Intelligent Systems and Applications: Proceedings of the 2021 Intelligent Systems Conference (IntelliSys) Volume 1: 294 (Lecture Notes in Networks and Systems)

Editor’s Preface
Contents
Late Fusion of Convolutional Neural Network with Wavelet-Based Ensemble Classifier for Acoustic Scene Classification
	1 Introduction
	2 Proposed Methodology
		2.1 Pre-processing and Feature Extraction
		2.2 Convolutional Neural Network
		2.3 Wavelet Scattering
		2.4 Ensemble Classifiers
		2.5 Fusion of CNN and Classifiers
	3 Results and Discussion
	4 Conclusion
	References
Deep Learning and Social Media for Managing Disaster: Survey
	1 Introduction
	2 Background and Related Works
		2.1 Recent Surveys
		2.2 Disaster
		2.3 Disaster Management
	3 Disaster Management Models
		3.1 Discussion About Disaster Management Models
	4 Social Media
	5 Retrieving Relevant Information from Social Media
		5.1 Classification Algorithms
		5.2 Machine Learning (ML)
		5.3 Deep Learning (DL)
	6 Conclusion and Future Works
	References
A Framework for Adaptive Mobile Ecological Momentary Assessments Using Reinforcement Learning
	1 Introduction
	2 Related Work
	3 Adaptive Mobile EMA
		3.1 An Unbiased Formulation for Mobile EMA
		3.2 Using Reinforcement Learning Framework for Adaptive Mobile EMA
	4 A Two-Level User State Model
	5 K-Routine Mining Algorithm
		5.1 Mining K-Routines
		5.2 Merging K-Routines
		5.3 Mapping K-Routines
	6 Designing Adaptive mEMA Method Using RL
		6.1 RL Algorithm
		6.2 State Space for Adaptive mEMA
		6.3 Action Space for Adaptive mEMA
		6.4 Reward Signal for Adaptive mEMA
		6.5 Experience Replay for Sample Efficiency Using Dyna-Q
		6.6 Performance Evaluation
	7 Experiments
		7.1 Data
		7.2 Baseline Methods
		7.3 Experimental Settings and Research Questions
	8 Results
		8.1 Comparisons Within RL Strategies
		8.2 Comparisons Between RL Strategies and Baseline Methods
		8.3 Performance by Data Segments
	9 Discussion
	10 Conclusion
	References
Reputation Analysis Based on Weakly-Supervised Bi-LSTM-Attention Network
	1 Introduction
	2 Related Work
		2.1 Machine Learning for Sentiment Analysis
		2.2 Deep Learning for Sentiment Analysis
	3 Weakly-Supervised Deep Embedding
		3.1 The Classic WDE Network Architecture
		3.2 Model Enhancement – WDE-BiLSTM-Attention
	4 Experiments
		4.1 Oversampling
		4.2 Baselines and Comparison
		4.3 Sentiment Classification
		4.4 Topic Mining Based on T-LDA
	5 Conclusion
		5.1 Deficiency and Future Work
	References
Multi-GPU-based Convolutional Neural Networks Training for Text Classification
	1 Introduction
	2 Related Work
		2.1 Data Parallelism Approaches
		2.2 Communications in Distributed Environment
	3 Distributed CNN for Text Categorization
		3.1 Motivation and Objective
		3.2 Baseline Model
		3.3 A Parallel CNN Algorithm for Text Classification
	4 Experimental Results
		4.1 Experimental Protocol
		4.2 Experiment 1: Sequential CNN Training
		4.3 Experiment 2: Sequential vs Distributed Training
		4.4 Experiment 3: Varying the Number of GPUs
	5 Conclusion
	References
Performance Analysis of Data-Driven Techniques for Solving Inverse Kinematics Problems
	1 Introduction
	2 Testing Model
	3 Forward Kinematics
	4 Analytical Approach
		4.1 Results of Analytical Techniques
		4.2 Limitation and Critical Analysis of Analytical Techniques
	5 Neural Network Approach
		5.1 Preparation of Data Set
		5.2 The Neural Network Architecture
	6 Experimental Results and Validation
	7 Conclusion and Future Work
	References
Machine Learning Based H2 Norm Minimization for Maglev Vibration Isolation Platform
	1 Introduction
	2 Vibration Isolator Modelling
		2.1 Derivation of the Balancing Levitation Force
