Cognitive Informatics, Computer Modelling, and Cognitive Science: Application to Neural Engineering, Robotics, and Stem: Volume 2: Application to Neural Engineering, Robotics, and STEM

Cognitive Informatics, Computer Modeling, and Cognitive Science
Copyright
Dedication
Contents
List of contributors
Editors’ biographies
Authors’ biography
Preface
Acknowledgments
1 Approaches from cognitive neuroscience and comparative cognition
	1.1 Introduction
	1.2 Cognitive science
	1.3 Neuroscience
	1.4 Python
	1.5 Review of literature
	1.6 Cognitive neuroscience/physiology
	1.7 Cognitive psychology
	1.8 Conclusion
	References
	Further reading
2 Functional neuroanatomy and disorders of cognition
	Abbreviations
	2.1 Introduction
	2.2 Neuroanatomy of memory encoding
		2.2.1 Medial temporal lobe
		2.2.2 Diencephalon
		2.2.3 Basal forebrain
	2.3 Mechanisms underlying memory formation
	2.4 Neurotransmitters involved in cognition
		2.4.1 Classical neurotransmitters
			2.4.1.1 Acetylcholine
			2.4.1.2 Glutamate
			2.4.1.3 γ-Aminobutyric acid
			2.4.1.4 Dopamine
			2.4.1.5 Serotonin (5-hydroxytryptamine)
			2.4.1.6 Agmatine
		2.4.2 Neuropeptides
			2.4.2.1 Cocaine- and amphetamine-regulated transcript
			2.4.2.2 Neuropeptide Y
			2.4.2.3 α-Melanocyte stimulating hormone
		2.4.3 Neurosteroids
	2.5 Cognition-related diseases
		2.5.1 Alzheimer’s disease
			2.5.1.1 Extraneuronal plaque deposition of β-amyloid
			2.5.1.2 Intraneuronal accumulation of neurofibrillary tangles
		2.5.2 Lewy body diseases
	2.6 Conclusion
	2.7 Acknowledgment
	References
	Further reading
3 A cognitive system of elderly exercise evaluation with sensors and robots
	3.1 Introduction
	3.2 System overview
	3.3 Elderly exercise measurement
	3.4 Exercise evaluation
	3.5 Feedback by robot interface
	3.6 Multiple Kinect application for occlusion problem
		3.6.1 Frame synchronization
		3.6.2 Sensing data integration without calibration
	3.7 Conclusion
	Acknowledgment
	References
4 Models of making choice and control over thought for action
	4.1 Outline of review
	4.2 Introduction
	4.3 Models of perceptual decision
		4.3.1 Fast decision-making
		4.3.2 Intuitive decision-making
	4.4 Models of economic decision
	4.5 Models of movement inhibition
		4.5.1 Proactive control
		4.5.2 Estimation of stopping efficacy
		4.5.3 Trigger failures
		4.5.4 Bayesian rational decision-making
		4.5.5 Optimal Bayesian statistical inference
		4.5.6 Decision process as optimal stochastic control
		4.5.7 Linear approach to threshold explaining space and time model for decisions in space and time
	4.6 Discussion
	Conflict of interest
	Acknowledgments
	References
	Further reading
5 Speech recognition technique for identification of raga
	5.1 Introduction
	5.2 Speech recognition
	5.3 Applications of speech recognition
	5.4 Speech analyses in music information retrieval
	5.5 A brief history of Indian music
	5.6 Mathematical structure of Carnatic music
	5.7 Digital speech processing
	5.8 Proposed methodology for classification of raga
	5.9 A practical example using Praat
	5.10 Conclusion
	Reference
	Further reading
6 Future of cognitive science
	6.1 Introduction
	6.2 Role of cognitive science in varied domains
		6.2.1 Cognitive science for big data
		6.2.2 Cognitive science for philosophy
		6.2.3 Brain–machine interface
		6.2.4 Cognition science for psychology
		6.2.5 Cognition social science
		6.2.6 Role of cognitive science in linguistics
		6.2.7 Cognitive control
		6.2.8 Cognitive image processing
	6.3 Future of cognitive neuroscience and cognitive enhancement
		6.3.1 Scope for neuroscience research and challenges
		6.3.2 Cognitive enhancement
		6.3.3 Ethical issues and concerns of cognitive enhancement
	6.4 Conclusion
	References
7 Application of virtual reality systems to psychology and cognitive neuroscience research
	7.1 Introduction
