Data Science: New Issues, Challenges and Applications (Studies in Computational Intelligence, 869)

✍ Scribed by Gintautas Dzemyda (editor), Jolita Bernatavičienė (editor), Janusz Kacprzyk (editor)

• Publisher: Springer
• Year: 2020
• Tongue: English
• Leaves: 325
• Category: Library

✦ Synopsis

This book contains 16 chapters by researchers working in various fields of data science. They focus on theory and applications in language technologies, optimization, computational thinking, intelligent decision support systems, decomposition of signals, model-driven development methodologies, interoperability of enterprise applications, anomaly detection in financial markets, 3D virtual reality, monitoring of environmental data, convolutional neural networks, knowledge storage, data stream classification, and security in social networking. The respective papers highlight a wealth of issues in, and applications of, data science.
Modern technologies allow us to store and transfer large amounts of data quickly. They can be very diverse - images, numbers, streaming, related to human behavior and physiological parameters, etc. Whether the data is just raw numbers, crude images, or will help solve current problems and predict future developments, depends on whether we can effectively process and analyze it. Data science is evolving rapidly. However, it is still a very young field.
In particular, data science is concerned with visualizations, statistics, pattern recognition, neurocomputing, image analysis, machine learning, artificial intelligence, databases and data processing, data mining, big data analytics, and knowledge discovery in databases. It also has many interfaces with optimization, block chaining, cyber-social and cyber-physical systems, Internet of Things (IoT), social computing, high-performance computing, in-memory key-value stores, cloud computing, social computing, data feeds, overlay networks, cognitive computing, crowdsource analysis, log analysis, container-based virtualization, and lifetime value modeling. Again, all of these areas are highly interrelated. In addition, data science is now expanding to new fields of application: chemical engineering, biotechnology, building energy management, materials microscopy, geographic research, learning analytics, radiology, metal design, ecosystem homeostasis investigation, and many others.

