Preface
Organization
Keynotes
Machine Learning Algorithms for Brain-Machine Interfaces
Digital Twins of the Ocean
On the Use (and Misuse) of Differential Privacy in Machine Learning
Learning on Graphs
Contents – Part II
Contents – Part I
Artifical Intelligence, Generation and Creativity
Erato: Automatizing Poetry Evaluation
1 Introduction
2 Related Work
3 What Characterizes a Good Poem?
4 Erato: A Framework for Poetry Evaluation
4.1 General Structure
4.2 Available Modules
4.3 Extending Erato for Specific Purposes
5 Case Study: Human and Machine Poetry
5.1 Computer-Generated Poetry
5.2 Human-Written Poetry
5.3 Analysis
6 Conclusion and Future Directions
References
A Path to Generative Artificial Selves
1 Introduction
2 Creativity as Restructuring a Manifold
3 Selfhood
4 Reflexively Autocatalytic Foodset-Derived Networks (RAFs)
5 RAF Models of Emergent Cognition
6 Discussion
6.1 Related Research
6.2 Future Work: Experimental Testing and Validation
7 Conclusions
References
Human+Non-human Creative Identities. Symbiotic Synthesis in Industrial Design Creative Processes
1 Technologies and Creative Processes
2 AI-Tools and Design Practice
3 An Evolving Symbiotic Creative Ecology
References
AIGenC: AI Generalisation via Creativity
1 Introduction
2 Functional Creativity, Concept Space and Affordances
3 A Framework for Concept Transfer and Functional Creativity
3.1 Deep Reinforcement Learning
3.2 Concept Processing Component
3.3 Reflective Reasoning Component
3.4 Blending Component
4 Discussion
References
Creativity, Intentions, and Self-Narratives: Can AI Really Be Creative?
1 Introduction
2 Creativity
3 Process Creativity and Intentions
4 Intentions and AI
5 Creativity in the Prompts
6 Self-Narratives
7 Conclusion
References
Evolving Urban Landscapes
1 Introduction
2 Related Work
3 Approach
3.1 Visual Grammar
3.2 Lexicon
3.3 Rules
3.4 Implementation
4 Assessing Creativity
4.1 Definition of Creativity
4.2 Creativity in the Context of Our System
4.3 Questionnaire
4.4 Results Analysis
5 Final Remarks
References
Emotion4MIDI: A Lyrics-Based Emotion-Labeled Symbolic Music Dataset
1 Introduction
2 Related Work
2.1 Text Emotion Classification
2.2 Emotion-Labeled Symbolic Music Datasets
3 Methodology
3.1 Model
3.2 Training
3.3 Inference
4 Results
4.1 Emotion Classification on the GoEmotions Dataset
4.2 Labeled MIDI Dataset
5 Conclusion and Future Work
References
Artificial Intelligence and Law
On the Assessment of Deep Learning Models for Named Entity Recognition of Brazilian Legal Documents
1 Introduction
2 Related Works
3 Method
3.1 Hyperparameters Tuning
4 Results and Discussion
4.1 Experimental Setup
4.2 Hyperparameters Evaluation for LeNER-Br
4.3 Hyperparameter Evaluation for PL-Corpus
4.4 Comparison and Discussion
5 Conclusion and Future Works
References
Anonymisation of Judicial Rulings for Legal Analytics Purposes: Ethics, Law, and Compliance
1 Introduction
2 Advancements and Benefits of Legal Analytics
2.1 A Case Study: The Legal Analytics for Italian Law (LAILA) Project
3 Anonymisation of Judicial Rulings for Legal Analytics Purposes
3.1 The Legal Framework
3.2 Anonymisation Measures Taken by Judicial Offices
3.3 Anonymisation of Court Decisions in the Context of the LAILA Project
4 Conclusions: At the Crossroads of Law and Ethics
References
LeSSE—A Semantic Search Engine Applied to Portuguese Consumer Law
1 Introduction
2 Related Work
3 Legal Semantic Search Engine
3.1 Datasets
3.2 System Overview
3.3 Semantic Pipeline
3.4 Lexical Pipeline
3.5 Results Selection and Presentation
3.6 Model Training and Optimization
4 Performance of LeSSE in Consumer Law
5 Performance of LeSSE in the Absence of Manual Annotations
6 Conclusions and Future Work
References
Does ChatGPT Pass the Brazilian Bar Exam?
