Organisers of Special Sessions are responsible for:
- Select a topic of interest to conference delegates.
- Obtain papers on this topic, normally a minimum of 5 for an invited special session, but often more. At least 60% of the papers must be by authors that are neither session chairs, from their team nor reviewers for the session.
- If there are not sufficient papers, final accepted papers will be moved to the general track.
- Manage the review process for these papers on due time and deadlines.
- Provide suitable reviewers for the reviews of the papers.
- Ensure the final versions of the papers are uploaded before the deadline.
- Attend the conference and chair the session.
- Provide a list of international reviewers (name, affiliation, country) who have already accepted to review the papers.
- Disseminate widely a call for papers for the special session.
Special Session 1
Special Session on Machine Learning and Computer Vision in Industry 4.0
Enrique Dominguez, University of Málaga
Jose Garcia Rodriguez, University of Alicante
Ramon Moreno Jiménez, Grupo Antolin
In the coming years, the use of machine learning and computer vision in industry is a trend that is changing not only large corporations, but also small and medium-sized businesses. Thanks to these technologies, the innovation in the industrial sector is giving rise to the named “smart factories”, allowing them to obtain multiple advantages.
This special session tries to provide a common platform for academics, developers, and industry-related researchers to discuss, share experiences and explore the new technological advances. The objective is to integrate an international scientific community working on industrial applications of machine learning and computer vision for fruitful discussions and ideas on the evolution of these technologies.
Special Session 2
Crossing the border between Soft Computing and Machine Learning
Giuseppe Psaila, University of Bergamo
Paolo Fosci, University of Bergamo
Soft Computing and Machine Learning have provided a large variety of tehcniques to process
large volumes of data in a way that traditional
crisp computing is not able to provide.
In fact, approximate matching, flexible models and
machine learning are able to achieve incredible results in many application scenarios.
However, what is exactly soft computing? and what is exactly machine learning?
Is it possible to identify a border between them?
Can soft computing provide the methods to explain machine learning?
The special session solicits papers that study (and possibly identify) the border between soft computing and machine learning (giving clear definitions of what soft computing is and what machine learning is), as well as papers that try to
jointly applly soft computing and machine learning, to “cross the broder” and adopt them in a sinergic way.
Papers that foster the discussion on the topic are wlecome.
Position papers are welcome.
Topics of interest.
The topics of interest for the special session encompass (but are not limited to) the following ones.
– Characterizing the border between softo computing and machine learning.
– Machine-learning techniques based on fuzzy sets.
– Differences between Machine learning and Deep learning.
– Neural Network: architectures, models and meaning of trained models.
– Differences among neural-network techniques.
– The rationale behind neural-network architectures.
– Fuzzy sets and soft computing used to describe models produced by machine-learning techniques.
– Imprecision and uncertainty in databases and datasets.
– Possibilistic approaches to machine learning.
– Comparisong of probnabilistic approaches and possibilistic approaches in soft computing and/or machine learning.
– Soft querying and soft information retrieval.
– Soft computing for explainable word embedding.
Special Session 3
Time Series Forecasting in Industrial and Environmental Applications (TSF)
Federico Divina, Pablo de Olavide University
Mario Giacobini, University of Torino
José Luis Vázquez Noguera, Universidad Nacional de Asunción
Miguel García Torres, Pablo de Olavide University of Seville
José F. Torres, Pablo de Olavide University of Seville
Time series can be found in almost all disciplines nowadays. Thus, time series forecasting is becoming a consolidated discipline that provides meaningful information in a wide variety of application areas, turning their efficient analysis into the utmost relevance for the scientific community.
This session pays attention to the extraction of useful knowledge from time series in the context of industrial and environmental applications. The analysis of very large time series, given its relevance in the emergent context of big data, is also encouraged. Topics of interest for the special session, always in the context of industrial and environmental applications, include but are not limited to:
1. Machine learning applied to time series forecasting.
2. Deep learning applied to time series forecasting.
3. New approaches for big data time series forecasting.
4. Hybrid systems for time series analysis.
5. Ensemble approaches for time series analysis.
Special Session 4
Optimization, Modeling and Control by Soft Computing Techniques (OMCS)
Eloy Irigoyen, University of the Basque Country
Matilde Santos Peñas, Complutense University of Madrid
José Luis Calvo Rolle , University of A Coruña
María Tomas-Rodríguez, University of London
Mikel Larrea Suika, University of Basque Cuntry
Nowadays, regardless any context, due to different reasons as climate change, aging population, and new engineering and industrial requirements, it is necessary to achieve some key challenges such as: minimize energy consumption and emissions to the atmosphere, improve the quality of life of humans, or reach industrial enhancement requests. People involved with industry and academic can not be oblivious to these facts, because that will determine the future. In this sense, the industry and some other fields like buildings management, assistive technologies, or real engineering applications play an important role into the different emerging techniques developed with the aim to achieve the previously cited objectives. Obviously, traditional techniques have met to the demands so far, but for future, new advanced improvements are needed.
This session provides an interesting opportunity to present and discuss the latest theoretical advances and real-world applications in Optimization, Modeling and Control Systems by means of Soft Computing models, including among others, the following topics: ç
• Energy efficiency and optimization.
• Control Techniques efficiency and optimization.
• Traditional systems improvement.
• Industrial control new techniques.
• Modeling of complex systems.
• Process optimization new techniques.
• Fault Detection and Diagnosis.
• Techniques to improve robustness against system failure.
• Computational intelligence developments aimed to human beings.
• Engineering and industrial soft computing applications.
