14th International Conference on Soft Computing Models in Industrial and Environmental Applications (SOCO 2019)

Name: 14th International Conference on Soft Computing Models in Industrial and Environmental Applications (SOCO 2019) | Seville, Spain, May 13-15, 2019, Proceedings
Brand: Springer
Price: 213.99 EUR
Availability: OnlineOnly

Seville, Spain, May 13-15, 2019, Proceedings

Francisco Martínez Álvarez Alicia Troncoso Lora José António Sáez Muñoz Héctor Quintián Emilio Corchado(Editor)

Springer (Publisher)

1st Edition

Published on 30. April 2019

XXI, 609 pages

E-Book

PDF with digital watermarking

System requirements

978-3-030-20055-8 (ISBN)

€213.99incl. 7% vat

System requirements

for PDF with digital watermarking

E-Book Single Licence

Available for download

Description

More details

Other editions

Content

Intro
Preface
Organization
General Chairs
International Advisory Committee
Program Committee Chairs
Program Committee
Special Sessions
Soft Computing Methods in Manufacturing and Management Systems
Sec7
Soft Computing Applications in the Field of Industrial and Environmental Enterprises
Sec9
Optimization, Modeling and Control by Soft Computing Techniques
Sec11
Soft Computing in Aerospace, Mechanical and Civil Engineering: New Methods and Industrial Applications
Sec13
SOCO 2019 Organizing Committee
Contents
Machine Learning
Indexes to Find the Optimal Number of Clusters in a Hierarchical Clustering
1 Introduction
2 Related Work
2.1 Clustering Methods
2.2 Validation Indexes
3 Our Proposal
3.1 Implementation
4 Experimentation
4.1 Working Environment and Datasets
4.2 Experimental Results
5 Conclusions
References
Analysis and Application of Normalization Methods with Supervised Feature Weighting to Improve K-means Accuracy
1 Introduction
2 Hypothesis and Foundations
3 Proposed Two-Stage Methodology for Normalization and Feature Weighting
3.1 First Stage: Normalization Methods
3.2 Second Stage: Feature Weighting Strategy
4 Results
5 Conclusions
References
Classifying Excavator Operations with Fusion Network of Multi-modal Deep Learning Models
Abstract
1 Introduction
2 Related Work
3 Proposed Method
3.1 Video-Based Model
3.2 Sensor-Based Model
3.3 Fusion Network
4 Experiments
4.1 Dataset and Experimental Settings
4.2 Result Analysis
5 Conclusion
Acknowledgement
References
A Study on Trust in Black Box Models and Post-hoc Explanations
1 Introduction
2 Intelligibility and Trust
2.1 Human Subject Studies and Trust Measures
2.2 Post-hoc Explanation Approaches
3 Method
3.1 Participants
3.2 Materials
3.3 Design
3.4 Procedure
4 Results
4.1 Trust Variables
5 Conclusion
References
A Study on Hyperparameter Configuration for Human Activity Recognition
1 Introduction
2 Related Work
3 Activity Recognition Overview
4 Experimental Results
4.1 The PAMAP2 Dataset
4.2 Experimental Setup
4.3 HAR Accuracy Results
4.4 Execution Time and Energy Consumption
5 Conclusion
References
A Fuzzy Approach for Sentences Relevance Assessment in Multi-document Summarization
Abstract
1 Introduction
2 Proposed Method
2.1 Preprocessing
2.2 Semantic Graph Generation
2.3 Graph Merging Process
2.4 Concepts Clustering
2.5 Fuzzy Relevance Assessment of the Sentences
2.6 Summary Construction
3 Experimental Results
4 Conclusions and Future Works
Acknowledgments
References
Online Estimation of the State of Health of a Rechargeable Battery Through Distal Learning of a Fuzzy Model
1 Introduction
2 Description of the Proposed Model
2.1 IC Curves and Analysis
2.2 Proposed Model and Learning Methodology
2.3 Fuzzy Rule-Based Model
3 Empirical Study
3.1 Experimental Setup
3.2 Numerical Results
4 Concluding Remarks
References
A Proposal for the Development of Lifelong Dialog Systems
1 Introduction and Related Work
2 Statistical Dialog Management Methodologies
3 User Intention Modeling
4 Emotional State Recognition
5 The Enhanced UAH Dialog System
6 Experiments
7 Conclusions and Future Work
References
Smart Cities and IOT
Real-Time Big Data Analytics in Smart Cities from LoRa-Based IoT Networks