		2.2 Isolator Dynamics
		2.3 State-Space Framework of Single Axis Levitation
		2.4 Four Pole Electromagnet Configuration
	3 Experimental Setup
		3.1 Hardware
		3.2 General Structure
	4 FSF Controller Syntheses
		4.1 H2 SF Controller Structure
	5 Deep Reinforcement Learning Algorithm
	6 Experimental Results
	7 Conclusions
	References
A Vision Based Deep Reinforcement Learning Algorithm for UAV Obstacle Avoidance
	1 Introduction
	2 Related Work
		2.1 Reinforcement Learning for Obstacle Avoidance
		2.2 Exploration
	3 Methodology: Towards Improving Exploration
		3.1 Training Setup
		3.2 Convergence Exploration
		3.3 Guidance Exploration
	4 Results and Discussion
	5 Conclusion
	References
Detecting and Fixing Nonidiomatic Snippets in Python Source Code with Deep Learning
	1 Introduction
	2 Related Work
	3 Method
		3.1 Formal Approach
		3.2 Neural Architectures
	4 Dataset Generation
		4.1 Template Generation
		4.2 Augmentation of Templates
	5 Evaluation
		5.1 Automated and Manual Evaluation
		5.2 Precision
		5.3 Recall
		5.4 Precision of Subsystems
	6 Conclusion
	A  Appendix
	References
BreakingBED: Breaking Binary and Efficient Deep Neural Networks by Adversarial Attacks
	1 Introduction
	2 Compression of Deep Neural Networks
		2.1 Knowledge Distillation
		2.2 Pruning
		2.3 Binarization
	3 Adversarial Attacks
		3.1 White-Box Attacks
		3.2 Black-Box Attacks
	4 Breaking Binary and Efficient DNNs
		4.1 CNN Compressed Variants
		4.2 Evaluation of Robustness
		4.3 Class Activation Mapping on Attacked CNNs
		4.4 Robustness Evaluation on ImageNet Dataset
		4.5 Discussion
	5 Conclusion
	References
Parallel Dilated CNN for Detecting and Classifying Defects in Surface Steel Strips in Real-Time
	1 Introduction
	2 Related Work
	3 Dataset and Augmentation
	4 Proposed DSTEELNet Architecture
	5 Experiments
		5.1 Experiment Metrics
		5.2 Setup
		5.3 Results
		5.4 Computational Time
	6 Conclusion
	References
Selective Information Control and Network Compression in Multi-layered Neural Networks
	1 Introduction
	2 Theory and Computational Methods
		2.1 Network Compression
		2.2 Controlling Selective Information
		2.3 Selective Information-Driven Learning
	3 Results and Discussion
		3.1 Experimental Outline
		3.2 Selective Information Control
		3.3 Generalization Performance
		3.4 Interpreting Compressed Weights
	4 Conclusion
	References
DAC–Deep Autoencoder-Based Clustering: A General Deep Learning Framework of Representation Learning
	1 Introduction
	2 Overview of Deep Autoencoder-Based Clustering
	3 Deep Autoencoder for Representation Learning
		3.1 Encoder
		3.2 Decoder
		3.3 Objective Function
	4 Experimental Results
		4.1 Data Set
		4.2 Measurement Metrics
		4.3 Experiment Setup
		4.4 Results on MNIST
		4.5 Results on Other Datasets
	5 Limitation
	6 Conclusion
	References
Enhancing LSTM Models with Self-attention and Stateful Training
	1 Introduction
	2 Background
		2.1 Feed-Forward Networks, Recurrent Neural Networks, Back Propagation Through Time
		2.2 Long Short-Term Memory and Truncated BPTT
		2.3 Self-attention
		2.4 Experimental Rationale
	3 Methodology
		3.1 Statefulness
		3.2 LSTM and Attention
	4 Data
		4.1 Data Characteristics
		4.2 Data Sets
	5 Models
		5.1 Architectures
		5.2 Hyperparameters
	6 Experiments and Results
		6.1 Model-to-Model and Model-to-Study Comparisons
	7 Discussion: Training Behavior
	8 Conclusions
	References
Domain Generalization Using Ensemble Learning
	1 Introduction
	2 Related Work
		2.1 Ensemble Learning
		2.2 Transfer Learning
		2.3 Domain Generalization
	3 Methods
		3.1 Data Preparation
		3.2 Experiments
		3.3 Hyperparameter Tuning
	4 Results
