Machine Learning and Python for Human Behavior, Emotion, and Health Status Analysis

Cover
Half Title
Title Page
Copyright Page
Table of Contents
Preface
Acknowledgments
About the Author
Chapter 1: Smart Assisted Homes, Sensors, and Machine Learning
	1.1 Smart Homes
		1.1.1 Technologies
		1.1.2 Benefits
		1.1.3 Challenges
	1.2 Example Smart Assisted Homes
	1.3 Events in Smart Assisted Homes
	1.4 Sensors in Smart Homes
		1.4.1 Wearable Sensors
		1.4.2 Ambient Sensors
	1.5 Machine Learning
		1.5.1 Supervised Learning
		1.5.2 Unsupervised Learning
		1.5.3 Semi-Supervised Learning
	1.6 Deep Machine Learning
		1.6.1 Transfer Learning
	1.7 Limitations of Machine Learning
		1.7.1 Underfitting and Overfitting
		1.7.2 Other Limitations
	1.8 Conclusion
	References
Chapter 2: Python and Its Libraries
	2.1 Python’s Key Features
	2.2 Python in Practice
	2.3 Python Libraries and Frameworks
	2.4 Python’s Community to Learn
	2.5 Python’s Impact on Education
	2.6 Python 2 Versus Python 3
	2.7 Python’s Role in Data Science and Machine Learning
	2.8 Challenges and Considerations
	2.9 Python Basics
	2.10 Built-in Python Libraries
	2.11 Data Manipulation Libraries
		2.11.1 NumPy
		2.11.2 Statistical Analysis
	2.12 Pandas
		2.12.1 Pandas Data Structures
		2.12.2 Basic Operations
	2.13 Data Visualization
		2.13.1 Matplotlib
		2.13.2 Seaborn
	2.14 Conclusion
	References
Chapter 3: Feature Analysis Using Python
	3.1 Feature Extraction
	3.2 Principal Component Analysis (PCA)
	3.3 Kernel Principal Component Analysis (KPCA)
	3.4 Feature Extraction Using ICA
	3.5 Linear Discriminant Analysis (LDA)
	3.6 Conclusion
	References
Chapter 4: Deep Learning and XAI with Python
	4.1 Introduction
	4.2 Non-Deep Machine Learning
		4.2.1 Support Vector Machines
		4.2.2 Random Forests
		4.2.3 AdaBoost and Gradient Boosting
		4.2.4 Nearest Neighbors
		4.2.5 More Examples
	4.3 Deep Machine Learning
		4.3.1 Convolutional Neural Networks
		4.3.2 Pre-trained CNN Models
		4.3.3 Long Short-Term Memory (LSTM)
		4.3.4 Neural Structured Learning
	4.4 Explainable AI (XAI)
		4.4.1 Local Explanations
		4.4.2 Visual Explanations
		4.4.3 Feature Relevance Explanations
	4.5 Conclusion
	References
Chapter 5: Behavior and Health Status Recognition
	5.1 Wearable Sensor-Based Behaviour Recognition
		5.1.1 Mobile Health Dataset and Application
		5.1.2 PUC-Rio Dataset
		5.1.3 ARem Dataset
		5.1.4 WISDM Dataset
		5.1.5 Real-Time HAR Using Wearable Sensor
	5.2 Video Camera-Based Behavior Recognition
	5.3 Ambient Sensor-Based Behavior Recognition
		5.3.1 Real-Time Home Monitoring Using Ambient Sensors
		5.3.2 Occupancy Prediction Dataset
	5.4 Health Status Monitoring
		5.4.1 LSTM for Prediction of Health Status
		5.4.2 ARIMA for Prediction of Health Status
		5.4.3 Case Study of Oxygen Saturation, Pulse, and Respiration Prediction
		5.4.4 Sleep Quality Analysis
		5.4.5 Case Study of Sleep Quality Analysis
	5.5 Synthetic Data Generation
	5.6 Conclusion
	Acknowledgement
	References
Chapter 6: Emotion Recognition
	6.1 Image-based Emotion Recognition
	6.2 Case Studies for Image-Based Emotion Recognition
		6.2.1 Local Directional Strength Pattern (LDSP)
		6.2.2 Principal Component Analysis on LDSP
		6.2.3 Linear Discriminant Analysis on PCA
		6.2.4 Facial Expression Modeling
		6.2.5 Experiments on Depth Dataset
		6.2.6 Experiments on RGB-based Public Database
		6.2.7 Experiments on Depth-based Public Database
	6.3 Sample Code for Image-based Emotion Recognition
	6.4 Real-Time Image-based Emotion Recognition
	6.5 Voice-based Emotion Recognition
	6.6 Case Studies on Voice-based Emotion Recognition
		6.6.1 Signal Pre-processing
		6.6.2 Feature Extraction with MFCC
		6.6.3 Emotion Modelling
		6.6.4 Experiments and Results
	6.7 Sample Code for Voice-based Emotion Recognition
	6.8 Conclusion
	References
Index