 Deep Learning for Character Recognition
		4.1 Brief Introduction
		4.2 Datasets
		4.3 Practical Example
	5 Deep Learning for Smart Grids
		5.1 Brief Introduction
		5.2 Datasets
		5.3 Practical Example
	6 Deep Learning in Renewable Energy and Sustainable Development
		6.1 Brief Introduction
		6.2 Datasets
		6.3 Practical Example
	7 Conclusion
	References
Deep Learning for Character Recognition
	1 Character Recognition
		1.1 Challenges in Character Recognition
	2 Deep Learning Approach on Character Recognition
		2.1 Convolutional Neural Networks
	3 Review on Various Character Sets
	4 Implementation of Character Recognition Using Keras and TensorFlow
	5 Summary
	References
Keras and TensorFlow: A Hands-On Experience
	1 TensorFlow Architecture
	2 Introduction to Keras
	3 Installation of TensorFlow and Keras in Jupyter Notebooks: Hardware Aspects
	4 Installation of TensorFlow and Keras in Jupyter Notebooks: Software Aspects
	5 Linear Regression Using Keras: Case Study
	6 Binary Classification Using Keras: Case Study
	7 Multiclass Classification: Case Study
	References
Deploying Deep Learning Models for Various Real-Time Applications Using Keras
	1 Keras
	2 Keras Models
		2.1 Sequential Model
		2.2 Keras Functional Model
		2.3 Standard Network Models
			2.3.1 Multilayer Perceptron (MLP)
			2.3.2 Convolutional Neural Network (CNN)
			2.3.3 Recurrent Neural Networks (RNN)
		2.4 Shared Layers Model
		2.5 Multiple Input and Output Models
	3 Comparison of Frameworks
	4 An Illustration of the Sequential Model
	5 Unstructured data and Structured Data
		5.1 Unstructured Data
		5.2 Structured Data
	6 Deploying Deep Learning Workstation
	7 Binary Classification
	8 Multiclass Classification
	9 Linear Regression Using Keras
	10 Conclusion
	References
Advanced Deep Learning Techniques
	1 ConvNets
		1.1 Introduction to ConvNets
		1.2 Layers
		1.3 Construction and Architecture
	2 RNN, LSTM and GRU
	3 Sequence Processing Using ConvNets
	4 Keras Callbacks and TensorBoard
		4.1 Callbacks
		4.2 TensorBoard
	5 Deep Dream and Neural Style Transfer
	6 Variational Autoencoders
	7 DCGAN
	References
Potential Applications of Deep Learning in Bioinformatics Big Data Analysis
	1 Introduction
	2 Bioinformatics Big Datasets
	3 Concepts in Deep Neural Network
	4 Applications of DNN in Bioinformatics
		4.1 Deep Learning for Omics Research
		4.2 Deep Learning for Protein Structure
		4.3 Deep Learning for Biomedical Image Processing
		4.4 Biomedical Signal Processing
		4.5 Multimodal Deep Learning
	5 Conclusions
	References
Dynamic Mapping and Visualizing Dengue Incidences in Malaysia Using Machine Learning Techniques
	1 Introduction
	2 Background
		2.1 Dengue in Malaysia
	3 Area of Study
	4 Data Collection
		4.1 Mathematical Modelling Using Machine Learning
		4.2 Gaussian Mixed Modelling
		4.3 The k-means Algorithm
	5 k-means Algorithm
		5.1 k-means Clustering Algorithm to Create Initial Vulnerability Map
		5.2 K-Nearest Neighbors’ Algorithm (K-NN)
		5.3 Expectation Maximization (EM) Algorithm
		5.4 Model Selection for EM Algorithm
		5.5 Results and Discussion
		5.6 K-Nearest Neighbor (K-NN) Classification Results
		5.7 Density Plot
	6 Conclusion and Discussion
	References
Vector-Borne Disease Outbreak Prediction Using Machine Learning Techniques
	1 Introduction
	2 Dataset and Description of Model
		2.1 Demographical Data
		2.2 Meteorological Data
		2.3 Disease Outbreak Prediction
			2.3.1 Cost Function
			2.3.2 Feed-Forward
			2.3.3 Backpropagation
			2.3.4 Gradient Descent
		2.4 Methods for Evaluation
	3 Methods
		3.1 Data Imputation and Normalisation
		3.2 ANN-Based Multimodal Disease Outbreak Prediction (ANN-MDOP) Algorithm
			3.2.1 Text Data Representation
			3.2.2 Input Layer of ANN
			3.2.3 Hidden Layer of ANN
			3.2.4 Output Layer of ANN
			3.2.5 Activation Function
			3.2.6 Data Normalisation
			3.2.7 Training the Parameters for ANN-MDOP
	4 Results Obtained from Experiments
		4.1 Effect of Neurons and Hidden Layer
		4.2 Comparison of Dropout Rate
		4.3 Iteration Effect
	5 Analysis of Results
		5.1 Positive Case and Weather Data (P&W-Data)
	6 Conclusion
	References
Eukaryotic Plasma Cholesterol Prediction from Human GPCRs Using K-Means with Support Vector Machine
	1 Introduction
		1.1 Definition of Cell Membrane
		1.2 Components of Cell Membrane
			1.2.1 Phospholipid Bilayer
			1.2.2 Carbohydrates
			1.2.3 Proteins
			1.2.4 Cholesterol
		1.3 G-Protein-Coupled Receptor
	2 Flow of Work Elaboration (Fig. 5)
	3 Methodology Discussion
		3.1 K-Means Clustering
		3.2 Support Vector Machine
	4 Experimental and Result Analysis
	5 Conclusion
	References
A Survey on Techniques for Early Detection of Diabetic Retinopathy
	1 Introduction
	2 Literature Review
		2.1 The Traditional Computer Vision Algorithms for Detection of DR
		2.2 Deep Learning Approaches for Detection of DR
		2.3 Deep Learning Approaches for Detecting Lesions in DR Using CNN-Based Object Detection Models
		2.4 Deep Learning Approaches for DR Detection Using Segmentation
		2.5 Inference
	3 Conclusion and Future Work
	References
Index