	5 Conclusion
	References
Research on Text Classification Modeling Strategy Based on Pre-trained Language Model
	1 Introduction
	2 Related Work
	3 Model Architecture
		3.1 Model Input
		3.2 Transformer
		3.3 Capsule Networks
		3.4 Model Framework
	4 Experiment Design and Analysis
		4.1 Experiment Corpus
		4.2 Evaluation Metrics
		4.3 Experimental Setup
		4.4 Comparative Experiment
		4.5 Ablation Experiment
		4.6 Experiment Analysis
	5 Conclusion and Future Work
	References
Discovering Nonlinear Dynamics Through Scientific Machine Learning
	1 Introduction
	2 Scientific Machine Learning Models
		2.1 Physics-Informed Neural Networks
		2.2 Universal Differential Equations
		2.3 Hamiltonian Neural Networks
		2.4 Neural Ordinary Differential Equations (Neural ODE)
	3 Physical Experiments
		3.1 Quadruple Spring Mass System
		3.2 Pendulum
		3.3 Simulated Pendulum
		3.4 Simulation of Wind Forced Pendulum
		3.5 Physical Experimental Pendulum
	4 Learning the Nonlinear Dynamics with Scientific Machine Learning
		4.1 What Do These SciML Models Learn?
		4.2 Can SciML Predict the Future?
		4.3 Can HNN Solve Complex Dynamic Problems?
	5 Conclusion
	References
Tensor Data Scattering and the Impossibility of Slicing Theorem
	1 Introduction
	2 Tensor
	3 Pick and Slice
	4 Tensor Variator and Its Provision Tensor
	5 Nondeterministic of Applying Variator
	6 Scattering
		6.1 Scatter APIs in Two Popular Deep Learning Frameworks
		6.2 Defining Scattering
		6.3 Sliceable Scattering
	7 Sparse Tensor with X-Sparse Representation
		7.1 The Limitations in Current Scattering APIs
		7.2 X-Sparse Tensor
		7.3 Counting Sparsity and Analyzing Performance
		7.4 Mocking Current Scattering APIs
	8 Conclusion
	References
Scope and Sense of Explainability for AI-Systems
	1 Introduction
	2 Superhuman Abilities of AI
	3 Forms of Explainability
	4 Complex Dynamical Systems
	5 Stability and Chaos
	6 Nonclassical Approaches, Training of Attractors
	7 Causality of Results?
	8 Conclusions
	References
Use Case Prediction Using Deep Learning
	1 Introduction
	2 Related Work
		2.1 Parts of Speech
		2.2 Deep Learning
	3 Proposed Approach
	4 Experiments and Results
		4.1 Datasets Description
		4.2 Metrics
	5 Conclusions
	References
VAMDLE: Visitor and Asset Management Using Deep Learning and ElasticSearch
	1 Introduction
	2 Background
		2.1 Visitor Management and Asset Management
		2.2 CNN and MobileNet
		2.3 Deep Transfer Learning
		2.4 ElasticSearch
		2.5 High Performance Computing
	3 Design
		3.1 Architectural Design
		3.2 UI and UX Design
	4 Implementation and Evaluation
		4.1 Dataset
		4.2 Image Pre-Processing and Data Augmentation
		4.3 Deep Transfer Learning Model
		4.4 Android Application
		4.5 Evaluation of the Proposed System
	5 Conclusion
	References
Wind Speed Time Series Prediction with Deep Learning and Data Augmentation
	1 Introduction
	2 Related Work
	3 Background
		3.1 Recurrent Neural Networks
		3.2 Data Augmentation
	4 Methodology
		4.1 Time Series Selection
		4.2 Time Series Imputation
		4.3 Data Augmentation
		4.4 Scaling
		4.5 Modelling
		4.6 Evaluation
	5 Results
	6 Discussion
	7 Conclusion and Future Work
	References
Evaluation for Angular Distortion of Welding Plate
	1 Introduction
	2 Equipment and CNN
	3 Experiment
	4 Validation of CNN
	5 Conclusions
	References
A Framework for Testing and Evaluation of Operational Performance of Multi-UAV Systems
	1 Introduction
	2 Literature Review
	3 Problem Description
		3.1 Terminology
		3.2 Problem Statement
	4 Proposed Framework
		4.1 Overview of the Proposed Framework
		4.2 Modes of Operation
		4.3 Scenarios
		4.4 Perception Inference Engine (PIE)
		4.5 True Scenario
