Data Science: Theory, Analysis and Applications

Cover
Half Title
Title Page
Copyright Page
Dedication
Contents
Preface
Editors
Contributors
PART I: Data Science: Theory, Concepts, and Algorithms
	Chapter 1 Framework for Visualization of GeoSpatial Query Processing by Integrating MongoDB with Spark
		1.1 Introduction
			1.1.1 Integration of Spark and MongoDB
		1.2 Literature Survey
		1.3 Proposed System
			1.3.1 Methodology for Processing Spatial Queries
			1.3.2 Spark Master-Slave Framework
			1.3.3 Algorithms for Sharding
				1.3.3.1 Algorithm for Range Sharding
				1.3.3.2 Algorithms for Zone Sharding
			1.3.4 Dataset and Statistics
		1.4 Results and Performance Evaluation
		1.5 Conclusion
		References
	Chapter 2 A Study on Metaheuristic-Based Neural Networks for Image Segmentation Purposes
		2.1 Introduction
		2.2 Supervised Image Segmentation
		2.3 Literature Review
		2.4 Artificial Neural Networks
		2.5 Optimization
		2.6 Metaheuristic Algorithms
			2.6.1 Genetic Algorithm
			2.6.2 Particle Swarm Optimization Algorithm
			2.6.3 Imperialist Competitive Algorithm
		2.7 Optimization of the Neural Networks Weights Using Optimization Algorithms
		2.8 Experimental Setup and Method Analysis
		2.9 Conclusions
		References
	Chapter 3 A Study and Analysis of a Feature Subset Selection Technique Using Penguin Search Optimization Algorithm
		3.1 Introduction
		3.2 Literature Review
		3.3 Proposed Work
			3.3.1 Pseudocode of the Proposed FS-PeSOA Algorithm
			3.3.2 Discussion
				3.3.2.1 Hunting Strategy of Penguins
				3.3.2.2 Fitness Function Evaluation
				3.3.2.3 Position Update Logic
				3.3.2.4 Oxygen Update Logic
		3.4 Result Analysis
		3.5 Conclusions
		References
	Chapter 4 A Physical Design Strategy on a NoSQL DBMS
		4.1 Introduction
		4.2 Motivation Example
		4.3 Neo4j
		4.4 Design Guidelines
		4.5 Physical Design
			4.5.1 Query Rewriting Using Path Redundancy Pattern
			4.5.2 Query Rewriting Using Minimal Query Pattern
			4.5.3 Path Materialization
			4.5.4 Index Creation
		4.6 Experimental Study
			4.6.1 Experimental Design
			4.6.2 Impact of the Proposed Physical Design on Query Performance for a 1 GB Database
			4.6.3 Impact of the Proposed Physical Design on Query Performance for a 10 GB Database
			4.6.4 Impact of the Proposed Physical Design on Query Performance for a 100 GB Database
		4.7 Related Work
		4.8 Discussion
		4.9 Future Research Directions
		4.10 Conclusion
		References
	Chapter 5 Large-Scale Distributed Stream Data Collection Schemes
		5.1 Introduction
		5.2 Data Collection Scheme for Distributed TBPS
			5.2.1 Assumed Environment
				5.2.1.1 Assumed TBPS Architecture
				5.2.1.2 Assumed Overlay for Distributed TBPS
			5.2.2 Proposed Method
				5.2.2.1 Methodology Principle
				5.2.2.2 Collective Store and Forwarding
				5.2.2.3 Adaptive Data Collection Tree
			5.2.3 Evaluation
				5.2.3.1 Simulation Parameters
				5.2.3.2 Simulation Results
		5.3 Data Collection Scheme Considering Phase Differences
			5.3.1 Problems Addressed
				5.3.1.1 Assumed Environment
				5.3.1.2 Input Setting
				5.3.1.3 Definition of a Load
			5.3.2 Proposed Method
				5.3.2.1 Skip Graphs
				5.3.2.2 Phase Differences
			5.3.3 Evaluation
				5.3.3.1 Collection Target Nodes
				5.3.3.2 Communication Loads and Hops
		5.4 Discussion
		5.5 Related Work
		5.6 Conclusion
		Acknowledgements
		References
PART II: Data Design and Analysis
	Chapter 6 Big Data Analysis and Management in Healthcare
		6.1 Introduction
		6.2 Preliminary Studies
		6.3 Healthcare Data
		6.4 Need of Big Data Analytics in Healthcare
		6.5 Challenges in Big Data Analysis in Healthcare
			6.5.1 Capture
