Advanced visual interfaces : supporting artificial intelligence and big data applications : AVI 2020 Workshops, AVI-BDA and ITAVIS, Ischia, Italy, June 9, 2020 and September 29, 2020, Revised Selected Papers

This book constitutes the thoroughly refereed post-workshop proceedings of the AVI 2020 Workshop on Road Mapping Infrastructures for Artificial Intelligence Supporting Advanced Visual Big Data Analysis, AVI-BDA 2020, held in Ischia, Italy, in June 2020, and the Second Italian Workshop on Visualization and Visual Analytics, held in Ischia, Italy, in September 2020. The 14 regular papers in this volume present topics such as big data collection, management and curation; big data analytics; big data interaction and perception; big data insight and effectuation; configuration and management of big data storage and compute infrastructures, services, and tools; advanced visual interaction in big data applications; user empowerment and meta design in big data applications; prediction and automation of big data analysis workflows; as well as data visualization; information visualization; visual analytics; infographics; and design. Preface of AVI-BDA Workshop Preface of ITAVIS Workshop Organization of AVI-BDA Workshop Organization of ITAVIS Workshop Contents AI2VIS4BigData: A Reference Model for AI-Based Big Data Analysis and Visualization 1 Introduction and Motivation 2 State of the Art 2.1 Reference Models 2.2 IVIS4BigData Reference Model 2.3 AIGO's AI System Lifecycle Reference Model 3 Conceptual Modeling 3.1 AI Model Types 3.2 AI Data 3.3 AI User Stereotypes 3.4 AI2VIS4BigData Reference Model 4 Remaining Challenges and Outlook 5 Conclusion References A Visual Analytics Technique to Compare the Performance of Predictive Models 1 Introduction 2 Background and Related Work 3 The Three Visualizations 3.1 Predictive Models Comparison Matrix 3.2 Two Models Pie-Chart Matrix 3.3 Instance Level Explanation (ILE) Visualization 4 Conclusion References Affective Analytics and Visualization for Ensemble Event-Driven Stock Market Forecasting 1 Introduction 2 Background 2.1 Stock Marketing 2.2 Introduction to Artificial Intelligence 2.3 Artificial Intelligence in Stock Market Prediction 2.4 AI Stock Market Prediction with Financial Indicators 2.5 AI Stock Market Prediction with Textual Data 2.6 AI Stock Market Prediction with Twitter Data Analysis 2.7 AI Stock Market Prediction with News Data Analysis 2.8 Big Data Visualisation of Streaming Data 2.9 Contribution 3 Big Data Pre-processing and Visualization of Tweets, Business NEWS and Financial Indicators 3.1 Data Pre-processing 3.2 Feature Extraction Techniques 3.3 Formulation of the Feature Engineered Data Set and Label 3.4 Twitter Data Accumulation and Visualization 3.5 Business NEWS Data Accumulation and Visualization 4 Architecture for Stock Market Prediction 4.1 Hybrid Architecture Based on Best Model Selection Strategy 4.2 Hybrid Architecture Based on Shallow Transfer Learning Model 4.3 Hybrid Architecture Based on Engineered Feature Dataset 5 Evaluation Metrics 6 Results 6.1 Hybrid Architecture Based on Best Model Selection Strategy 6.2 Hybrid Architecture Based on Shallow Transfer Learning Model 6.3 Hybrid Architecture Based on Engineered Feature Dataset 7 Validation of the AI2VIS4BigData Reference Model 8 Conclusion and Future Work References Understanding the Role of (Advanced) Machine Learning in Metagenomic Workflows 1 Introduction 1.1 Importance of Microbiome Analysis 1.2 Machine Learning Trends 2 Genetics and Metagenomics 2.1 Microorganisms 2.2 Differences Between DNA and RNA 2.3 Genes and Proteins 2.4 Genomic vs Metagenomic Studies 2.5 Phylogenetic Trees and Taxonomies 2.6 Structure of Metagenomic Studies 