		4.6 Evaluator
	5 Synthetic Data Generation
	6 Hardware Implementation
	7 Experiments and Discussion
		7.1 Data Collection and Model Selection
		7.2 Deployment of PIE
		7.3 Deployment of PIE in Hardware
	8 Conclusion and Future Work
	References
Addressing Consumer Demands: A Manufacturing Collaboration Process Using Blockchain for Knowledge Representation
	1 Introduction
	2 Background
		2.1 Blockchain
		2.2 Related Work
	3 Proposed Solution
		3.1 Collaborative Network of Entities
		3.2 Reasoning and Interaction
		3.3 Knowledge Representation
	4 Conclusion and Future Work
	References
Cellular Formation Maintenance and Collision Avoidance Using Centroid-Based Point Set Registration in a Swarm of Drones
	1 Introduction
	2 Proposed Approach
		2.1 Obstacle Detection
		2.2 Collision Avoidance
		2.3 Re-formation
	3 Simulation and Results
	4 Conclusion
	References
The Simulation with New Opinion Dynamics Using Five Adopter Categories
	1 Introduction
	2 Theory
		2.1 Opinion Dynamics
		2.2 Diffusion of Innovations
	3 Modeling
	4 Simulations
		4.1 Manipulating the Initial Distribution of Opinions
		4.2 Manipulation of Confidence Coefficient 2D2ij
		4.3 Manipulating Mass Media Effects
		4.4 Manipulating Network Connection Probabilities
	5 Discussion
	6 Conclusion
	References
Intrinsic Rewards for Reinforcement Learning Within Complex 2D Environments
	1 Introduction
	2 Related Work
	3 Data
	4 Methods
		4.1 Reinforcement Learning Background
		4.2 Model Policies
		4.3 Model Inputs
		4.4 Model Architecture
	5 Metrics
		5.1 Quantitative Agent Comparison
		5.2 Qualitative Comparison
	6 Results and Discussion
		6.1 Experiment Setup
		6.2 Quantitative Results
		6.3 Qualitative Results
	7 Conclusion and Future Work
	References
Analysis of Divided Society at the Standpoint of In-Group and Out-Group Using Opinion Dynamics
	1 Introduction
	2 Trust-Distrust Model
		2.1 Theory of Trust-Distrust Model
		2.2 Two-Agents Calculation
		2.3 Calculation for 300 Persons
	3 Model Setting for Social Simulation
	4 Results
		4.1 Calculation for the First Model
		4.2 Calculation for the Second Model
	5 Discussion
	6 Conclusion
	References
Simulation of Intragroup Alignment Using a New Model of Opinion Dynamics
	1 Introduction
	2 Theory
	3 Simulation Model of Intragroup Alignment
	4 Results
		4.1 Trust to a Candidate from Voters
		4.2 Sub-leaders
	5 Discussion
	6 Conclusion
	References
Random Forest Classification with MapReduce in Holonic Multiagent Systems
	1 Introduction
	2 Related Work
	3 Background
		3.1 Multiagent Learning
		3.2 Holonic Multiagent Systems
		3.3 Decision Trees and Random Forests
	4 Materials and Methods
		4.1 Y-Combinator
		4.2 Decision Tree Classification
		4.3 Random Forest Classification
		4.4 System Components
	5 Results
	6 Discussion
	7 Conclusion
	References
Monitoring Goal Driven Autonomy Agent\'s Expectations Generated from Durative Effects
	1 Introduction
	2 Related Work
	3 Preliminaries
	4 Two Basic Operations
	5 Informed Expectations with Durative Effects
	6 Regression Expectations
	7 Goldilocks Expectations
	8 Property of Regression
	9 Empirical Evaluation
	10 Conclusions
	References
Sublinear Regret with Barzilai-Borwein Step Sizes
	1 Introduction
		1.1 Contributions
	2 Problem Formulation
		2.1 Algorithms for Online Optimization Problem
		2.2 Quasi-Newton Methods
	3 The Barzilai-Borwein Quasi-Newton Method
	4 Regret Bounds
	5 Conclusions
	References
Fluid Dynamics of a Pandemic in a Spatial Social Network: A Reflective Measure of the Spreading
	1 Introduction
	2 Literature Review
	3 Methodology
		3.1 Preliminaries
		3.2 Argumentative Game Theoretical Approach in Social Network