			6.5.2 Cleaning
			6.5.3 Storage
			6.5.4 Security
			6.5.5 Stewardship
			6.5.6 Querying
			6.5.7 Reporting
			6.5.8 Visualization
			6.5.9 Updating
			6.5.10 Sharing
		6.6 Collection of Healthcare Data
			6.6.1 Importance in Healthcare Data Collection
			6.6.2 Complications and Clarifications of Healthcare Data Collection
			6.6.3 Current Data Collection Methods
			6.6.4 Advanced Data Collection Tools
			6.6.5 Healthcare Data Standards
			6.6.6 Inferences of Patient Data Collection in Healthcare
		6.7 Analysis of Healthcare Data
		6.8 Healthcare Data Management
			6.8.1 Big Data and Care Management
			6.8.2 Advantages of Healthcare Data Management
		6.9 Big Data in Healthcare
			6.9.1 Big Data and IoT
			6.9.2 Patient Prophecies for Upgraded Staffing
			6.9.3 Electronic Health Records
			6.9.4 Real-Time Warning
			6.9.5 Augmenting Patient Engagement
			6.9.6 Using Health Data for Informed Strategic Planning
			6.9.7 Extrapolative Analytics in Healthcare
			6.9.8 Diminish Fraud and Enrich Security
			6.9.9 Telemedicine
			6.9.10 Assimilating Big Data per Medical Imaging
			6.9.11 A Method to Avert Pointless ER (Emergency Room) Visits
		6.10 Future for Big Data in Healthcare
		6.11 Conclusion
		References
	Chapter 7 Healthcare Analytics: A Case Study Approach Using the Framingham Heart Study
		7.1 Introduction and Background to the Case Study: Framingham Heart Study
		7.2 Literature Review
		7.3 Introduction to the Data Analytics Framework
			7.3.1 Step 1. Defining the Healthcare Problem
			7.3.2 Step 2. Explore the Healthcare Data
			7.3.3 Step 3. Predict What Is Likely to Happen; or Perform Classification Analysis
			7.3.4 Step 4. Check the Modeling Results
			7.3.5 Step 5. Optimize (Find the Best Solution)
			7.3.6 Step 6. Derive a Clinical Strategy for Patient Care and Measure the Outcome
			7.3.7 Step 7. Update the CDS System
		7.4 Data Exploration and Understanding of the Healthcare Problem
		7.5 Machine-Learning Model Application
		7.6 Evaluation of the Machine-Learning Model Results
		7.7 Conclusion
		7.8 Future Direction
		Acknowledgements
		References
	Chapter 8 Bioinformatics Analysis of Dysfunctional (Mutated) Proteins of Cardiac Ion Channels Underlying the Brugada Syndrome
		8.1 Introduction
		8.2 Results
			8.2.1 Brief Description of Unique BrS-Related Proteins
			8.2.2 PIM-Based Analysis of the Unique BrS-Related Proteins
			8.2.3 Intrinsic Disorder Analysis of the BrS-Related Proteins
			8.2.4 Kolmogorov–Smirnov Test
		8.3 Discussion
		8.4 Materials and Methods
			8.4.1 Evaluation of Polar Profile
				8.4.1.1 Weighting of Polar Profiles
				8.4.1.2 Comparison of Polar Profiles
				8.4.1.3 Graphics of Polar Profiles
			8.4.2 Evaluation of Intrinsic Disorder Predisposition
			8.4.3 Data Files
			8.4.4 Kolmogorov–Smirnov Test
			8.4.5 Test Plan
				8.4.5.1 Polar Profile
		8.5 Conclusions
		References
	Chapter 9 Discrimination of Healthy Skin, Superficial Epidermal Burns, and Full-Thickness Burns from 2D-Colored Images Using Machine Learning
		9.1 Introduction
		9.2 Literature Review
			9.2.1 Skin Burns
			9.2.2 Causes of Burn Injuries
			9.2.3 Burns Category
			9.2.4 Burn Assessment Techniques
				9.2.4.1 Clinical Assessment
				9.2.4.2 Blood Perfusion Measurement
		9.3 Machine Learning
			9.3.1 Convolutional Neural Networks
				9.3.1.1 Convolution Layer
				9.3.1.2 Pooling Layer
				9.3.1.3 Output/Classification Layer
			9.3.2 Training a ConvNet
			9.3.3 Common ConvNet Models
				9.3.3.1 AlexNet
				9.3.3.2 GoogleNet
				9.3.3.3 VGGNet
				9.3.3.4 Residual Network
		9.4 Goals and Methodology
			9.4.1 Image Acquisition and Preprocessing
