Special Session 1: Fuzzy Systems and Intelligent Data Analysis


Data and text have increased immeasurably in most science, engineering, and societal domains which brings proper challenges for extracting valuable information from data. The volume, variety, velocity, and noise related veracity of data are often the causes of inherent uncertainty in representing and analysing data.

Fuzzy systems have proven to be a powerful approach providing efficient techniques, flexible and interpretable methods to deal with the imprecision and uncertainty in data thanks to the processing power and linguistic expressiveness of fuzzy sets. Also, data analysis plays a key role on deriving new concepts and new relations between concepts opening new possibilities to analyse data at higher semantic levels.

The aim of this session is to offer a forum for both academic and industrial communities to share and disseminate innovative research efforts and developments on fuzzy systems and intelligent analysis contributing to tackling the challenges to extract value from heterogeneous, high dimensional, noisy or otherwise uncertain data in real-world problems.

The topics include but are not limited to:

• Granular modelling, classification, and regression
• Fuzzy clustering
• Neurocomputing, fuzzy neural nets, and deep learning
• Fuzzy evolutionary computation
• Visualization
• Methods to deal with uncertainty and interpretability in data processing
• Artificial Intelligence for uncertain environments


Susana Nascimento (NOVA University, Lisboa, Portugal)
Antonio J. Tallón-Ballesteros (University of Huelva, Huelva, Spain)
José Valente de Oliveira (University of Algarve, Faro, Portugal)
Boris Mirkin (National Research University, HSE, Moscow, Russian Federation)

Special Session 2: Machine Learning towards Smarter Multimodal Systems


Multimodal Systems research is focused in methods and tools “to create, access, and interact all forms of digital content in any device or scenario”. Multimodal human-computer interaction is related to the user interaction with the virtual and physical scenario through the most natural modes of communication. Nowadays, with the advent of Ubiquitous Computing (Ubicomp), which focuses on a human-centred paradigm, multimodal systems aim to provide interaction with adaptive content, services, and interfaces towards each one of its users, according to the context of the applications’ scenarios. However, the provision of that appropriated personalised interaction is a true challenge due to different reasons, such as the user interests, heterogeneous environments and devices, or the dynamic user behaviour and data capture.
Machine Learning (ML) algorithms analyse historical data, build models & predict outcomes. Machine Learning has got vast industry transformation capabilities and is used across sectors for business efficiency. Applications that provide large amounts of data regarding the users' interaction, such as usage data, preferences or even context of usage, should use machine learning to reason about the users. It should be used to acquire models of individual users interacting with the information system and grouping them into communities or stereotypes with common interests. In this regard, ML techniques can be excellent tools to cope with difficult problems that arise when implementing smarter multimodal systems, ranging from representation, translation, alignment or fusion between modalities, especially for Ubicomp scenarios. There are many applications which can be tackled by ML techniques, such as the personalization of the user interaction, the distribution of the user interface across different devices, or new user-interface approaches, towards intelligent user interactions with multimodal systems to better reach the Ubicomp vision.
This special session aims to cover a wide range of works and recent advances on the application of ML techniques to enhance the user interaction in Multimodal Systems. We hope that this session can provide a common forum for researchers and practitioners to exchange their ideas and report their latest findings in the following topics.


We encourage submissions addressing:
• Intelligent User Interaction.
• Personalization of the User Interaction.
• Recommendation in multimodal systems.
• Modelling human behavior.
• Machine Learning for Accessibility.
• Prediction of future users’ activities and interactions.
• Distribution of the User interface across different devices.
• Novel user-interface techniques.
• Machine Learning for healthcare applications.
• Machine Learning for Creativity.
• Machine Learning towards Ubicomp systems.
• Any application of ML techniques to HCI problems.


Prof. Dr. Nuno Correia, NOVA LINCS, DI, FCT, NOVA University of Lisboa, Portugal,
Prof. Dr. Rui Neves Madeira, NOVA LINCS, DI, FCT, NOVA University of Lisboa and Sustain.RD center, ESTSetúbal, Polytechnic Inst. Setúbal, Portugal,
Prof. Dr. Susana Nascimento, NOVA LINCS, DI, FCT, NOVA University of Lisboa, Portugal,

Special Session 3: Data Selection in Machine Learning


Data selection focuses on reducing the training time and, at the same time, taking advantage to do better predictions. Too much information is not handy at all since uninformative samples or features may be learnt and consequently the ability to generalize could be hindered. Addressing any problem may mean not having prior knowledge and even to become able, through data selection and even transformation measure, to learn the important data for the forthcoming prediction on unseen data. Depending on the followed methodology to conduct the process model for data mining, the data selection may be named with different names although the core is the same. Tools based on graphical user interfaces are of particular interest in the sense that may make easier the procedure to refine the raw data and eventually to get the ready data to face the mining phase. Data pre-processing deals with many tasks such as data cleansing, attribute selection, instance selection, noise reduction and detecting wrong or distorted labels. Visual data analytics is on the rise especially in multi-dimensional business applications. We encourage to submit very recent applications and if possible unprecedented. Additionally, new theoretical or empirical approaches are welcome.