	4 Illustrative Example
	5 Conclusion
	References
Affective Story-Morphing: Manipulating Shelley’s Frankenstein under Program Control using Emotionally Intelligent Agents
	1 Introduction and Motivation
	2 Story Morphing in the Affective Reasoner
	3 How Development Proceeds
	4 Additional Aspects of the Affective Reasoner
		4.1 Humor
		4.2 Case-Based Reasoning
		4.3 Applications
		4.4 Users as Agents
	5 Morphing the Monster
		5.1 A Paraphrase of the Original Narrative—Snippet One
		5.2 Story Morph Snippet Two
		5.3 Story Morph Snippet Three
		5.4 Story Morph Four
		5.5 Story Morph Five
		5.6 Story-Morph Snippet Six
		5.7 Story-Morph Snippet Seven
		5.8 Story-Morph Snippet Eight
		5.9 Story-Morph Snippet Nine
		5.10 Story-Morph Snippet Ten
		5.11 Some Finer-Grained Variations
	6 Implementation
	7 Conclusions and Summary
	References
Dynamic Strategies and Opponent Hands Estimation for Reinforcement Learning in Gin Rummy Game
	1 Introduction
	2 Gin Rummy Rules
	3 Related Work
	4 Static Strategies
		4.1 Discard Strategy
		4.2 Draw Strategy
		4.3 Opponent Hand’s Estimation Strategy
		4.4 Knock Strategy
	5 Dynamic Strategies
		5.1 Dynamic Knock Strategy
		5.2 Dynamic Draw/Discard Strategy
	6 Experimental Results
	7 Conclusion and Future Work
	References
Wireless Sensor Network Smart Environment for Precision Agriculture: An Agent-Based Architecture
	1 Introduction
		1.1 Agriculture Evolution
		1.2 Agriculture 4.0 Conceptual Model
	2 Enabling Technologies for Precision Agriculture
	3 Multi-agent Architecture for Precision Agriculture
		3.1 Modeling the Precision Agriculture Smart Environment
	4 Agent-Based PA Implementation Directives
		4.1 Hardware Specifications
		4.2 Software Specifications
		4.3 Experiment Environment Setting
	5 Conclusions
	References
Autonomy Reconsidered: Towards Developing Multi-agent Systems
	1 Introduction
	2 Related Literature
	3 Behavior, Success, and Autonomy
		3.1 Absolute Autonomy: Behavior, Success, Fulfillment
		3.2 Relative Autonomy: Levels, Asymmetries, Deficiencies
	4 Multi-agent Systems
		4.1 Group Potential: Synergy and Interference
		4.2 Augmentation and Diminishment
	5 Summary
	References
A Real-Time Intelligent Intra-vehicular Temperature Control Framework
	1 Introduction
	2 Background
		2.1 Object Detection
		2.2 Convolutional Neural Network (CNN)
		2.3 Controller Area Network (CAN) Bus
		2.4 Message Queuing Telemetry Transport (MQTT)
	3 Proposed System
		3.1 Microcontroller M1
		3.2 Microcontroller M2
		3.3 Cloud Communication
	4 Results
	5 Conclusion
	References
Intelligent Control of a Semi-autonomous Assistive Vehicle
	1 Introduction
	2 The Wheelchair
	3 Control
		3.1 Modelling
		3.2 Controller Design
		3.3 Path-Following
	4 Conclusions and Future Work
	References
One Shot Learning Approach to Identify Drivers
	1 Introduction
	2 The New Approach
	3 Discussion and Results
	4 Conclusions and Future Work
	References
Facial Recognition Software for Identification of Powered Wheelchair Users
	1 Introduction
		1.1 Facial Recognition Systems
		1.2 API
		1.3 Software Libraries
	2 Facial Recognition System
	3 Results
	4 Discussion and Conclusions
	References
Intelligent User Interface to Control a Powered Wheelchair Using Infrared Sensors
	1 Introduction
	2 The New System
	3 Testing
	4 Conclusions and Future Work
	References
A Classification Based Ensemble Pruning Framework with Multi-metric Consideration
	1 Introduction
	2 Related Work
	3 The Proposed Framework
		3.1 Problem Statement
		3.2 Overview of the Proposed Framework
		3.3 Ensemble Pruning with Classification Based Optimization
		3.4 Multi-Metric Consideration and Its Optimization
	4 Empirical Results