		7.1.1 Cognitive science
		7.1.2 Virtual reality
	7.2 Literary survey review
		7.2.1 Cognitive neuroscience/physiology
		7.2.2 Cognitive psychology
	7.3 Conclusion
	References
	Further reading
8 Electrodermal activity and its effectiveness in cognitive research field
	8.1 Introduction
	8.2 History of electrodermal activity signal, psychophysiological, and physiological mechanism behind electrodermal activity
		8.2.1 Application of electrodermal activity
		8.2.2 Electrodermal activity as an indicator of general arousal
		8.2.3 Electrodermal activity in different sleep stages
		8.2.4 Electrodermal indices of emotion and stress
	8.3 Experiment design—a good experiment design
		8.3.1 Experimental design
			8.3.1.1 Experiment design
			8.3.1.2 Types of experiments
			8.3.1.3 Hypothesis
			8.3.1.4 Stimulus
			8.3.1.5 Measure of performance
		8.3.2 External and internal influences
		8.3.3 Climatic conditions
		8.3.4 Internal or physiological influences
		8.3.5 Demographic characteristics
	8.4 Electrodermal activity signal collection sites and pretreatment of sites
		8.4.1 Electrodermal activity signal collection sites
		8.4.2 Pretreatment of sites
	8.5 Artifacts removal from the electrodermal activity signal
	8.6 Analysis of electrodermal activity signal
		8.6.1 Phasic electrodermal activity
			8.6.1.1 Latency
			8.6.1.2 Amplitude
			8.6.1.3 Shape of electrodermal responses
		8.6.2 Area measurements
		8.6.3 Tonic electrodermal activity
	8.7 End remarks
	References
	Further reading
9 Study of modern brain-imaging and -signaling techniques for brain–computer interface
	9.1 Introduction
	9.2 Brain-imagining techniques
		9.2.1 Computer tomography
			9.2.1.1 Computer tomography head
				9.2.1.1.1 Benefits
				9.2.1.1.2 Risk and limitation
		9.2.2 Near-infrared spectroscopy–based imaging equipment
			9.2.2.1 Functional near-infrared spectroscopy
			9.2.2.2 Diffuse optical imaging or diffuse optical tomography
			9.2.2.3 High-density diffuse optical tomography
				9.2.2.3.1 Advantages and disadvantages of optical imaging
		9.2.3 Magnetic resonance imaging
			9.2.3.1 Magnetic resonance imaging head
			9.2.3.2 Functional magnetic resonance imaging
				9.2.3.2.1 Advantages of magnetic resonance imaging
				9.2.3.2.2 Disadvantages of magnetic resonance imaging
		9.2.4 Single-photon emission computed tomography
			9.2.4.1 Advantages of single-photon emission computed tomography
			9.2.4.2 Disadvantage of single-photon emission computed tomography
		9.2.5 Cranial ultrasound
			9.2.5.1 Advantages of cranial ultrasound
			9.2.5.2 Limitations of cranial ultrasound
	9.3 Brain-signaling techniques
		9.3.1 Electroencephalography
			9.3.1.1 Application of electroencephalography [28]
			9.3.1.2 Advantages of electroencephalography
			9.3.1.3 Disadvantages of electroencephalography
		9.3.2 Magnetoencephalography
			9.3.2.1 Advantages of magnetoencephalography
			9.3.2.2 Limitations of magnetoencephalography
		9.3.3 Electromyography
			9.3.3.1 Applications of electromyography
			9.3.3.2 Advantages of electromyography
			9.3.3.3 Limitations of electromyography
	9.4 Sleep-based disorder analysis using neurodiagnosis techniques
		9.4.1 Polysomnography
			9.4.1.1 Advantages of polysomnograhy
			9.4.1.2 Limitation of polysomnograhy
	9.5 Summary
	References
	Further reading
10 Reading an extremist mind through literary language: approaching cognitive literary hermeneutics to R.N. Tagore’s play T...
	10.1 Introduction
		10.1.1 Why transdisciplinary?
		10.1.2 Tagore’s The Post Office: a cognitive neurology
	10.2 Affecting factors to activate mirror neuron in R.N. Tagore
	10.3 Hypothesis
	10.4 Colonialism/nationalism or national extremism: symptoms psychoneurological disorders
	10.5 The mind of extremist: a neurological observation
	10.6 “Nation is the greatest evil for the Nation”?