1 Introduction
2 Related Works
3 System Architecture
3.1 LoRa Based Infrastructure
3.2 JSON Payload Buffering and Preprocessing
3.3 Real-Time Environment
3.4 Big Data Streaming Engine
4 Results
4.1 Dataset Construction and Linear Regression Parametrization
4.2 Experimental Setup
4.3 Analysis
5 Conclusions
References
Deep Learning in Modeling Energy Cost of Buildings in the Public Sector
Abstract
1 Introduction
2 Literature Review
3 Data and Sampling
4 Methodology
5 Results and Discussion
6 Conclusion
Acknowledgments
References
Framework for the Detection of Physiological Parameters with Musical Stimuli Based on IoT
Abstract
1 Introduction
2 Physiological Parameters, Emotions and External Stimuli
2.1 From the Theories of Emotion to Affective Computing
2.2 Physiological Parameters and Emotional States
2.3 IoT and Biosensors
3 Smoodsically. An Overview of the Framework Proposed
4 Case Study
4.1 EDA and Temperature Results
5 Conclusion and Future Work
Acknowledgments
References
Edge Computing Architectures in Industry 4.0: A General Survey and Comparison
1 Introduction
2 Internet of Things and Edge Computing
3 Edge Computing Reference Architectures
3.1 FAR-Edge RA
3.2 Edge Computing RA 2.0
3.3 Industrial Internet Consortium RA
4 Evaluation of the Edge Reference Architectures
5 Conclusions and Future Work
References
Predictive Maintenance from Event Logs Using Wavelet-Based Features: An Industrial Application
1 Introduction
2 From Event to Time Functions
3 Methods
3.1 Wavelets Transform
3.2 Random Forest
4 Experiments
4.1 Predictive Performance
4.2 Variable Importance
4.3 Observations Proximity
5 Discussion and Conclusion
References
Building Robust Prediction Models for Defective Sensor Data Using Artificial Neural Networks
1 Introduction
2 Proposed Approach
2.1 NoiseDrop
2.2 Input Drop
3 Experimental Setup
3.1 Data
3.2 Results
4 Conclusions
References
Temporal Data Analysis
Ensemble Deep Learning for Forecasting 222Rn Radiation Level at Canfranc Underground Laboratory
1 Introduction
2 Methods and Materials
2.1 Convolutional Neural Networks
2.2 Seasonal and Trend Decomposition Using Loess
2.3 Recurrent Neural Networks
2.4 Statistics
3 Experimental Results and Models Comparison
4 Conclusions
References
Search of Extreme Episodes in Urban Ozone Maps
1 Introduction
2 Methods and Materials
2.1 DBSCAN
2.2 Distance Metrics
3 Experimental Results
3.1 Outliers Detection with DBSCAN and L2 Norm
3.2 Outliers Detection with DBSCAN and L1-Norm
4 Conclusions
References
A Novel Heuristic Approach for the Simultaneous Selection of the Optimal Clustering Method and Its Internal Parameters for Time Series Data
1 Introduction
2 Proposed Harmony Search Algorithm for Optimal Clustering Configuration (HSOCC)
2.1 Encoding Solution
2.2 Steps of the HSOCC Algorithm
3 Simulation Results
4 Conclusions and Future Work
References
A Hybrid Approach for Short-Term NO2 Forecasting: Case Study of Bay of Algeciras (Spain)
Abstract
1 Introduction
2 Area and Data Description
3 Methods
3.1 LASSO
3.2 ANNs
4 Experimental Procedure
5 Results and Discussion
6 Conclusions
Acknowledgements
References
Context-Aware Data Mining vs Classical Data Mining: Case Study on Predicting Soil Moisture
1 Introduction
1.1 Related Work
1.2 Context-Aware DM vs Classical DM Concepts
2 Experimental Setup
2.1 Description (Reading) of the Existing Data
2.2 Preprocessing the Data
2.3 Preliminary Decisions Before Implementing the Data Mining Processes
3 Experiment Implementation and Results
3.1 Classical DM vs CADM Process Implementation
3.2 DM vs CADM Results
4 Conclusions
References
DTW as Alignment Function in the Context of Time Series Balancing
1 Introduction
2 The Proposal
2.1 The TS_SMOTE Algorithm
2.2 The Distance Functions and Alignment Issues in Balancing TS Problems
2.3 Building an Unaligned Dataset
3 Experiments and Results
3.1 Materials and Methods
3.2 Numerical Results
4 Conclusions
References
Feature Clustering to Improve Fall Detection: A Preliminary Study
1 Introduction
2 Peak Detection and Feature Extraction
