Progress in Artificial Intelligence 21st EPIA Conference on Artificial Intelligence, EPIA 2022, Lisbon, Portugal, August 31-September 2, 2022, Proceedings

This book constitutes the proceedings of the 21st EPIA Conference on Artificial Intelligence, EPIA 2022, which took place in Lisbon, Portugal, in August/September 2022. The 64 papers presented in this volume were carefully reviewed and selected from 85 submissions. They were organized in topical sec...

Full description

Saved in:
Bibliographic Details
Main Authors Marreiros, Goreti, Martins, Bruno, Paiva, Ana, Ribeiro, Bernardete, Sardinha, Alberto
Format eBook
LanguageEnglish
Published Cham Springer International Publishing AG 2022
Springer International Publishing
Edition1
SeriesLecture Notes in Computer Science. Lecture Notes in Artificial Intelligence
Subjects
Online AccessGet full text
ISBN9783031164736
3031164733

Cover

Table of Contents:
  • Intro -- Preface -- Organization -- Contents -- AI4IS - Artificial Intelligence for Industry and Societies -- Estimating the Temperature on the Reinforcing Bars of Composite Slabs Under Fire Conditions -- 1 Introduction -- 2 Transient Thermal Problem -- 2.1 Physical Multidomains -- 2.2 Boundary Conditions Corresponding to a Standard Fire -- 2.3 Analytical Method Provided by the Standard Eurocode -- 3 Improving the Analytical Method with Numerical Results -- 3.1 Computational Solution by Finite Elements Method -- 3.2 Improving the New Proposal with an Optimization Method -- 3.3 Improving the New Proposal by the Linear Least Squares Method -- 3.4 Comparison of the Results -- 4 Conclusion -- References -- Hierarchically Structured Scheduling and Execution of Tasks in a Multi-agent Environment -- 1 Introduction -- 2 Related Work -- 2.1 Resource Management -- 2.2 Hierarchical Reinforcement Learning -- 3 Background -- 3.1 Markov Decision Problems -- 3.2 Reinforcement Learning -- 3.3 Multi-agent Markov Decision Problems -- 3.4 Multi-agent Reinforcement Learning -- 3.5 Hierarchical Reinforcement Learning -- 4 Problem Setting -- 4.1 High-Level -- 4.2 Low-Level -- 4.3 Implementation Details -- 5 Evaluation -- 5.1 Environment Settings -- 5.2 Experimental Results -- 5.3 Additional Experiments -- 6 Conclusion -- References -- AIL - Artificial Intelligence and Law -- Content-Based Lawsuits Document Image Retrieval -- 1 Introduction -- 2 Proposal -- 3 Application -- 3.1 Contextualization and Database -- 3.2 Preprocessing and Feature Extraction -- 3.3 Similarity Calculation and Result Presentation -- 4 Experiments -- 4.1 CNN's Choice -- 4.2 Hybrid Algorithm Evaluation -- 5 Conclusion -- References -- Lawsuits Document Images Processing Classification -- 1 Introduction -- 2 Related Work -- 3 Background -- 4 Proposal -- 4.1 Preprocessing -- 4.2 Classification
  • 3 MedCloudCare -- 3.1 React Frontend -- 3.2 Django Backend API -- 3.3 Machine Learning Modelling -- 3.4 Docker and Kubernetes for ML Inference -- 4 Results -- 5 Conclusions and Future Work -- References -- An Active Learning-Based Medical Diagnosis System -- 1 Introduction -- 2 Related Work -- 3 Materials -- 4 Methods -- 4.1 Preprocessing -- 4.2 Active Learning Component -- 4.3 Metrics -- 4.4 Experimental Design -- 5 Results -- 6 Discussion -- 7 Conclusion -- References -- Comparative Evaluation of Classification Indexes and Outlier Detection of Microcytic Anaemias in a Portuguese Sample -- 1 Background -- 2 Methods -- 2.1 Data Description -- 2.2 Classification Indexes Evaluation -- 3 Results and Discussion -- 3.1 Descriptive Data Analysis -- 3.2 Indexes Evaluation -- 3.3 Binary and Multi-class Microcytic Anaemia Classifiers -- 3.4 Outliers Detection -- 4 Conclusions -- References -- A General Preprocessing Pipeline for Deep Learning on Radiology Images: A COVID-19 Case Study -- 1 Introduction -- 2 Proposed Approach -- 2.1 CT Scans Preprocessing for 3D Deep Learning Architecture -- 3 CT Scan Normalization -- 3.1 Deep Learning Architecture for CT Scan Classification -- 4 Experiment and Results -- 4.1 Dataset and Preprocessing -- 4.2 Implementation Details and Model Training -- 4.3 Results -- 5 Conclusion -- References -- AIPES - Artificial Intelligence in Power and Energy Systems -- Automatic Configuration of Genetic Algorithm for the Optimization of Electricity Market Participation Using Sequential Model Algorithm Configuration -- 1 Introduction -- 2 Related Work -- 3 Automatic Configuration of Genetic Algorithm for Market Participation Portfolio Optimization -- 3.1 Portfolio Optimization -- 3.2 Genetic Algorithm -- 3.3 Sequential Model Algorithm Configuration (SMAC) -- 4 Experimental Findings -- 5 Conclusions -- References
  • Modeling Stand-Alone Photovoltaic Systems with Matlab/Simulink
  • 5 Experiment -- 5.1 Preprocessing -- 5.2 Classification -- 6 Conclusion -- References -- A Rapid Semi-automated Literature Review on Legal Precedents Retrieval -- 1 Introduction -- 2 Materials and Methods -- 2.1 Rapid Reviews -- 2.2 Literature Review Automation -- 2.3 Keyword Identification -- 2.4 Electronic Databases and Eligibility Criteria -- 2.5 Data Extraction and Pre-processing -- 2.6 Screening Based on Keyword Frequency -- 2.7 Topic Modelling for Eligibility -- 2.8 Full-Text Screening for Inclusion -- 3 Results -- 3.1 Descriptive Analytics -- 3.2 Content Analysis -- 4 Discussion -- 4.1 RQ1 (How has the Task of Automating the Identification of Previous Relevant Cases been Addressed by the Researchers?) and RQ2 (What Types of Techniques are Covered in the Reviewed Studies?) -- 4.2 RQ3: What are the Most Promising Strategies and Research Gaps in Automating the Retrieval of Legal Precedents? -- 5 Conclusions, Limitations, and Future Research -- References -- The European Draft Regulation on Artificial Intelligence: Houston, We Have a Problem -- 1 Presentation of the European Draft Regulation -- 2 Risk-Based Approach -- 3 The Conformity Assessment -- 4 Difficulty in Complying with Some Requirements -- 5 Contextualization Within the EU Legal Framework -- 6 Lack of Protection of CITizen's Rights -- 7 Discouragement to Technological Development -- 8 Houston, We Have a Problem -- References -- Traffic Stops in the Age of Autonomous Vehicles -- 1 Background -- 1.1 Autonomous Vehicle Development -- 1.2 Likely Impacts of Autonomous Vehicles on Traffic Stops -- 2 The Legal Complexities of AV Traffic Stops -- 2.1 Level 2 (Semi-autonomous) Vehicles -- 2.2 Level 3 (Semi-autonomous) Vehicles -- 2.3 Level 4 &amp -- 5 (Fully Autonomous) Vehicles -- 3 Potential Solutions -- References -- UlyssesSD-Br: Stance Detection in Brazilian Political Polls