	10.7 Amal as a religion under control
	References
	Further Reading
	Recommended Reading
11 REAH: Resolution Engine for Anaphora in Hindi dialogue
	11.1 Introduction
		11.1.1 Categorization of Hindi anaphora
		11.1.2 Boundaries in anaphora resolution
			11.1.2.1 Nonavailability of freeware Hindi discourse
			11.1.2.2 Efficiency of linguistic preprocessor
			11.1.2.3 No benchmark for POS tagging
			11.1.2.4 Lack of efficient named entity recognizer
	11.2 The state-of-the-art
		11.2.1 Background of the authors
	11.3 The resolution engine
		11.3.1 The preprocessing phase
			11.3.1.1 Data annotation
			11.3.1.2 Defining the term patterns
			11.3.1.3 Removal of irrelevant chunks and nonanaphoric
			11.3.1.4 Identification of intermediate clause
			11.3.1.5 Extraction of relevant noun phrases
			11.3.1.6 Distance factors
			11.3.1.7 Identifying inanimate entity
		11.3.2 Anaphora resolution phase
			11.3.2.1 Constraints
			11.3.2.2 Identifying the equivalence class
				11.3.2.2.1 Algorithm for resolving first-person pronouns
				11.3.2.2.2 Algorithm for resolving second-person pronouns
				11.3.2.2.3 Algorithm for resolving third-person pronouns
				11.3.2.2.4 Algorithm for resolving reflexive pronouns
				11.3.2.2.5 Algorithm for resolving locative pronouns
				11.3.2.2.6 Algorithm for resolving demonstrative pronouns
	11.4 Test datasets
	11.5 Experiments and evaluations
	11.6 Conclusion
	References
12 Surveying various effective modes and research trends on cognitive Internet of Things over wireless sensor network
	12.1 Introduction
	12.2 Objects with computing devices and AI
		12.2.1 Internet of Things
		12.2.2 Objects with computing devices and computerized ones
		12.2.3 Objects with computing devices is not AI
		12.2.4 Need for AI in Internet of Things
	12.3 Intellectual AI and Intellectual compute
		12.3.1 Intellectual AI and cognition, AI
		12.3.2 Intellectual computing
		12.3.3 Further than mechanization
	12.4 Objects with computing devices and Intellectual computing
		12.4.1 The Intellectual Internet of Things
		12.4.2 Ownership of Intellectual Internet of Things
		12.4.3 The pillars of Intellectual Internet of Things
		12.4.4 Challenge of Intellectual Internet of Things
	12.5 Value of Intellectual Internet of Things
	12.6 Areas where we used
		12.6.1 Well turned-out livelihood
		12.6.2 Elegant health
		12.6.3 Household appliances
		12.6.4 Smart cities
		12.6.5 Wiki City
		12.6.6 Synchronized analytics
	12.7 Usecase
	12.8 Conclusion
	References
	Further reading
13 Time and feature specific sentiment analysis of product reviews
	13.1 Introduction
	13.2 Related work
	13.3 Proposed model
	13.4 Need of feature specificity
	13.5 The aging factor
	13.6 Experimental setup
		13.6.1 Collection and preparing of dataset
		13.6.2 Define feature dictionary for product
		13.6.3 Preprocess, tokenize, and vectorize the dataset
		13.6.4 Classify the review tokens under the features in the feature dictionary
		13.6.5 Find the sentiments of the review tokens for each feature
		13.6.6 Multiply the polarity with the aging factor to get the sentiment score of the review term
		13.6.7 Sum up the results for each feature
		13.6.8 Visualize the results
	13.7 Result and discussion
	13.8 Conclusion and future work
	References
14 Language learnability analysis of Hindi: a comparison with ideal and constrained learning approaches
	Glossary
	14.1 Introduction
	14.2 Language acquisition theories
	14.3 Evaluation models
		14.3.1 Bayesian segmentation
		14.3.2 Bayesian inference
	14.4 Data preparation for learnability analysis
		14.4.1 Transliteration
		14.4.2 Syllabification
		14.4.3 Phonemization
	14.5 Results and discussions
	14.6 Conclusion and future work
	Acknowledgments
	References
	Further reading
15 A special report on changing trends in preventive stroke/cardiovascular risk assessment via B-mode ultrasonography
	15.1 Introduction
		15.1.1 Article search strategy
	15.2 Risk assessment using traditional methods
	15.3 Fundamentals of machine learning
		15.3.1 Types of machine learning techniques
		15.3.2 General framework of machine learning
			15.3.2.1 Feature engineering: extraction and selection
			15.3.2.2 Data partitioning
			15.3.2.3 Training model design
			15.3.2.4 Prediction or testing model
			15.3.2.5 Performance evaluation of machine learning systems
		15.3.3 Machine learning–based algorithms
	15.4 Risk assessment in machine learning framework
		15.4.1 Image-based stroke risk assessment using machine learning
		15.4.2 Cardiovascular diseases risk assessment using machine learning
		15.4.3 Cardiovscular disease/stroke risk assessment indices
	15.5 Medical implications of machine learning–based risk assessment
	15.6 Deep learning–based cardiovascular risk stratification
	15.7 Challenges in machine learning design
	15.8 Conclusion
	Acknowledgments
	Funding
	Disclosure
	References
	Appendix: performance evaluation parameters
16 A healthcare text classification system and its performance evaluation: a source of better intelligence by characterizin...