1 Introduction
2 GPT in Law
3 Experiment Design
4 Results and Discussion
5 Conclusions and Further Work
References
A Semantic Search System for the Supremo Tribunal de Justiça
1 Introduction
2 Related Work
3 Data
4 Semantic Search System Architecture
5 Legal Language Model
5.1 Domain Adaptation
5.2 Semantic Textual Similarity
5.3 Natural Language Inference
5.4 Multilingual Knowledge Distillation
5.5 Metadata Knowledge Distillation
6 Evaluation
6.1 Language Model Evaluation
6.2 Search System Evaluation
7 Conclusion
References
Artificial Intelligence in Power and Energy Systems
The AI Act Meets General Purpose AI: The Good, The Bad and The Uncertain
1 AI Act: The Regulation of GPAI
1.1 Context
1.2 Definition: Dimensions of Generality
1.3 Regulation: Challenges and Risks
2 AIA Draft
2.1 AI Requirements and Obligations
2.2 Key Elements: Value Chain and Cooperation
2.3 Exemptions
3 Conclusions
References
Rule-Based System for Intelligent Energy Management in Buildings
1 Introduction
2 Proposed Model
2.1 Power Consumption State Ruleset
2.2 Air Conditioning System Ruleset
2.3 Brightness Ruleset
3 Rulesets Evaluation
3.1 Consumption State Ruleset
3.2 Brightness Ruleset Case Study
3.3 Air Conditioning System Ruleset Case Study
4 Conclusions
References
Production Scheduling for Total Energy Cost and Machine Longevity Optimization Through a Genetic Algorithm
1 Introduction
2 Related Works
3 Proposed Methodology
4 Genetic Algorithm Implementation
4.1 Initial Population Procedure
4.2 Crossover Procedure
4.3 Mutation Procedure
4.4 Selection Procedure
5 Case Study
6 Results and Discussion
7 Conclusions
References
A Novel Federated Learning Approach to Enable Distributed and Collaborative Genetic Programming
1 Introduction
2 Genetic Programming
3 Federated Learning
4 Methodology
5 Case Study
6 Discussion and Results
7 Conclusion
References
Artificial Intelligence in Medicine
A Scoping Review of Energy Load Disaggregation
1 Introduction
2 Methodology
3 Results
3.1 Applied Domains
3.2 Data and Data Sources
3.3 Related Methods
4 Discussion
5 Conclusion
References
Deep Learning Survival Model to Predict Atrial Fibrillation From ECGs and EHR Data
1 Introduction
2 Materials and Methods
2.1 Data
2.2 Model Development
2.3 Experimental Setting
2.4 Evaluation Metrics
3 Results
4 Discussion
5 Conclusion
References
Generalization Ability in Medical Image Analysis with Small-Scale Imbalanced Datasets: Insights from Neural Network Learning
1 Introduction
2 Methods
2.1 Definition of Neural Network Architecture Components
2.2 Generalization Ability
2.3 Model Complexity
3 Results and Discussion
4 Conclusion
References
Multi-omics Data Integration and Network Inference for Biomarker Discovery in Glioma
1 Introduction
2 Materials and Methods
2.1 Graphical Lasso
2.2 Network Distance
2.3 Data Description
2.4 Pipeline and Implementation
2.5 Network Validation
3 Results
3.1 Variable Selection
3.2 Protein Networks
3.3 Validation Outcomes
4 Discussion
References
Better Medical Efficiency by Means of Hospital Bed Management Optimization—A Comparison of Artificial Intelligence Techniques
1 Introduction
2 Background
2.1 Resources Planning in Hospital Settings
2.2 Related Work
3 Materials and Methods
3.1 Methodologies
3.2 Tools and Algorithms
3.3 Data Sets
4 Experiments
4.1 Problem Formulation
4.2 Data Provided
4.3 Data Preparation
4.4 Domain and Fitness Function
4.5 Optimization Techniques
4.6 Evaluation
5 Results and Discussion
5.1 Algorithm Settings
5.2 Results
6 Conclusions
References
AI-Based Medical Scribe to Support Clinical Consultations: A Proposed System Architecture
1 Introduction
2 Literature Review
2.1 Digital Medical Scribe
2.2 Automatic Speech Recognition and Natural Language Processing Algorithms
3 System Architecture
4 Conclusion and Further Work
References