Special Session 5
Special Session on Soft Computing Applied to Renewable Energy Systems
Jesús Enrique Sierra García, Universidad de Burgos
Matilde Santos Peñas, Complutense University of Madrid
Pawel Martynowicz, AGH University of Science and Technology
Fares M´zoughi, University of Basque Cuntry
Payam Aboutalebi, University of Basque Cuntry
Abstract Coal-fired power plants have been identified as one of the significant causes of climate change. Although CO2 emissions are mainly produced by thermal power plants, these energy sources are still widely used nowadays. As a result, there is a widespread consensus that renewable energy sources such as wind, marine, hydro, and solar must be considered to mitigate climate change and reduce air pollution. Consequently, research on renewable energies and, particularly, on control and efficiency is encouraged to contribute to this sustainable trend. Expert systems, fuzzy control, neural networks, genetic algorithms, artificial immune networks, swarming particle techniques, ACO, reinforcement learning, and other soft computing techniques have been shown to be effective in many different fields. They can be applied to tackle complex problems where conventional methods are less efficient or unsuccessful.
The aim of this special session is to provide a platform for researchers, engineers, and industrial professionals from different fields to share and exchange their ideas, research results and experiences in the field of soft computing techniques applied to renewable energy systems. Contributions to this special session are welcome to present and discuss novel methods, algorithms, frameworks, architectures, platforms, and applications.
Topics Session topics include, but are not limited to, the following strategies and approaches applied to renewable energy systems:
• Intelligent control: fuzzy control, neuro-control, neuro-fuzzy, intelligent-PID control, …
• Optimization by heuristic techniques in system engineering and control
• Modelling and identification by Soft Computing techniques
• Identification and control by hybrid intelligent strategies
Special Session 6
Special Session Soft Computing Methods in Manufacturing and Management Systems
Damian Krenczyk, Silesian University of Technology (Poland)
Anna Burduk, Wroclaw University of Science and Technology (Poland)
Bożena Skołud, Silesian University of Technology (Poland)
Wojciech Bożejko, Wroclaw University of Science and Technology (Poland)
Marek Placzek, Silesian University of Technology (Poland)
Abstract: Management of manufacturing systems involves development of detailed solutions related to decision making and problem solving processes. There are many important decisions to be taken and high complexity problems to solve (NP-hard), related to e.g. processes organization, planning and control of manufacturing systems. Special attention is paid to inexact solutions for which there is no known algorithm that can obtain an exact solution in polynomial time. The aim of this session is to present the results of research related to management of production systems. Taking into account the complexity of problems related to production management, soft computing and intelligent methods may deliver the most adequate answers. Topics session should be related to application of soft computing methods and tools to problem solving in:
• Manufacturing Systems Integration
• Optimization of Manufacturing Systems
• Modelling and Design
• Control and Supervision
• Industry 4.0
• Production Planning and Scheduling
• Virtual Organisation
• Data Mining and Data Recognition
• Production System Organization
• Production Management
• Discrete Optimization
• Line Balancing
• Parallel Algorithms
• Artificial Intelligence
Special Session 7
On pre-processing big data in machine learning
Antonio J. Tallón-Ballesteros, University of Huelva (Spain)
Simon Fong, University of Macau (Macau SAR)
Luís Cavique, Universidade Aberta (Portugal)
Abstract: The Knowledge Doubling Curve was proposed by Richard Buckminster Fuller almost forty years ago. It states that the (computational) knowledge is doubled every 18 months. Currently, pre-processing big data is a step of utmost importance in machine learning applications. Yottabyte is up to now the highest magnitude defined to store the information. Nonetheless, in some cases to process amount of data higher than two gigabytes is almost computationally intractable even if the application runs with thirty two gigabytes of main memory in an advanced computer. Clearly, we need new paradigms to transform the raw data into input data for any machine learning task. Plain text files may not be opened using any text-editor and this is why a built-in data editor is very a convenient for the big data analyst. Data are overwhelming and the research value is just to be able to capture the most important information in order to predict with a high performance and very quickly. To process the data we need to conduct a machine learning task which may include different kinds of pre-processing. This special session welcomes works concerning real-world applications where data pre-processing in machine learning plays an important role. The scope may be concerning supervised, unsupervised or semi-supervised tasks to prepare the data. The topics of interest for this thematic session comprise, but are not limited to: – Data selection – Feature selection – Outlier detection – Outlier removal – Noise smoothing – Instance selection – Random projection – Informative projection – Data normalization – Text-based user interfaces to pre-process big data.
Special Session 8
Special Session on Tackling Real World Problems with Artificial Intelligence
Enrique de la Cal, University of Oviedo (Spain)
José R. Villar, University of Oviedo (Spain)
Beatriz de la Iglesia, University of East Anglia (UK)
Noelia Rico, University of Oviedo (Spain)
Fernando Moncada, University of Oviedo (Spain)
Samad B. Khojasteh, University of Oviedo (Spain)
Enol García González, University of Oviedo (Spain)
Víctor M. González, University of Oviedo (Spain)
Víctor M. Álvarez, University of Oviedo (Spain)
Mirko Fáñez, University of Oviedo (Spain)
Alberto Gallucci, University of Oviedo (Spain)
Paloma Valverde, Technological Institute of Castilla y León (Spain)
Petrica Pop, Technical University Cluj Napoca (Romania)
Abstract: In recent years, society has perceived the impact and relevance of Artificial Intelligence in everyday life, and so does the economy. Not only the scientific media has echoed the AI good performance: the industry and corporates have positively considered the inclusion of AI in their production systems. Therefore, a wide variety of AI applications have been designed and developed to cope with all this plethora of problems. This special session focuses on this issue, pursuing to show how many different possibilities AI can offer in solving real-world problems.