	16.1 Introduction
	16.2 Brief literature survey and our proposed model
		16.2.1 Our model
	16.3 Data types
		16.3.1 Data type 1: TwitterA dataset
		16.3.2 Data type 2: WebKB4 dataset
		16.3.3 Data type 3: Disease dataset
		16.3.4 Data type 4: Reuters (R8) dataset
		16.3.5 Data type 5: SMS dataset
	16.4 Methodology
		16.4.1 Brief discussion on classifiers
			16.4.1.1 Support vector machine
			16.4.1.2 Multilayer perceptron
			16.4.1.3 AdaBoost
			16.4.1.4 Stochastic gradient descent
			16.4.1.5 Decision tree
	16.5 Experiment protocol
		16.5.1 Experimental protocol 1: system classifier accuracy computation over all parameters
			16.5.2 Experimental protocol 2: effect of training data size on classification accuracy
			16.5.3 Experimental protocol 3: overall mean performance using all parameters: D, C, K, and T
			Sensitivity
			Specificity
			Positive predictive value
			Accuracy
	16.6 Results
		16.6.1 Results of protocol #1: system accuracy computation over all parameters
		16.6.2 Results of protocol #2: effect of the training data size on classification accuracy
		16.6.3 Results for the protocol #3: overall mean performance over all D, C, K, and T
	16.7 Hypothesis validation and performance evaluation
		16.7.1 Hypothesis validation
			16.7.1.1 System performance linking misrepresentation ratio with area under the curve of machine learning system
			16.7.1.2 Effect of misrepresentation ratio on machine learning classification accuracy
			16.7.1.3 Effect of misrepresentation ratio on mean area under the curve for all classifiers and all data types
		16.7.2 Individual receiver operating characteristic plots for all K protocols, D data types, and C classifiers
		16.7.3 Reliability and stability analysis
			16.7.3.1 Reliability index
			16.7.3.2 Stability index
	16.8 Discussion
		16.8.1 Benchmarking
		16.8.2 A special note on classifier, ground truth labels and misrepresentation ratio
		16.8.3 Strength weakness and extensions
	16.9 Conclusion
	Acknowledgment
	Funding
	Conflict of interest
	References
	Appendix A Types of dataset used in the study
		A.1 TwitterA dataset
		A.2 WebKB4 dataset
		A.3 Disease dataset
		A.4 Reuters (R8) dataset
		A.5 SMS dataset
	Appendix B Labels used in different text data types
	Appendix C Receiver operating characteristic curves
		C1 Receiver operating characteristic curves for K2 protocol using five classifiers
		C2 Receiver operating characteristic curves for K4 protocol using five classifiers
		C3 Receiver operating characteristic curves for K5 protocol using five classifiers
		C4 Receiver operating characteristic curves for K10 protocol using five classifiers
		C5 Receiver operating characteristic curves for JK protocol using five classifiers
	Appendix D Area under the curve tables
	Appendix E Postive predictive value tables
	Appendix F Sensitivity tables
	Appendix G Specificity tables
	Appendix H List of abbreviations/symbols
Index