		4.1 Compared Methods
		4.2 Experiments on Benchmark Datasets
	5 Application to Fraud Detection Tasks
	6 Conclusion
	References
Customs Risk Assessment Based on Unsupervised Anomaly Detection Using Autoencoders
	1 Introduction
	2 Data and Methodology
		2.1 Autoencoder
		2.2 Variational Autoencoder
	3 Results
		3.1 Autoencoder on ENS Data
		3.2 Autoencoder on Synthetic Data
	4 Future Work and Conclusions
		4.1 Conclusions
		4.2 Future Work
	References
Best Next Preference Prediction Based on LSTM and Multi-level Interactions
	1 Introduction
	2 LSTM Based Recommendations
	3 DeepCBPP for Next Preference Predictions
	4 LSTM Model Architectures
	5 Performance Evaluation
	6 Conclusions and Future Work
	References
Achieving Trust in Future Human Interactions with Omnipresent AI: Some Postulates
	1 Introduction
	2 Defining Omnipresent AI
		2.1 What is an Omnipresent AI?
		2.2 Interaction Models for Omnipresent AI
		2.3 Interaction and Trust
		2.4 The Aspirations of Omnipresent AI
	3 Towards Postulates of Human-Omnipresent AI Interaction
		3.1 Proposing a Natural Communication Method
		3.2 Presence and Personality
		3.3 Proprioception and the Understanding of Context
	4 The Postulates of a Trustworthy Human-Omnipresent AI Interaction
	5 Speculative Application of AI-Human Interaction in an Autonomous Vehicle
	6 Conclusion
	References
A Decentralized Explanatory System for Intelligent Cyber-Physical Systems
	1 Introduction
	2 Background and Related Works
	3 Smart Home Scenarios
	4 Decentralizing Explanatory Reasoning
		4.1 Solution Overview
		4.2 A Decentralized Knowledge
		4.3 A Unifying Algorithm: D-CAS
		4.4 Generating an Explanation
	5 Implementation and Results
		5.1 The Window Blinds
		5.2 The Ventilation Monitoring System
	6 Discussions and Future Works
	7 Conclusion
	References
Construction Control Organization with Use of Computer and Information Technologies in Context of Sustainable Development Providing
	1 Introduction
	2 Materials and Methods
	3 Results
	4 Discussion
	5 Conclusion
	References
Computational Rational Engineering and Development: Synergies and Opportunities
	1 Introduction and Motivation
	2 Recent and Past Perspectives on Computer Systems for Automation of Engineering and Development
	3 Computational Rationality in Engineering Development
		3.1 Domain Characteristics: Problem-Solving and Decision-Making in the Context of Industrial Design, Engineering, and Development
		3.2 Interdisciplinary Opportunities and Synergies
	4 Discussion and Perspectives
		4.1 Mind the Gap: Intelligent Systems for Design, Engineering, and Development
		4.2 Open Challenges and Prospective Research Directions
	5 Concluding Remarks
	References
QPSetter: An Artificial Intelligence-Based Web Enabled, Personalized Service Application for Educators
	1 Introduction
	2 Motivation
	3 Related Work
	4 System Architecture
		4.1 Scraper Module
		4.2 Educational Artificial Intelligence (EAI) Module
		4.3 The Database
		4.4 The Q-Adder
		4.5 The User Interface
	5 Conclusions
	References
Is It Possible to Recognize a Philosophical Zombie and How to Do It
	1 Introduction
	2 Why is It Necessary to Think About Philosophical Zombies
	3 How to Recognize a Philosophical Zombie
	4 Could Artificial Intelligent Systems Get Qualia
	5 Conclusion
	References
Dynamic Analysis of Bitcoin Fluctuations by Means of a Fractal Predictor
	1 Introduction
	2 Related Work
	3 Theoretical Framework
		3.1 Bitcoin
		3.2 Fractal Theory
	4 Proposal
		4.1 Theoretical Definition
		4.2 Methodology
	5 Experimental Results
		5.1 Results
		5.2 Discussion
	6 Conclusions
	References
Are Human Drivers a Liability or an Asset?