Combining Neighbor Models to Improve Predictions of Age of Onset of ATTRv Carriers
1 Introduction
2 Background
2.1 Ensemble Learning
2.2 Related Work
3 Single Learning Approach and Combination Strategies
3.1 Prediction Problem and Single Learning Approach
3.2 Data and Evaluation Strategy
3.3 Combination Strategies
3.4 Evaluation
4 Results and Discussion
5 Conclusions and Future Work
References
Unravelling Heterogeneity: A Hybrid Machine Learning Approach to Predict Post-discharge Complications in Cardiothoracic Surgery
1 Background
2 Dataset
3 Methodology
3.1 Unsupervised Learning Strategy
3.2 Supervised Learning Strategy
4 Results
4.1 Clustering
4.2 Classification
5 Discussion
6 Conclusion
References
Leveraging TFR-BERT for ICD Diagnoses Ranking
1 Introduction
2 Related Work
3 Methodology
3.1 Overview
3.2 Learning-to-Rank System
3.3 Fine-Tuned Language Representation Model
4 Experiments and Results
4.1 Dataset
4.2 Training Parameters
4.3 Metrics
4.4 Additional Approaches
5 Results
5.1 Overall ICD Code Ranking
5.2 Windowed Training Approach
5.3 Summarization Training Approach
6 Discussion
7 Conclusion and Future Work
References
Artificial Intelligence and IoT in Agriculture
Evaluating the Causal Role of Environmental Data in Shellfish Biotoxin Contamination on the Portuguese Coast
1 Introduction
2 Methods
2.1 Data Sources and Preparation
2.2 Time Series Modelling
3 Results and Discussion
3.1 Correlation Analysis
3.2 Causality Analysis
3.3 Time Series Regression Models
3.4 Dynamic Bayesian Network
4 Conclusion
References
Sound-Based Anomalies Detection in Agricultural Robotics Application
1 Introduction
2 System Design
3 Materials and Methods
3.1 Data Acquisition
3.2 Pre-processing and Dataset Generation
3.3 Data Training and Evaluation
4 Results and Discussion
5 Conclusion
References
Can the Segmentation Improve the Grape Varieties' Identification Through Images Acquired On-Field?
1 Introduction
2 Methods
2.1 Data Collection and Preparation
2.2 Experiments Design
2.3 Data Segmentation
2.4 Data Classification and Evaluation
3 Results and Discussion
3.1 Segmentation Experiment
3.2 Classification Experiments
4 Conclusion
References
Enhancing Pest Detection Models Through Improved Annotations
1 Introduction
2 Background and Related Work
2.1 Object Detection Models
2.2 Data-Centric AI—Improving Annotations
3 Methodology
3.1 Dataset Creation
3.2 Architecture Selection
3.3 Improving Annotations Quality
4 Results and Discussion
5 Conclusions and Future Work
References
Deep Learning-Based Tree Stem Segmentation for Robotic Eucalyptus Selective Thinning Operations
1 Introduction
2 Materials and Methods
2.1 Data Acquisition
2.2 Data Annotation and Augmentation
2.3 Training and Testing of Deep Learning Models
2.4 Automatic Stem Selection Algorithm
3 Results
4 Conclusions
References
Segmentation as a Pre-processing for Automatic Grape Moths Detection
1 Introduction
2 Methodology
2.1 Dataset and Data Preparation
2.2 Segmentation Task
2.3 Detection Task
3 Results and Discussion
3.1 Segmentation Task
3.2 Detection Task
4 Conclusion
References
Artificial Intelligence in Transportation Systems
Safety, Stability, and Efficiency of Taxi Rides
1 Introduction
1.1 Step 1: Collecting Harassment Claims
1.2 Step 2: Penalizing Harassing Drivers
1.3 Step 3: Matching Drivers and Passengers
2 Related Work
3 Our Novel Model
3.1 Preferences Layers
3.2 Matchings
4 Phase 1: Computing Jointly Safe Sub-Matchings
4.1 Joint Safety
4.2 Joint Safety and Efficiency
5 Phase 2: Computing Stable Sub-Matchings
5.1 Stability
5.2 Stability and Efficiency
6 Conclusion and Future Work
References
Improving Address Matching Using Siamese Transformer Networks
1 Introduction
2 Background and Related Work
3 Data Description and Preparation
3.1 Addresses Structure
3.2 Datasets