3 Roles of Machine Learning in Metagenomics 3.1 Preparation 3.2 Aggregation 3.3 Annotation 3.4 Analysis 3.5 Visualization 4 Metagenomic Processing Pipelines 4.1 Galaxy 4.2 MG-RAST 4.3 MGnify (EBI Metagenomics) 4.4 Qiime 4.5 MetaPlat 5 Example: Rumen Microbiome Analysis with MetaPlat 5.1 Visualization of Gene Dependencies Using Shotgun Sequencing 5.2 Comparison of Rumen and Feces Microbiomes Using Amplicon Sequencing 6 Challenges for Machine Learning in Metagenomics 6.1 Choosing the Right Model 6.2 Accessibility 6.3 Explainability 6.4 Reproducibility 6.5 Biological Diversity 6.6 Big Data 6.7 High Dimensionality and Low Number of Samples 7 Applying the AI2VIS4BigData Reference Model 8 Conclusion References Intelligent Advanced User Interfaces for Monitoring Mental Health Wellbeing 1 Introduction 2 Questionnaires Assessing Depressive States 3 Behavioral and Physiological Parameters Indicating Depressive States 4 The MENHIR (Mental Health Monitoring Through Interactive Conversations) Project 5 Automatic Detection of Depressive States Exploiting Speech 6 Conclusions References Towards Explainable Artificial Intelligence and Explanation User Interfaces to Open the ‘Black Box’ of Automated ECG Interpretation 1 Artificial Intelligence 2 Automated ECG Interpretation-Challenges and Opportunities 3 Artificial Intelligence Transparency 4 Explainable Artificial Intelligence in ECG Interpretation 5 Model Validation 6 Conclusion and Discussion References Recognition and Visualization of Facial Expression and Emotion in Healthcare 1 Introduction 2 State of the Art 3 Methodology 3.1 Video Recognition Process 3.2 Emotion Visualization Process 4 Implementation Methods 4.1 Real-Time Video Streaming 4.2 Real-Time Video Analysis 4.3 Detection and Recognition of Emotion 4.4 Emotion Visualization 4.5 AI2VIS4BigData Model 5 Conclusion and Future Work References Machine Learning in Healthcare: Breast Cancer and Diabetes Cases 1 Introduction 1.1 Image Processing 1.2 Machine Learning (ML) 2 Machine Learning Applications 2.1 ML in Breast Cancer Research 2.2 ML in Diabetes Research 3 Other Potential Areas 4 Summary 5 Validation of the AI2VIS4BigData Reference Model 6 Conclusion References AI2VIS4BigData: Qualitative Evaluation of an AI-Based Big Data Analysis and Visualization Reference Model 1 Introduction and Motivation 2 AI2VIS4BigData Reference Model 2.1 IVIS4BigData 2.2 AIGO'S AI System Lifecycle 2.3 AI Model Types, AI Data, AI User Stereotypes 2.4 AI2VIS4BigData Reference Model and Subjective Decisions in Model Derivation 3 Expert Round Table 3.1 Method 3.2 Results 4 Workshop Participant Survey 4.1 Method 4.2 Results 5 Validation and Limitations 6 Conclusion and Outlook References Progressive Visualization of Epidemiological Models for COVID-19 Visual Analysis 1 Introduction 2 Related Work 3 Epidemiological Models Background 4 Progressive Visualization Technique 5 Evaluation 6 Conclusions and Future Works References An Experience on Cooperative Development of Interactive Visualizations for the Analysis of Urban Data 1 Introduction and Motivation 2 Developing Interactive Visualizations According to CMD 3 Dataset 4 The Implemented Interactive Visualizations 5 Conclusions References Explaining AI Through Critical Reflection Artifacts 1 Introduction 2 Reaching the General Audience 2.1 Critical Reflection Artifacts 3 Discussion References Information Visualization and Visual Analytics at IVU Lab 1 Introduction 2 Visual Analytics 3 Time Series 4 Other Topics and Research Directions References Visual Analytics for Financial Crime Detection at the University of Perugia 1 Research Context and Motivation 2 Main Research Activities 3 Future Research Directions References Author Index