3 Data Modeling and Classification
4 Experimental Design
4.1 Dataset Description
4.2 Cross Validation
5 Obtained Results and Discussion
6 Conclusions
References
Data Generation and Preparation
Creation of Synthetic Data with Conditional Generative Adversarial Networks
1 Introduction
2 Methodology
2.1 Generative Adversarial Networks
2.2 Conditional Adversarial Networks
2.3 Dataset
2.4 Software and Experimental Setting
3 Results
3.1 Generating New Credit Card Data with CGANs
3.2 Similarity of the Data
3.3 Classification Results
4 Discussion
References
Data Selection to Improve Anomaly Detection in a Component-Based Robot
1 Introduction and Previous Work
2 Anomaly Detection
2.1 Support Vector Machines
2.2 Metrics
3 Experiments and Results
3.1 Dataset
3.2 Results on the Whole Dataset
3.3 Results on Trial 21
4 Conclusions and Future Work
References
Addressing Low Dimensionality Feature Subset Selection: ReliefF(-k) or Extended Correlation-Based Feature Selection(eCFS)?
1 Introduction
2 Background
3 The Proposed Approach
4 Experimentation
5 Results
6 Conclusions
References
A Predictive Maintenance Model Using Recurrent Neural Networks
1 Introduction
2 Background
3 Case Study
4 LSTM Architecture
5 Results
5.1 Network Classifier
5.2 Regressive Network
5.3 Final Model
6 Conclusions
References
Soft Computing Applications
Prototypical Metric Transfer Learning for Continuous Speech Keyword Spotting with Limited Training Data
1 Introduction
2 Related Work
3 Dataset
4 Approach
4.1 Data Preprocessing
4.2 Feature Engineering
4.3 Deep Learning Architectures
5 Experiments and Results
6 Conclusion
References
Characteristic of WiFi Network Based on Space Model with Using Turning Bands Co-simulation Method
1 Introduction
2 Related Work
3 Turning Bands Method
4 Characteristic of PWR-WiFi Network
5 Preliminary and Structural Analysis of Data
6 Space Model of PWR-WiFi Network Efficiency Made by Co-simulation TBM
7 Conclusions
References
Inconsistency Detection on Data Communication Standards Using Information Extraction Techniques: The ABP Case
Abstract
1 Introduction
2 Background and Related Works
3 Research Method
3.1 Initial Works on the Text Sentences: Tagging, Setting Chunking, Heuristic Split and Parsing
3.2 Searching and Extraction Algorithms
4 Results
4.1 Results Obtained from the ABP_Standard_Text
4.2 Results Obtained from the ABP_NO_Standard_Text
5 Conclusions
6 Future Works
References
Mobile Architecture for Forest Fire Simulation Using PhyFire-HDWind Model
Abstract
1 Introduction
2 Background
2.1 PhyFire-HDWind Models
3 Proposed System
3.1 Input Spatial Data
3.2 Pre-process
3.3 Process
3.4 Post-process and Results Display
4 Results and Conclusions
Acknowledgments
References
A Proposal of Robust Leak Localization in Water Distribution Networks Using Differential Evolution
1 Introduction
2 Model-Based Leak Localization Methods in WDN
2.1 Leak Detection and Localization Methods Based in LSM
2.2 Leak Localization Methods Using Optimization Algorithms
2.3 Topological Differential Evolution (T-DE)
3 Hanoi Network
4 Experiments and Results
5 Conclusions
References
Neural Model of a Specific Single Proton Exchange Membrane PEM Fuel Cell
1 Introduction
2 Background
2.1 PEM Fuel Cells
2.2 Artificial Neural Networks
3 Experimental Design
3.1 Commercial PEM Fuel Cell Description
3.2 Data Gathering Process
3.3 Modeling Process
4 Experimental Results
5 Conclusions
References
Special Session - Soft Computing Methods in Manufacturing and Management Systems
A Hybrid Heuristic Algorithm for Multi-manned Assembly Line Balancing Problem with Location Constraints
Abstract
1 Introduction
2 Multi-manned Assembly Line's Workstations with Location Constraints
3 mLALBP Model
3.1 Problem Definition
3.2 Problem Assumptions
3.3 Model Notation
4 The Proposed Hybrid Heuristic Method
5 An Illustrative Example
6 Summary
References
A Comparison Analysis of the Computer Simulation Results of a Real Production System
Abstract
1 Introduction
2 The Model of the Production System
2.1 The Simulation Model of the Production System Built with FlexSim
