Hybrid Artificial Intelligent Systems 16th International Conference, HAIS 2021, Bilbao, Spain, September 22-24, 2021, Proceedings

This book constitutes the refereed proceedings of the 16th International Conference on Hybrid Artificial Intelligent Systems, HAIS 2021, held in Bilbao, Spain, in September 2021.The 44 full and 11 short papers presented in this book were carefully reviewed and selected from 81 submissions.

Saved in:
Bibliographic Details
Main Authors Sanjurjo González, Hugo, Pastor López, Iker, García Bringas, Pablo, Quintián, Héctor, Corchado, Emilio
Format eBook
LanguageEnglish
Published Netherlands Springer Nature 2021
Springer International Publishing AG
Springer International Publishing
Edition1
SeriesLecture Notes in Computer Science
Subjects
Artificial intelligence
Special computer methods
Online AccessGet full text
ISBN9783030862718
3030862712
3030862704
9783030862701

Cover

Table of Contents:
  • 5 Performance Evaluation -- 5.1 Datasets -- 5.2 Experimental Setup -- 5.3 Experimental Results -- 6 Conclusions and Future Work -- References -- Learning Bipedal Walking Through Morphological Development -- 1 Introduction -- 2 Designing a Morphological Development Strategy -- 3 Experimental Setup -- 4 Results and Discussion -- 5 Conclusion and Future Work -- References -- Ethical Challenges from Artificial Intelligence to Legal Practice -- 1 Introduction. From Justice to eJustice -- 2 Objectives -- 3 Methodology -- 4 Developing. The Application of Algorithms in Justice -- 4.1 Artificial Legal Intelligence -- 4.2 Jurimetrics. Legal Decision Support System -- 4.3 Legal Expert Systems -- 5 Results: First Ethical Questions Derived from the Use of AI in Legal Procedures and Processes -- 5.1 Publicity Principle -- 5.2 Principle of Contradiction and Right of Defense -- 5.3 Principle of Procedural Equality or Equality of the Parts -- 5.4 Influence of AI on the Principles of Criminal Process -- 5.5 Preventive and Predictive AI -- 6 Discussion. Future Challenges: Understandability, Intepretability, Explicability, Transparency, and Algorithmic Traceability as Ethical Challenges -- 6.1 A Relationship Between the Jurimetrics with the Global Framework of the eJustice that Guarantees the Nature of the Latter as an Effective Attorney of Justice -- 6.2 Respect for Procedural Data by Means of Big Data Techniques that Respect the Rights that Are Set Out in Spanish or European Legislation -- 6.3 Transparency in a Digitized Jurisdictional Framework -- 6.4 Establishment of Responsibilities for the People Who Define the Prediction Algorithms on the Prosperousness of a Claim -- 6.5 Irreplaceability of Lawyers or Other Judicial Agents by AI Legal Systems
  • 6.6 Guarantee that the Digitization of the Judicial Process Does not Destroy the Analysis and Legal Argumentation that Any Matter Requires -- 7 Conclusions -- 7.1 Need for Codes of Ethics and Legal Regulations Regarding the Use of AI Technologies -- 7.2 Protection of Privacy -- 7.3 Algorithmic Ethics and Research in Explanatory AI (XAI) -- References -- Visual Analysis and Advanced Data Processing Techniques -- Classification of Burrs Using Contour Features of Image in Milling Workpieces -- 1 Introduction -- 2 Inspection Method -- 2.1 Image Processing -- 2.2 Region of Interest Identification -- 2.3 Feature Vector Calculation -- 2.4 Classification -- 3 Experimental Results -- 4 Conclusions -- References -- Adaptive Graph Laplacian for Convex Multi-Task Learning SVM -- 1 Introduction -- 2 Multi-Task Learning SVMs -- 2.1 Convex Multi-Task Learning SVM -- 2.2 Convex Graph Laplacian MTL SVM -- 3 Convex Adaptative Graph Laplacian MTL SVM -- 4 Synthetic Experimental Results -- 5 Discussion and Conclusions -- References -- Fight Detection in Images Using Postural Analysis -- 1 Introduction -- 2 Related Work -- 3 Existing Datasets for Fighting Detection -- 4 Methodology -- 4.1 Data Gathering -- 4.2 Angles Extraction from Skeleton -- 4.3 Dataset Generation -- 5 Experimentation and Results -- 5.1 Feature Subset Selection -- 5.2 Analysis of Network Learning Curves -- 6 Conclusions -- References -- Applying Vector Symbolic Architecture and Semiotic Approach to Visual Dialog -- 1 Introduction -- 2 Related Works -- 3 Vector Semiotic Architecture for Visual Dialog -- 4 Experiments -- 5 Discussion -- 6 Conclusion -- References -- Transfer Learning Study for Horses Breeds Images Datasets Using Pre-trained ResNet Networks -- 1 Introduction and Motivation -- 2 The Proposal -- 3 Numerical Results -- 3.1 Materials and Methods -- 3.2 Results -- 4 Conclusion and Future Work
