Proceedings of the 15th International Conference on Ubiquitous Computing & Ambient Intelligence (UCAmI 2023): Volume 2 (Lecture Notes in Networks and Systems)

This book serves as a comprehensive compilation of groundbreaking research endeavors within the realms of ambient intelligence and ubiquitous computing. These initiatives are pivotal in enabling both researchers and practitioners to discern recent breakthroughs and emerging frontiers in these fields. Encompassing a wide array of domains, including Ambient Active and Assisted Living (A3L), the Internet of Things (IoT), Smart Environments, Data Science, and Human-Ambient Interaction, acts as a valuable resource for scholars, professionals, and graduate students alike. The primary aim of this book is to empower individuals within the academic and professional community to harness this wealth of knowledge. It equips them to tackle innovative challenges and engineer smart and ubiquitous solutions that will shape the landscape of the next decade. By amalgamating insights from various facets of ambient intelligence and ubiquitous computing, this book encourages cross-disciplinary collaboration and fosters a holistic understanding of the field. Thus, it not only highlights the recent strides in these areas but also serves as a roadmap for future exploration and innovation, paving the way for a smarter and more interconnected world. Preface Organization Contents About the Editors Smart Environment Reinforcement Learning Model in Automated Greenhouse Control 1 Introduction 2 Related Work 3 Reinforcement Learning Using Predictive Control in the Greenhouse 4 Internet of Things and User-Centered Design for the RL Model 5 Application in an Automated Greenhouse 6 Conclusions References Performance Evaluation of Face Detection Algorithms for an Emotion Recognition Application in a School in the Department of Nariño - Colombia 1 Introduction 2 Related Work 3 Configuration of the Working Space and Capture of Images 4 Performance Evaluation of Face-Detection Algorithms 5 Conclusions References An Open Framework for Nonverbal Communication in Human-Robot Interaction 1 Introduction 2 Nonverbal Human-Robot Interaction 2.1 Types of Nonverbal Interactions in HRI 2.2 Design Scenarios 3 Architecture of the Nonverbal HRI Framework 3.1 The Eva Social Robot 3.2 MQTT Broker 3.3 Wearable and Environmental Sensors 3.4 Rhasspy 3.5 EvaNotebook 3.6 Data Analysis Models 4 Implementation of Scenarios 4.1 Scenario 1: Gaze and Proxemics to Initiate an Interaction 4.2 Scenario 2: Suggesting a Cooking Recipe 5 Conclusion References A First Approach to the Generation of Linguistic Summaries from Glucose Sensors Using GPT-4 1 Introduction 2 Chatgpt 3 Generation of Linguistic Summaries 4 Linguistic Summaries Using ChatGPT 4.1 Methodology 4.2 Description of the Problem 4.3 Simplification Process 5 Experimentation 6 Evaluation 7 Conclusion References Efficient and Accountable Industry 5.0 Production Scheduling Mechanism for Mass Customization Scenarios 1 Introduction 2 Industry 5.0 Technologies and Mass Customization Solutions 3 Efficient and Accountable Industry 5.0 Production Scheduling 3.1 Prosumer Environment and Price Calculation 3.2 Transparent and Accountable Blockchain-Enabled Marketplace 3.3 Efficient Production Scheduling 4 Experimental Validation: Simulation and Results 5 Conclusions and Future Work References Non-invasive Synthesis from Vision Sensors for the Generation of 3D Body Landmarks, Locations and Identification in Smart Environments 1 Introduction 2 Materials and Methods 2.1 Sensor, Deployment and Setup 2.2 Body Segmentation and 3D Landmark Body Estimation 2.3 Computing 2D Location of Inhabitants 2.4 Non-supervised Deep Learning Tracking in Multi-occupancy 2.5 Face-Based Identification 2.6 Representing Data in a Virtual Environment 3 Case Study 