✦ Table of Contents

Preface
Contents
Object Detection in Aerial Photos Using Neural Networks
1 Introduction
2 Distributed Data Processing System
3 Convolutional Neural Networks
4 Object Detection Methods by Convolutional Neural Network Review
5 Object Detection
6 Coordinate Rectangles Merging Algorithm
7 Adaptive Threshold Changing
8 Conclusion
References
Modelling and Control of Human Response to a Dynamic Virtual 3D Face
1 Introduction
2 Experiment Planning and Data Analysis
3 Input-Output Model Building
4 Predictor-Based Control with Constraints
5 Modelling Results
6 Conclusions
References
Knowledge-Based Transformation Algorithms of UML Dynamic Models Generation from Enterprise Model
1 Introduction
2 Business and IT Alignment Influence to Knowledge-Based IS Engineering
3 Importance of UML Elements Role Variations After Generating from EM
4 UML Models Transformation Algorithms from EM
5 Conclusions
References
An Approach for Networking of Wireless Sensors and Embedded Systems Applied for Monitoring of Environment Data
1 Introduction
2 Related Works on the Solutions of Smart WSN
3 The Architecture of Buoy’s System-Based on WSN for Monitoring of Sea Water Parameters
3.1 Embedded System Deployment in Sensor Node
3.2 Communication of the Network Sensors and Master Node
3.3 Background of Architecture of Buoy—Sensor Node
4 Possibilities to Apply Data Gathering Methods and Means for Situation Recognition
5 Working Principles of Smart Components Based on WSN System for Recognition of Situations
5.1 Data Acquiring and Combining
5.2 Detailing of Monitoring Parameters of Water Reservoir for Evaluation of Situations
5.3 Requirements for Intellectualization Stage in Environment Monitoring System
6 Experimental Results of Autonomous Uninterrupted Work of WSN System
7 Conclusions
References
Non-standard Distances in High Dimensional Raw Data Stream Classification
1 Introduction
2 The Research Problem
3 Current Knowledge
3.1 Stream Classification Algorithms
3.2 Stream Classification Performance Measures
3.3 Motivation
4 Classifier Model
4.1 Normalized Compression Distance
4.2 Lempel-Ziv Jaccard Distance
4.3 MOA Environment
5 Results of the Experiments
6 Conclusions
References
Data Analysis in Setting Action Plans of Telecom Operators
1 Introduction
2 Presentation of the Problem
3 Main Telecom Operators Portfolio
4 Methodological Aspects
5 Proposed Implementation Roadmap
6 Proposed/Recommended Architecture for Information System
7 Hub Services Based on Location Positions
8 Conclusion
References
Extending Model-Driven Development Process with Causal Modeling Approach
1 Introduction
2 Causality and Causal Modeling of Enterprise
3 Causal Model of Enterprise
4 Evaluation of MDA/MDD Process Gaps
5 The Causality Driven MDA/MDD Process
6 A Prior Knowledge and Causal Modeling of Enterprise
7 The Causal Knowledge Discovery Technique
8 Experimental Research
9 Conclusions
References
Discrete Competitive Facility Location by Ranking Candidate Locations
1 Introduction
2 Facility Location Problems
2.1 Customer Choice Rules
2.2 Discrete Competitive Facility Location Problems
3 Solution of Single-Objective DCFLP/EF
3.1 Ranking-Based Discrete Optimization Algorithm
3.2 Performance of RDOA
4 Solution of Multi-objective DCFLP/FE
4.1 Multi-objective Random Search with Ranking
4.2 Performance of MORSR
5 Conclusions
References
Investigating Feature Spaces for Isolated Word Recognition
1 Introduction
2 Feature Spaces
2.1 Fractal Dimension Features
2.2 Waveforms
2.3 Hartley Spectrum
2.4 Cochlegram
3 Convolutional Neural Network Architecture
4 Experimental Results
5 Conclusions
References
Developing Algorithmic Thinking Through Computational Making
1 Introduction
2 Programming and Problem-Based Learning
3 Computational Thinking and Making
4 How to Support Teachers in CT Implementation?
4.1 Computational Thinking Teaching Material
4.2 Computational Thinking Assessments
4.3 Computational Thinking Development Technologies
5 Discussion and Conclusions
References
Improving Objective Speech Quality Indicators in Noise Conditions
1 Introduction
2 Characteristics of the Lombard Speech
3 Objective Quality Level Indicators
4 Single-Ended Measurements
5 PESQ MOS Measurement Method
6 Recordings
6.1 Illustration of Lombard Effect in Speech Utterances
7 Recording Modifications
7.1 Frequency-Domain Filtering
7.2 Manipulation of Duration
7.3 Applying Modification to Pitch
7.4 Modification of the Vocal Tract
7.5 Manipulation of Formants
8 Estimation of PESQ MOS Values for the Analyzed Set of Recordings
8.1 Mixing the Recording with Pink Noise and Babble Speech
8.2 PESQ Estimation
9 Classification
10 Conclusions
References
Investigation of User Vulnerability in Social Networking Site
1 Introduction
2 Related Works
3 Data Set Description
4 Models for Data Analysis
5 Experimental Study
6 Conclusions
References
Zerocross Density Decomposition: A Novel Signal Decomposition Method
1 Introduction
2 Method
2.1 Derivation
2.2 Algorithm
2.3 Measures of Evaluation
3 Illustrative Example and Results
3.1 Modal Analysis
3.2 Comparison with EMD
3.3 Noise Robustness Analysis
4 Conclusion
References
DSS—A Class of Evolving Information Systems
1 Introduction
2 Context
2.1 Trends in the Evolution of Controlled Objects
2.2 Evolutions in Automation
2.3 Enabling Technologies
3 Computer-Aided Decision Support
3.1 Decisions and Participants to Decision-Making Processes
3.2 Decision Support Systems (DSS)
4 Enabling Information and Communication Technologies (I&CT) and Their Impact
4.1 Artificial Intelligence (AI) and Cognitive Systems
4.2 Big Data
4.3 Cloud and Mobile Computing
5 Conclusions
References
A Deep Knowledge-Based Evaluation of Enterprise Applications Interoperability
1 Introduction
2 Enterprise Applications Interoperability
3 Recent Findings in Interoperability Solutions
4 Interoperability Evaluation Using MDA, EA Approach
5 Experiment Environment Setups
6 Experiment Results
7 Further Work
8 Conclusions
References
Sentiment-Based Decision Making Model for Financial Markets
1 Introduction
2 Sentiment Based Market Forecasting
3 Development of Composite Indicator and Its Inclusion to Investment Decision Making Process
4 The Tool and Simulation of the Investment Process by Applying Composite Indicator
5 Conclusion
References