4 Model Implementation
4.1 Proposed Model: Bi-Encoder + Cross-Encoder
4.2 Training Overview
4.3 Model Evaluation
5 Results and Discussion
5.1 Inference Time
5.2 Accuracy Results—Test Dataset
6 Conclusions
References
An Ethical Perspective on Intelligent Transport Systems
1 Introduction
2 Related Work
3 The Relevance of Transport for Ethics
3.1 Transport Impacts that May Raise Ethical Concerns
3.2 From an Ethical Standpoint
4 Ethical Theories and Transport
4.1 Common Ethical Principles
4.2 Forms of Ethical Theories
5 Ethics for Intelligent Transport Systems
5.1 Novel Ethical Issues
5.2 Novel Dimensions of Ethical Concern
6 Conclusions
References
Using CDR Data to Understand Post-pandemic Mobility Patterns
1 Introduction
2 Related Work
3 Data Exploration
3.1 COVID-19 Statistics
3.2 Study Area
3.3 Call Detail Records
4 Mobility Analysis
4.1 Comparison and Discussion
5 Conclusions
References
Artificial Intelligence in Smart Computing
Using Artificial Intelligence for Trust Management Systems in Fog Computing: A Comprehensive Study
1 Introduction
2 Related Reviews
3 Non-AI Methods for Trust Management in the Fog
4 Adopting AI Techniques for Trust Management in the Fog
5 Discussion and Insights
6 Conclusion
References
Source-Code Generation Using Deep Learning: A Survey
1 Introduction
1.1 Related Surveys
1.2 Survey Organization
2 Surveyed Approaches
2.1 Text-To-Code
2.2 Image-To-Code
3 Datasets
4 Metrics
5 Conclusion
References
An IoT-Based Framework for Sustainable Supply Chain Management System
1 Introduction
2 Literature Review
3 System Design
3.1 Research Approach
3.2 Implementation of ICT, IoT, and Big Data in Smart Supply Chain
3.3 Key Performance Indicators
3.4 Design
4 Findings
5 Discussion and Conclusions
References
Artificial Intelligence for Industry and Societies
Tool Wear Monitoring Using Multi-sensor Time Series and Machine Learning
1 Introduction
2 State of the Art
2.1 Tool Wear Sensing and Prediction
2.2 Machine Learning Models
3 Methodology
3.1 Context
3.2 Initial Approach: Dataset Validation and Baseline
3.3 Focus on Machine Learning Algorithms Using Statistical Features
3.4 Focus on Stationary Components of the Signals
3.5 Mixing Spectrogram Approach and Stationary Components of the Signals
4 Results
5 Conclusion—Discussion
References
Digital Twins: Benefits, Applications and Development Process
1 Introduction
2 Benefits of Digital Twins
3 Digital Twin Applications
4 Digital Twin Application Development
5 Conclusions
References
Using Deep Learning for Building Stock Classification in Seismic Risk Analysis
1 Introduction
2 Related Work
2.1 Building Classification
2.2 Creating Exposure Models
3 Materials and Methods
4 Results and Analysis
5 Conclusions
References
Data Mining Models to Predict Parking Lot Availability
1 Introduction
2 Background
2.1 Smart Cities
2.2 Data Mining
2.3 Related Work
3 Material and Methods
3.1 Methods
3.2 Technologies and Tools
4 Work
4.1 Data Analysis (Data Comprehension)
4.2 Data Preparation
4.3 Modelling
4.4 Evaluation
5 Discussion
6 Conclusion
References
Advancements in Synthetic Data Extraction for Industrial Injection Molding
1 Introduction
2 Related Work
2.1 Synthetic Data Generation
2.2 Injection Molding Simulation
2.3 ML in Injection Molding
3 Methods
3.1 Simulation of Production Processes
3.2 Labeling the Simulated Product Cycles
3.3 Enrichment of Training Sets
3.4 Testing and Evaluation
4 Experimental Setup
4.1 Real Training and Validation Data
4.2 Synthetic Data
4.3 LSTM Training
5 Results
5.1 Quality of Classification
6 Evaluation
7 Conclusion
References
Vision Transformers Applied to Indoor Room Classification
1 Introduction
2 Related Work
3 Image Classification Models
3.1 Vision Transformer
3.2 Benchmark Models
4 Methodology
5 Results
6 Conclusion
References
Author Index