3.1 Results 3.2 Limitations of the Study 4 Conclusions and Ongoing Works References A Model to Support the Prediction of Indicators in the Diagnosis and Intervention of Autism Spectrum Disorder 1 Introduction 2 Case Study 3 Conclusions References ADLnet: A 1d-CNN for Activity of Daily Living Recognition in Smart Homes 1 Introduction 2 Related Works 3 Proposed Solution 4 Experimental Results 5 Conclusions References NeeMAS: A Need-Based Multi-agent Simulator of Human Behavior for Long-Term Drifts in Smart Environments 1 Introduction 2 Related Works 3 Proposed Approach 3.1 Behavioural Drifts 4 Experimental Results 5 Conclusions and Future Work References Predictive Modeling of Body Shape Changes in Individuals on Dietetic Treatment Using Recurrent Networks 1 Introduction 2 Related Works 2.1 3D Human Body Representation 2.2 Neural Networks for Learning Data Sequences 3 BSP-RNN: Predicting Body Shape Using Recurrent Networks 3.1 BSP-RNN Workflow 3.2 Estimation of Body Parameters 3.3 Prediction of Body Shape 4 Experiments 4.1 Experimental Setup 4.2 Experimental Results 5 Conclusions References Understanding Students' Perspectives About Human-Building Interactions in the Context of Smart Buildings 1 Introduction 2 Related Work 3 Methods 4 Results 5 Discussion 6 Conclusions and Future Directions References IoT-Driven Real-Time Glucose Monitoring: Empowering Diabetes Care and Prevention 1 Introduction 2 Related Works 3 Proposed Real-Time Glucose Monitoring System 3.1 Proposed Glucose Sensor 3.2 Architecture of the Proposed System 4 Quantitative Evaluation of the Effectiveness of the Proposed Real-Time Glucose Monitoring System 5 Alignment with Sustainable Development Goals 6 Conclusions and Futures Works References Edge IoT System for Wearable Devices: Real-Time Data Processing, Inference, and Training for Activity Monitoring and Health Evaluation 1 Introduction 2 Related Works 2.1 Wearable Devices and Biomedical Kits 2.2 Data Fusion from Heterogeneous Devices 3 Matherials and Methods 3.1 System Architecture 3.2 Biosignal Devices 3.3 Data Resampling and Synchronisation 3.4 AI Models to Fuse the Information of the Sensors 3.5 Online Training in NVIDIA Jetson TX2 4 Experimental Results 4.1 Dataset Details 4.2 Hardware Infrastructure 4.3 Results 5 Conclusions and Future Works References Analysing Requirements Specification Languages for Self-adaptive AAL Systems 1 Introduction 2 Case Study 3 Self-adaptive Requirements Specification 3.1 RELAX 3.2 Tropos4S 3.3 GODA 4 Experimental Design 5 Analysis of Results 6 Related Work 7 Conclusions References A Deep Learning and Probabilistic Approach to Recognising Activities of Daily Living with Privacy Preserving Thermal Sensors 1 Introduction 2 Methodology 2.1 Deployment of Thermopile Infrared Sensors Within a Smart Environment 2.2 Classifying Poses 2.3 Determining the Nearest Object 2.4 Inferring Sub-activities 2.5 Using a Hidden Markov Model to Represent Activities of Daily Living 3 Experimental Design 4 Experimental Results and Discussion 4.1 Pose Classification Results 4.2 Activities of Daily Living Recognition Results 5 Conclusions References Using Synthetic Data to Improve the Accuracy of Human Activity Recognition 1 Introduction 2 Background and Related Work 3 Methodology 4 Implementation and Results 5 Discussion and Conclusions References Smart-Lab IoT Research Platform with Modern Open Source Components 1 Introduction 2 Related Work 2.1 Solution Requirements in the Context of Research 3 Framework for an Open-Source IoT Research Platform Ecosystem 3.1 Architectural Choices 3.2 The Implemented Architecture 4 Case Study: Multi-modal Thermal Fall Detection 5 Conclusion References Secure, Transparent, Supply Chain Supporting Agri-Food Provenance and Sustainability 1 Introduction 2 Related Work 3 A Solution to Provide Secure, Dynamic, and Traceable Agrifood Supply Chains 3.1 Requirements Elicitation 3.2 Solution Design 4 Agri-food Use Cases 4.1 Challenges 4.2 How Does the Solution Address the Challenges? 