2.2 The Simulation Model of the Production System Built with Plant Simulation
3 The Comparative Analysis
4 Summary
References
Multiple Fault Diagnosis in Manufacturing Processes and Machines Using Probabilistic Boolean Networks
Abstract
1 Introduction
2 Probabilistic Boolean Networks and Their Use in Manufacturing Systems
3 PBNs for FDI in Manufacturing Systems
4 Experimental Results
5 Conclusions
Acknowledgements
References
Concurrent Planning and Scheduling of Heterogeneous Production System. Case Study
Abstract
1 Introduction
2 Methodology
3 Case Study
3.1 Production Order Description
4 Summary
References
Multi-domain, Advisory Computing System in Continuous Manufacturing Processes
Abstract
1 Introduction
2 Analysis of Loss of Signal Quality at Selected Points of the Computer System
3 Concept of a Multi-domain System
4 Summary
References
Assessment of Similarity of Elements as a Basis for Production Costs Estimation
Abstract
1 Introduction
2 Method of Description of Elements
2.1 Description of Shaft's Technological and Structural Features
2.2 Description of Shaft's Elementary Functional Surfaces
3 Assessment of Element's Similarity
4 Example of Method's Application
5 Summary
References
Special Session - Soft Computing Applications in the Field of Industrial and Environmental Enterprises
Outlier Generation and Anomaly Detection Based on Intelligent One-Class Techniques over a Bicomponent Mixing System
1 Introduction
2 Case of Study
2.1 Bicomponent Mixing System
2.2 Dataset Description
3 Techniques Applied to Validate the Proposed Model
3.1 One-Class Techniques
3.2 Artificial Outlier Generation
4 Experiments and Results
4.1 Approximate Convex Hull Classifier
4.2 Artificial Neural Network Autoencoder Classifier
4.3 SVM Classifier
5 Conclusions and Future Works
References
Material Flow Optimization Using Milk Run System in Automotive Industry
Abstract
1 Introduction
2 Material Flow and Related Work
3 Modelling in Material Flow
3.1 Particle Swarm Optimization Algorithm
3.2 Collection of Input Data
4 Experimental Results and Discussion
4.1 Milk Run System Implementation
4.2 Discussion, Benefits and Limitation of Milk Run System
5 Conclusion and Future Work
References
Smart PPE and CPE Platform for Electric Industry Workforce
1 Introduction
2 Related Work
3 Proposed Method
4 Results
5 Conclusion and Future Work
References
Acoustic Anomaly Detection Using Convolutional Autoencoders in Industrial Processes
1 Introduction
2 Related Work
3 Acoustic Anomaly Detection System
3.1 Feature Extraction
3.2 Convolutional Autoencoder
3.3 One-Class Support Vector Machine
4 Experiments and Results
4.1 Experimental Setup
4.2 Evaluation Criteria
4.3 Results
5 Conclusion
References
One-Class Classification to Predict the Success of Private-Participation Infrastructure Projects in Europe
Abstract
1 Introduction and Previous Work
2 Applied Classifiers
2.1 Support Vector Machines
2.2 k-Nearest Neighbor
2.3 Random Forest
3 Experiments and Results
3.1 Dataset
3.2 Results from the Energy Sector
3.3 Results from the Telecommunication Sector
4 Conclusions
References
Optimizing a Bi-objective Vehicle Routing Problem Appearing in Industrial Enterprises
1 Introduction
2 Multi-start Multiobjective Local Search Procedure
2.1 Phase 1. Solution Representation and Construction
2.2 Phase 2. Local Search
3 Computational Results
4 Conclusions
References
An Industrial Application of Soft Computing for the Design of Personalized Call Centers
1 Introduction and Background
2 Let's Go System
3 Experimental Set-Up
4 Discussion of the Experimental Results
5 Conclusions and Future Work
References
A Preliminary Study on Multivariate Time Series Clustering
1 Introduction
2 A Proposal for Multivariate TS Clustering
2.1 Finding Similarities Between Variables from an Example
2.2 RNN and Transfer Learning
2.3 Computing the Similarities Within a Multivariate TS Dataset
3 Experiment and Methods
4 Results and Discusion
5 Conclusions
References