	1 Introduction
	2 Do Near Misses Suggest that Collisions May Occur?
		2.1 Near Misses
		2.2 Bowties
	3 Method and Testing
	4 Results
	5 Discussion and Conclusions
	References
Negative Emotions Induced by Non-verbal Video Clips
	1 Introduction
	2 Experiment
	3 Method
		3.1 Participants
		3.2 Materials
		3.3 Procedure
	4 Results
	5 Conclusion
	References
Automatic Recognition of Key Modulations in Symbolic Musical Pieces Using Information Theory
	1 Introduction
	2 Related Work
	3 Key and Modulation
		3.1 Harmonic Analysis
	4 Information Theory
	5 Application and Analysis
		5.1 Results
	6 Discussion and Conclusions
	Appendix A
	References
Increasing Robustness for Machine Learning Services in Challenging Environments: Limited Resources and No Label Feedback
	1 Introduction
	2 Foundations
		2.1 Machine Learning
		2.2 Concept Drift
		2.3 Outlier Detection
	3 Problem Definition and Requirements
	4 Design Options
		4.1 Step 1: Data Validity
		4.2 Step 2: Model Robustness
	5 Evaluation
		5.1 Evaluation of Data Validity (Step 1)
		5.2 Evaluation of Model Robustness (Step 2)
		5.3 Evaluation of Overall Prediction Method
	6 Conclusion
	References
Development Support for Intelligent Systems: Test, Evaluation, and Analysis of Microservices
	1 Introduction: Microservices in General
	2 Challenges with Testing Microservices
	3 Analysis of Microservices
		3.1 Collecting Key Figures
		3.2 Approach
	4 Test Concepts
		4.1 Test Concept of Eberhard Wolff
		4.2 Test Concept of Sam Newman
		4.3 Test Concept of Google
		4.4 Test Concept of Netflix
	5 Tools for Analysis and Testing
		5.1 Tools for Isolated Testing
		5.2 Analysis
		5.3 Netflix
	6 Conclusion
	References
An Analysis with Dynamics Between Human Motivation and Messaging on Social Networking Services
	1 Introduction
		1.1 Back Ground and Purpose
		1.2 Structure of This Paper
	2 Issues of Previous Studies and Our Approach
		2.1 Issues of Previous Studies
		2.2 Our Approach
	3 The Mechanism of Our Messaging Model
		3.1 Event Driven Based
		3.2 Messaging and Motivation
		3.3 Messaging Strategy
	4 Simulations
		4.1 Initial Conditions
		4.2 Validation Test
		4.3 Increments of Modification, M q( t )
		4.4 Variable Reliability Factor, M r ( t ) and Trust Level, M tr( t )
		4.5 Personalization (Level 1): Random Variation of M q( t ), M r( t ) and M tr( t )
		4.6 Personalization (Level 2): Random Variation of Thresholds for Motivation of Each Node
		4.7 Personalization (Level 3): Random Variation Both M q( t ) and Thresholds
		4.8 Personalization (Level 4): Random Variation Both M q( t ) and Thresholds with P/N Opinions
	5 Discussions and Future Works
	6 Conclusions
	References
Author Index