5 Future Work 6 Conclusion References Dataset Generation and Study of Deepfake Techniques 1 Introduction 2 Deepfake Studies 2.1 Generation Techniques 2.2 Detection Techniques 3 Deepfake Dataset 3.1 Discovery Tools 3.2 The Generated Dataset 4 Conclusions and Further Works References Understanding the Association of Driving Safety and Visual Behaviors Collected Through Smart Sensing Technology 1 Introduction 2 Method 2.1 Study Procedure 2.2 Cognitive Skills Measures 2.3 On-Road Driving Assessment 2.4 Visual Behavior Measures 3 Results 4 Conclusions References Data Science Clustering ABI Patients for a Customized Rehabilitation Process 1 Introduction 2 Related Work 3 Clustering for ABI Patients 3.1 Dataset 3.2 Pre-processing 3.3 Clustering Process 3.4 Post-processing 4 Conclusions and Future Work References Comparative Study of Large Language Models as Emotion and Sentiment Analysis Systems: A Case-Specific Analysis of GPT vs. IBM Watson 1 Introduction 2 Materials and Methods 2.1 Data Processing 2.2 Sentiment Analysis and Emotion Recognition 2.3 Evaluation 3 Results 3.1 Sentiment Analysis 3.2 Emotion Recognition 4 Discussions 5 Conclusions References Meta-learning. An Approach Applied to Clinical Data 1 Introduction 2 Architecture 3 Case Study 3.1 Levels of Obesity 3.2 Early Detection of Diabetes 4 Conclusions References Bridging the Gap: Enhancing Geospatial Analysis with Natural Language and Scenario Generation Language 1 Introduction 2 Related Work 2.1 Using GPT-LLM to Generate Structured Output from Natural Language Instructions 2.2 The Scenario Generation Language 3 Implementing a Natural Language Interface for SGL 4 Preliminary Results 5 Generalizing the Process 6 Conclusions References DAKTILOS: An Interactive Platform for Teaching Mexican Sign Language (LSM) 1 Introduction 2 LSM Alphabet 3 Methods 4 Evaluation 5 Conclusions References Evaluation of Four Classification Algorithms Applied to Detection of Poisonous Mushrooms 1 Introduction 2 Materials and Methods 2.1 Data Description 2.2 Data Preprocessing 2.3 Classification Algorithms 2.4 Validation 3 Experiments and Results 4 Discussion and Conclusions References Traceable Health Data for Consciously Trained ML Models 1 Introduction 2 Blockchain and Web3 in Healthcare 3 Proposed Approach 3.1 The TRHEAD Reference Architecture for eHealth Data Collection and Usage 3.2 Architecture 4 Implementation Results: BCRAT 5 Conclusions and Discussion References Viric Learning - A Novel Transfer Learning Method 1 Introduction 1.1 Application of Nature Inspired Optimization Algorithms 2 Approach 2.1 Viric Learning Method 2.2 Activity Dataset 2.3 Experiment Baseline 3 Evaluation and Results 3.1 Viric Data vs. Natural Data 3.2 Experiments 4 Conclusions References Granular Linguistic Model Based Multimodal Data Integration for Automated Evaluation of Core Soft Skills 1 Introduction 2 Background 3 Methodological Proposal 3.1 Conceptualization Framework 3.2 Multimodal Feature Analyzer 3.3 Multimodal Soft Skills Interpreter 3.4 Prompt Generator 4 Case of Study 4.1 Conceptualization Framework 4.2 Multimodal Feature Analyzer 4.3 Multimodal Soft-Skill Interpreter 4.4 Prompt Generator 5 Conclusions References Understanding the Role of the User in Information Propagation on Online Social Networks: A Literature Review and Proposed User Model 1 Introduction 2 Literature Review 2.1 Research Question 2.2 Search of Primary Researches 2.3 Review of Articles for Inclusion and Exclusion 2.4 Classification of Articles 2.5 Data Extraction and Mapping 3 Review of the Results 4 Proposal of User Model in Information Propagation 5 Conclusions, Future Works and Limitations References Author Index