Data Democracy: At the Nexus of Artificial Intelligence, Software Development, and Knowledge Engineering

Cover
Data Democracy: At the Nexus of Artificial Intelligence,
Software Development, and
Knowledge Engineering
Copyright
Dedication
How To Use
Contributors
A note from the editors
Foreword
	References
Preface
Section I: The data republic
1 - Data democracy for you and me (bias, truth, and context)
	1. What is data democracy?
	2. Incompleteness and winning an election
	3. The story and the alternative story
	4. Nothing else matters
	References
2 - Data citizens: rights and responsibilities in a data republic
	1. Introduction
	2. A paradigm for discussing the cyclical nature of data–technology evolution
	3. Use cases explaining the black–red–white paradigm of data–technology evolution
		3.1 “Thank you visionaries”—black: 100 years of data science (1890–1990)
		3.2 “Progress to profit”—red: big data and open data in today's information economy
		3.3 “The past provides a lens for the future”—white: looking backward to see forward
	4. Preparing for a future data democratization
		4.1 The Datamocracy Framework helps envision the future
		4.2 Guiding principles within the framework
			4.2.1 Feature engineering should creatively use existing data to enhance models without introducing unintended bias. Ideally, inv ...
			4.2.2 Machine learning practice should protect the typical data citizen and not exploit their data literacy. Ideally, the data sc ...
			4.2.3 Data use should set ethical precedence in this revolution toward progress and harmony. Ideally, the data could be made full ...
	5. Practical actions toward good data citizenry
		5.1 Use data science archetypes
		5.2 Focus on the questions
		5.3 Collaborate within the process to build a new culture of data
		5.4 Label machine learning products for consumers
	6. Conclusion
	References
3 - The history and future prospects of open data and open source software
	1. Introduction to the history of open source
	2. Open source software's relationship to corporations
	3. Open source data science tools
	4. Open source and AI
	5. Revolutionizing business: avoiding data silos through open data
	6. Future prospects of open data and open source in the United States
	References
	Further reading
4 - Mind mapping in artificial intelligence for data democracy
	1. Information overload
		1.1 Introduction to information overload
		1.2 Causes of information overload
			1.2.1 Digital transformation
			1.2.2 Internet of things
			1.2.3 Social media
			1.2.4 Cybersecurity
			1.2.5 Internet web pages
			1.2.6 Emails
			1.2.7 Data openness
			1.2.8 Push systems
			1.2.9 Attention manipulation
			1.2.10 Spam email
			1.2.11 Massive open online courses (MOOCs)
		1.3 Consequences of information overload
			1.3.1 Anxiety, stress, and other pathologies
			1.3.2 Reduction in productivity
			1.3.3 Misinformation
			1.3.4 Poor decision-making
		1.4 Possible solutions
			1.4.1 Literature reviews
			1.4.2 Content management systems
			1.4.3 Open data portals (data democracy)
			1.4.4 Search engines
			1.4.5 Personal information agents
			1.4.6 Recommender systems
			1.4.7 Infographics
			1.4.8 Mind mapping
		1.5 Artificial intelligence in the reduction of information overload
	2. Mind mapping and other types of visualization
		2.1 Mind mapping in the visualization of open data
			2.1.1 Visualization of open data using mind mapping
			2.1.2 Visualization of big open data using mind mapping
		2.2 Visualization of content management systems using mind mapping
		2.3 Visualization of artificial intelligence results
			2.3.1 Types of applications of visualization in AI
			2.3.2 Exploratory data analysis as a first step in AI
			2.3.3 Software visualization and visual programming of AI applications
				2.3.3.1 An example: visualization of complex information in NLU applications
	3. Conclusions
	References
5 - Foundations of data imbalance and solutions for a data democracy
	1. Motivation and introduction
	2. Imbalanced data basics
		2.1 Degree of class imbalance
		2.2 Complexity of the concept
	3. Statistical assessment metrics
		3.1 Confusion matrix
		3.2 Precision and recall
		3.3 F-measure and G-measure
		3.4 Receiver operating characteristic curve and area under the curve
		3.5 Statistical assessment of the insurance dataset
	4. How to deal with imbalanced data
		4.1 Undersampling
			4.1.1 Random undersampling
			4.1.2 Tomek link
			4.1.3 Edited nearest neighbors
		4.2 Oversampling
			4.2.1 Random oversampling
			4.2.2 Synthetic minority oversampling technique
			4.2.3 Adaptive synthetic sampling
		4.3 Hybrid methods
	5. Other methods
	6. Conclusion
	References
Section II: Implications of a data democracy
6 - Data openness and democratization in healthcare: an evaluation of hospital ranking methods
	1. Introduction
	2. Healthcare within a data democracy—thesis
	3. Motivation
	4. Related works
	5. Hospitals' quality of service through open data
	6. Hospital ranking—existing systems
	7. Top ranked hospitals
	8. Proposed hospital ranking: experiment and results
	9. Conclusions and future work
	References
	Further reading
7 - Knowledge formulation in the health domain: a semiotics-powered approach to data analytics and democratization
	1. Introduction
	2. Conceptual foundations
		2.1 Semiotics
		2.2 Semantics: lexica and ontologies
		2.3 Syntagmatics: relationships and rules
		2.4 Syntactics: metadata
		2.5 Data interoperability and health information exchange
		2.6 Semiotics-based analytics
		2.7 Model-based analytics
			2.7.1 Information domain delineation: contexts and scope
			2.7.2 Data identification (exploration)
			2.7.3 Data preparation (data staging)
			2.7.4 Information model development
			2.7.5 Information presentation
			2.7.6 Heuristics-based analytics
	3. A semiotics-centered conceptual framework for data democratization
		3.1 Data democratization conceptual architecture
		3.2 Data democratization governance
	4. Conclusion
	References
8 - Landsat's past paves the way for data democratization in earth science
	1. Introduction
	2. Landsat overview
	3. Machine learning for satellite data
	4. Satellite images on the cloud
	5. Landsat data policy
	6. Conclusion
	References
9 - Data democracy for psychology: how do people use contextual data to solve problems and why is that important for AI systems?
	1. Introduction and motivation
	2. Understanding context
	3. Cognitive psychology and context
	4. The importance of understanding linguistic acquisitions in intelligence
	5. Context and data, how important?
	6. Neuroscience and contextual understanding
	7. Context and artificial intelligence
	8. Conclusion
	References
10. The application of artificial intelligence in software engineering: a review challenging conventional wisdom
	1. Introduction and motivation
	2. Applying AI to SE lifecycle phases
		2.1 Requirements engineering and planning
		2.2 Software design
		2.3 Software development and implementation (writing the code)
		2.4 Software testing (validation and verification)
		2.5 Software release and maintenance
	3. Summary of the review
	4. Insights, dilemmas, and the path forward
	References
	Further reading
Index
	A
	B
	C
	D
	E
	F
	G
	H
	I
	K
	L
	M
	N
	O
	P
	R
	S
	T
	U
	W
	X
	Z
Back Cover