Adaptive Fault-Tolerant Tracking Control Algorithm for IoT Systems: Smart Building Case Study
1 Introduction
2 Related Work
3 Proposed Control Model
3.1 Reference Input
3.2 State Predictor
3.3 Cooperative Control Algorithm
4 Results
5 Conclusions
References
Special Session - Optimization, Modeling and Control by Soft Computing Techniques
Low Voltage Grid Operation Scheduling Considering Forecast Uncertainty
1 Introduction
2 Problem Formulation
2.1 Problem Objectives
3 Solution Approach
3.1 Objective Function
3.2 Simulation Algorithm
4 Experimentation
4.1 Set up
4.2 Results
5 Conclusions
References
Iterative Learning Control for a Hydraulic Cushion
1 Introduction
2 Force Controller Design
3 Iterative Learning Control Design
4 Simulation Results
5 Conclusions
References
Opinion Mining to Detect Irony in Twitter Messages in Spanish
1 Introduction
2 Motivation and Related Work
3 Methodology and Experimentation
3.1 Building of Training Corpus
3.2 Pre-processing
3.3 Data Set
3.4 Feature Extraction
3.5 Vectorization or Bag-of-Words
3.6 Classification Algorithm
4 Results Evaluation and Discussion
4.1 Discussion
5 Conclusion and Future Work
References
An Efficient Soft Computing Approach for Solving the Two-Stage Transportation Problem with Fixed Costs
1 Introduction
2 Definition of the Problem
3 Description of the Proposed Soft Computing Approach
4 Computational Results
5 Conclusions
References
Takagi-Sugeno Fuzzy Incremental State Model for Optimal Control of a Ball and Beam Nonlinear Model
1 Introduction
2 Ball and Beam Nonlinear Model
3 Fuzzy Takagi-Sugeno Model and System Identification
3.1 Fuzzy T-S Model
3.2 Estimation of T-S Model Parameters
4 Incremental State Model
5 Fuzzy Controller and Observer Design Based on Incremental State Model
5.1 Fuzzy Controller for Zero Steady-State Error
5.2 Fuzzy Observer for Incremental State Model
6 Results
7 Conclusion
References
Time-Oriented System to Control Critical Medications
1 Introduction
2 System for Controlling Time Restrictions of Medicines
2.1 Design to Store Time Restrictions in Medical Treatments
2.2 Crawling the Web
2.3 Parsing the Web Content
3 Example of Use
3.1 Case of Use with the Active Principle Fluoroquinolone
4 Conclusions and Future Work
References
Special Session - Soft Computing in Aerospace, Mechanical and Civil Engineering: New Methods and Industrial Applications
An Introduction to Some Methods for Soft Computing in Fluid Dynamics
1 Introduction to Turbulent Flows and Big Data
2 Soft Computing in Fluid Dynamics
2.1 Singular Value Decomposition: POD and PCA
2.2 Dynamic Mode Decomposition
2.3 Spectral POD, Multi-scale Analysis and EMD
3 Machine Learning: POD and DMD
4 Concluding Remarks
References
A Data-Driven ROM Based on HODMD
1 Introduction
2 Construction of the Data-Driven ROM Based on HODMD
3 Results for the Complex Ginzburg-Landau Equation
4 Conclusions
References
Soft Computing Techniques to Analyze the Turbulent Wake of a Wall-Mounted Square Cylinder
1 Introduction
2 Methodology
2.1 Proper Orthogonal Decomposition
2.2 Spectral POD
2.3 Dynamic Mode Decomposition
2.4 Multi-scale Analyses: POD and DMD
3 Data Analysis
4 Conclusions
References
Generating Three-Dimensional Fields from Two-Dimensional Soft Computing Strategies
1 Introduction
2 Spatio-Temporal Higher Order Dynamic Mode Decomposition
2.1 HODMD Algorithm
3 Methodology for Constructing Three-Dimensional Approximations from a Set of Planes
4 Definition of the Three-Dimensional Toy Model
5 Reconstruction of the Three-Dimensional Fields
6 Conclusions
References
Low Cost Methods for Computing Instabilities in Boundary Layer Flows
1 Introduction
2 Theory
2.1 Streaky Base Flow Formulation
2.2 Stability Analysis
3 Results
3.1 Stability of Optimal Streaks
3.2 Stability of Streamwise Vortices
4 Conclusions
References
Author Index

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

14th International Conference on Soft Computing Models in Industrial and Environmental Applications (SOCO 2019)

Description

More details

Other editions

Additional editions

Content

System requirements