			9.4.2 Feature Extraction and Classification
		9.5 Results and Discussion
			9.5.1 Terms Related to Contingency Table
			9.5.2 Classifier Performance
		9.6 Conclusions
		References
	Chapter 10 A Study and Analysis of an Emotion Classification and State-Transition System in Brain Computer Interfacing
		10.1 Introduction
		10.2 Literature Review
		10.3 Proposed Work
			10.3.1 Classification Processes
				10.3.1.1 SVM Classifier
				10.3.1.2 KNN Classifier
				10.3.1.3 Random Forest Classifier
			10.3.2 State-Transition Machine
				10.3.2.1 Proposed Algorithm of Emotional State Transition Based on Channel Value for a Fixed Time Interval
		10.4 Result Analysis
			10.4.1 Requirement
			10.4.2 Result Comparisons of SVM, KNN, and Random Forest Classifiers
			10.4.3 SVM Polynomial Kernel Performance Analysis
			10.4.4 Analysis of the State-Transition Machine
			10.4.5 Comparison with Previous Works
			10.4.6 Computational Complexity
		10.5 Conclusion
		Acknowledgment
		References
PART III: Applications and New Trends in Data Science
	Chapter 11 Comparison of Gradient and Textural Features for Writer Retrieval in Handwritten Documents
		11.1 Introduction
		11.2 Literature Review
		11.3 Adopted Features
			11.3.1 Local Binary Pattern
			11.3.2 Histogram of Oriented Gradients
			11.3.3 Gradient Local Binary Pattern
			11.3.4 Pixel Density
			11.3.5 Run Length Feature
		11.4 Matching Step
		11.5 Experimental Evaluation
			11.5.1 Evaluation Criteria
			11.5.2 Experimental Setup
			11.5.3 Retrieval Results
		11.6 Discussion and Comparison
		11.7 Conclusion
		References
	Chapter 12 A Supervised Guest Satisfaction Classification with Review Text and Ratings
		12.1 Introduction
		12.2 Related Literature
			12.2.1 Guest Satisfaction and Online Reviews
		12.3 Methodology
			12.3.1 Data Description and Analysis
			12.3.2 Data Cleaning
			12.3.3 Latent Semantic Analysis
			12.3.4 Classifiers and Performance Measures
		12.4 Experimental Results
			12.4.1 Features Related to Guest Satisfaction
			12.4.2 Hotel Guest Satisfaction Prediction
		12.5 Discussions and Conclusion
		12.6 Implications
			12.6.1 Theoretical Implications
			12.6.2 Managerial Implications
		12.7 Limitations and Future Scope
		References
	Chapter 13 Sentiment Analysis for Decision-Making Using Machine Learning Algorithms
		13.1 Introduction
		13.2 Literature Review
			13.2.1 Related Studies and Techniques
		13.3 Methods
			13.3.1 Naïve Bayes Classifier
			13.3.2 Support Vector Machine
		13.4 Data Analysis and Algorithm Initialization
			13.4.1 Experimental Setup
			13.4.2 Data Preparation and Model Initialization
			13.4.3 Evaluation Measures
			13.4.4 Planned Approach
		13.5 Results and Model Evaluation
		13.6 Conclusion and Future Works
		13.7 Acknowledgment
		References
	Chapter 14 Deep Learning Model: Emotion Recognition from Continuous Action Video
		14.1 Introduction
		14.2 Related Works
		14.3 Learning
			14.3.1 Nondeep Learning-Based Approaches
				14.3.1.1 Dictionary Learning Approaches
				14.3.1.2 Genetic Programming Approaches
			14.3.2 Deep Learning-Based Approaches
				14.3.2.1 Generative/Unsupervised Approaches
				14.3.2.2 Discriminative/Supervised Approaches
			14.3.3 Convolutional Neural Network
				14.3.3.1 Convolutional Layer
				14.3.3.2 Subsampling Layers
				14.3.3.3 Rectified Linear Unit
				14.3.3.4 Fully Connected Layer
				14.3.3.5 Softmax Layer
				14.3.3.6 Output Layer
			14.3.4 Feedforward Deep Convolutional Neural Network
			14.3.5 VGG16 Model
		14.4 Performance Evaluation Metrics
			14.4.1 Dataset
			14.4.2 Performance Evaluation Metrics
		14.5 Results and Discussions
		14.6 Conclusion
		References
Index