  • Audio Feature Ranking for Sound-Based COVID-19 Patient Detection -- 1 Introduction -- 1.1 The Paper's Contributions -- 2 Audio Features Overview -- 2.1 Time Domain -- 2.2 Frequency Domain -- 2.3 Time-Frequency Domain -- 3 Experimental Method and Results -- 3.1 Research Questions -- 3.2 The Datasets -- 3.3 Feature Engineering -- 3.4 Results Description and Analysis -- 4 Related Work -- 5 Conclusion and Future Work -- References -- Using a Siamese Network to Accurately Detect Ischemic Stroke in Computed Tomography Scans -- 1 Introduction -- 2 Related Work -- 3 Dataset and Methods -- 3.1 Image Preprocessing -- 3.2 Symmetry Detection -- 3.3 Proposed Architecture -- 3.4 Loss Function -- 4 Results -- 5 Conclusions and Future Work -- References -- Determining Internal Medicine Length of Stay by Means of Predictive Analytics -- 1 Introduction -- 2 Background -- 2.1 Resources Planning in Hospital Settings -- 2.2 Related Works -- 3 Materials and Methods -- 3.1 Methodologies -- 3.2 Tools and Algorithms -- 3.3 Data Sets -- 4 Case Study -- 4.1 Business Understanding -- 4.2 Data Understanding -- 4.3 Data Preparation -- 4.4 Modeling -- 4.5 Evaluation -- 5 Results and Discussion -- 6 Conclusions -- References -- Improving the Prediction of Age of Onset of TTR-FAP Patients Using Graph-Embedding Features -- 1 Introduction -- 2 Feature Construction and Embedding-Based Network Representation Learning -- 3 Model Overview -- 3.1 Problem Definition -- 3.2 Defining and Using Embeddings -- 3.3 Experimental Setup -- 4 Results and Discussion -- 4.1 Overall Evaluation of the Different Embedding Variants -- 4.2 Studying the Effect of Predicting the Age of Onset of Asymptomatic Patients at 22 Years Old -- 5 Conclusions and Future Work -- References -- Cloud-Based Privacy-Preserving Medical Imaging System Using Machine Learning Tools -- 1 Introduction -- 2 Related Work
  • 1 Introduction -- 2 Related Works -- 3 UlyssesSD-Br Corpus -- 3.1 Data Collection -- 3.2 Annotation -- 3.3 Data Analysis -- 4 Experimental Setup -- 5 Results and Discussion -- 5.1 Experiments in UlyssesSD-Br Corpus -- 5.2 Experiments in Elicited and Twitter Corpora -- 5.3 Experiments in Elicited and Twitter Corpora Using the UlyssesSD-Br Model Knowledge -- 6 Conclusion -- References -- Unraveling the Algorithms for Humanized Digital Work Oriented Artificial Intelligence -- 1 Introduction -- 2 Living the Age of Artificial Intelligence -- 3 Algorithms in the Constitution of AI -- 4 Platforms and Algorithms: Working Against Workers -- 5 Conclusion -- References -- The Compatibility of AI in Criminal System with the ECHR and ECtHR Jurisprudence -- 1 The Use of AI Systems as an Aid in Determining Sentence Length -- 1.1 Supporting Decision in the Criminal Sentencing -- 2 The Use of AI in Light of ECHR -- 2.1 Right to a Fair Trial -- 2.2 No Punishment Without Law -- 2.3 Prohibition of Discrimination -- 3 Conclusions -- References -- Enriching Legal Knowledge Through Intelligent Information Retrieval Techniques: A Review -- 1 Introduction -- 2 Legal Knowledge Representation -- 3 Approaches for Information Retrieval in Legal Documents -- 3.1 Information Science -- 3.2 Artificial Intelligence -- 3.3 An Overview of the Related Background and Prior Research -- 4 Conclusion -- References -- AIM - Artificial Intelligence in Medicine -- Region of Interest Identification in the Cervical Digital Histology Images -- 1 Introduction -- 2 Related Work -- 3 Method -- 3.1 Cervical Slide Image Processing -- 3.2 Bounding Box (bb) Construction Procedure -- 3.3 Extracted bb Classification -- 3.4 Epithelium RoI Identification -- 4 Experiments -- 4.1 Dataset -- 4.2 Procedure -- 4.3 Assessment -- 4.4 Comparison with State-Of-The-Arts -- 5 Conclusion -- References