  • Intro -- Preface -- Organization -- Contents -- Data Mining, Knowledge Discovery and Big Data -- Document Similarity by Word Clustering with Semantic Distance -- 1 Introduction -- 2 Vector Space Model and Dimension Reduction -- 2.1 Vector Space Model -- 2.2 Distance by WordNet -- 2.3 Distance by Word2Vec -- 2.4 Word Clustering -- 2.5 Dimension Reduction by Semantic Distance -- 2.6 Cluster-Word Relation Matrix -- 3 Experimental Results -- 3.1 Conditions of Experiments -- 3.2 Similarity of Documents -- 3.3 Result of Each Method -- 4 BBC Dataset -- 5 Conclusion -- References -- PSO-PARSIMONY: A New Methodology for Searching for Accurate and Parsimonious Models with Particle Swarm Optimization. Application for Predicting the Force-Displacement Curve in T-stub Steel Connections -- 1 Introduction -- 2 Related Works -- 3 PSO-PARSIMONY Methodology -- 4 Case Study -- 5 Experiment and Results Discussion -- 6 Conclusions -- References -- Content-Based Authorship Identification for Short Texts in Social Media Networks -- 1 Introduction -- 2 Proposed Approach -- 3 Experiment -- 3.1 Data Gathering -- 3.2 Data Pre-processing -- 3.3 Word Embeddings -- 3.4 Tweet Mean Vectors and User Mean Vectors -- 3.5 Supervised Learning -- 4 Discussion -- 5 Conclusions -- References -- A Novel Pre-processing Method for Enhancing Classification Over Sensor Data Streams Using Subspace Probability Detection -- 1 Introduction -- 2 Related Work -- 3 Our Proposed Model -- 4 Experiment -- 5 Conclusions -- References -- Bio-inspired Models and Evolutionary Computation -- Remaining Useful Life Estimation Using a Recurrent Variational Autoencoder -- 1 Introduction -- 2 RUL Estimation -- 3 Model Architecture -- 4 Experimental Study -- 4.1 Data Pre-processing -- 4.2 Illustrative Example -- 4.3 Numerical Results -- 5 Concluding Remarks -- References
  • Open-Ended Learning of Reactive Knowledge in Cognitive Robotics Based on Neuroevolution -- 1 Introduction -- 2 Knowledge in the Epistemic Multilevel Darwinist Brain (e-MDB) -- 2.1 Knowledge Representation -- 2.2 Knowledge Acquisition and Decision Making -- 3 Policy Learning Process -- 4 Real Robot Example -- 5 Conclusions -- References -- Managing Gene Expression in Evolutionary Algorithms with Gene Regulatory Networks -- 1 Introduction -- 2 Biological Gene Expression -- 3 Gene Regulatory Networks -- 4 Proposed Algorithms -- 4.1 Chromosome Representation -- 4.2 Fitness Evaluation -- 4.3 Mutation -- 4.4 Crossover -- 5 Evaluation -- 6 Results -- 7 Conclusion -- 8 Future Work -- References -- Evolutionary Optimization of Neuro-Symbolic Integration for Phishing URL Detection -- 1 Introduction -- 2 Related Works -- 3 The Proposed Method -- 3.1 Collect and Structuralize Calibration Rules for Deep Learning Classifier -- 3.2 Deep Learning Calibration and Optimization to Improve URL Classification Performance -- 3.3 One-Dimensional Convolutional Recurrent Neural Network to Extract Sequence of Character and Word Features -- 4 Experimental Results -- 4.1 Data Collection -- 4.2 Comparison of Accuracy and Recall Performance -- 4.3 Classification Performance and Chi-Square Test -- 4.4 Analysis of Optimized Combination and Misclassified Cases -- 4.5 Generations for Genetic Algorithm -- 5 Conclusion -- References -- Interurban Electric Vehicle Charging Stations Through Genetic Algorithms -- 1 Introduction -- 2 Problem Description -- 3 Interurban EV Charging Stations Distribution -- 3.1 Utility -- 3.2 Uncovered Area and Distance to Nearest Station -- 3.3 Implementation and Fitness -- 4 Experimental Results -- 5 Conclusions -- References -- An Effective Hybrid Genetic Algorithm for Solving the Generalized Traveling Salesman Problem -- 1 Introduction
  • 2 Definition of the Generalized Traveling Salesman Problem -- 3 The Proposed Hybrid Algorithm -- 3.1 Chromosome Structure -- 3.2 Chromosome Optimization -- 3.3 Initial Population -- 3.4 Crossover -- 3.5 Mutation -- 3.6 Genetic Parameters -- 4 Preliminary Computational Results -- 5 Conclusions -- References -- A Simple Genetic Algorithm for the Critical Node Detection Problem -- 1 Introduction -- 2 Related Work -- 3 Maximum Components GA (MaxC-GA) -- 4 Numerical Experiments -- 5 Conclusions -- References -- Learning Algorithms -- Window Size Optimization for Gaussian Processes in Large Time Series Forecasting -- 1 Introduction -- 2 Methodology -- 2.1 Gaussian Processes -- 2.2 Kernel Selection -- 2.3 General Approach to Large Time Series -- 2.4 Data Set -- 3 Results and Analysis -- 3.1 Kernel and Parameter Choice -- 3.2 Hyperparameter Tuning and Interpretability -- 4 Conclusions -- References -- A Binary Classification Model for Toxicity Prediction in Drug Design -- 1 Introduction -- 2 Data and Methods -- 2.1 Dataset Creation -- 2.2 Data Processing -- 2.3 Proposed Model Architecture -- 3 Experiments and Results -- 3.1 Metrics -- 3.2 Scenarios -- 3.3 Results -- 4 Conclusions -- References -- Structure Learning of High-Order Dynamic Bayesian Networks via Particle Swarm Optimization with Order Invariant Encoding -- 1 Introduction -- 2 Background -- 2.1 High-Order Dynamic Bayesian Networks -- 2.2 Particle Swarm Structure Learning -- 3 Encoding and Operators -- 3.1 Natural Vector Encoding -- 3.2 Position and Velocity Operators -- 4 Results -- 4.1 Implementation -- 4.2 Experimental Comparison -- 5 Conclusions -- References -- Prototype Generation for Multi-label Nearest Neighbours Classification -- 1 Introduction -- 2 Related Work -- 3 The Reduction Through Homogeneous Clustering (RHC) Algorithm -- 4 The Proposed Multi-label RHC (MRHC) Algorithm
  • References