The topics of interest for this session include, but are not limited to:
- Data selection
- Data pre-processing
- Data cleansing
- Data engineering
- Attribute selection
- Instance selection
- Data fusion
- Data mining
- Text mining
- Speech mining
- Signal mining
- Stream mining
- Motif mining
- Itemset mining
- Sequential pattern mining
- Frequent pattern mining
- Infrequent pattern mining
- Rare pattern mining


Antonio J. Tallón-Ballesteros (University of Huelva, Spain)
Raymond Kwok-Kay Wong (University of New South Wales, Australia)
Ireneusz Czarnowski (Gdynia Maritime University, Poland)
Simon James Fong (University of Macau, China)

Special Session 4: Machine Learning in Healthcare


Machine learning researchers and data scientist face several challenges when modeling healthcare data. Firstly, there is the usual need for interpretable healthcare data models. Machine learning algorithms are usually designed to optimize a mapping between inputs and outputs without providing too much detail of how and why this is done (they tend to work as black boxes). This is a serious issue that keeps many machine learning models from being used for healthcare. For instance, clinicians are not only interested in a model that predicts extremely well the risk of a certain medical condition, but also in knowing what the model can tell them about what to do in order to reduce the risk. Secondly, healthcare data is usually extracted from several sources that are not necessarily fully compatible between them. For instance, data may come from a diverse of clinical studies that differ in the number and/or type of variables, and/or from clinical centers with different protocols. Thirdly, representation is another important challenge for healthcare data modeling with. Data typically come in different modalities such as images, time series, and relational data. However, many machine learning algorithms work under the assumption that datasets are “rectangular”, matrix-like, with rows representing data instances and columns, input variables. Many of them work better if further restrictions are imposed on data representation, such as the number of rows being much larger than the number of columns. Forcing healthcare data into a rectangular shape has the negative consequences of losing information, increasing the risk of other issues (additional missing values, a higher number of variables, misrepresentation, etc.) and, crucially, undermining model interpretability.

This special session aims at promoting the use of new machine learning developments for understanding healthcare data. We particularly encourage the submission of papers that deal with any of the aforementioned challenges. However, any aspect of machine learning research and application for healthcare will be welcome.

Topics of interest for this session include, but are not limited to:

* Novel machine learning algorithms aimed at modeling healthcare data.
* Comparative assessment of several machine learning algorithms used to model a healthcare task.
* Description of annotated and/or curated datasets large enough to be potentially used in new machine learning experiments for understanding healthcare data, if they are made accessible to the scientific community.
* New software tools specifically designed to analyze healthcare data using machine learning methods.


Dr Ivan Olier (Liverpool John Moores University, U.K.)
Dr Sandra Ortega-Martorell (Liverpool John Moores University, U.K.)
Prof Paulo Lisboa (Liverpool John Moores University, U.K.)
Dr Alfredo Vellido (Universitat Politècnica de Catalunya, Spain)

Special Session 5: Machine Learning in Automatic Control


Over the past two decades, an explosion of data is emerging from the physical world and it requires a rapprochement of areas such as machine learning, control theory, and optimization. The availability and scale of data, both temporal and spatial, brings a wonderful opportunity for our community to both advance the theory of control systems in a more data-driven fashion, as well as have a broader industrial and societal impact.

Interest in Machine Learning methodologies has been increasing exponentially. The objective of this special session is to focus on the inter-connection between Machine Learning and a broad and important discipline – Automatic Control. The intersection of these domains has the potential for very interesting and challenging research. Machine Learning involves the use of statistical techniques to enable computer systems to “learn” using data, typically using offline training methods. Control systems, on the other hand, focus on methods for automated regulation and tracking in engineering systems and industrial applications. Offline and online learning of the underlying control parameters have been analyzed extensively, over the past four to five decades, in sub-disciplines in control such as system identification and adaptive control. It is in these subdisciplines whereby parameter estimation occurs, either implicitly or explicitly, and thus an obvious intersection with ML arises.


The goal of this special session is to understand the aforementioned intersections between ML and automation. There are various challenges on the interface between the control community and the machine learning community. In particular, the special session goals are:

• Present state-of-the-art results in the theory and application of ML in automatic control, including topics such as statistical learning for control, reinforcement learning and deep neural networks in control, evolutionary strategies for control, data-driven control, adaptive control, dual control, online learning, active learning for control, model learning, and applications of learning control, etc.
• Bring together some of the leading researchers across the fields in order to promote cross-fertilization of results, tools, and ideas, and stimulate further progress in the area.
• Attract new researchers in these exciting problems, creating a larger yet focused community that thinks rigorously across the disciplines and ask new questions.

This special session will facilitate intensive discussion and exchange of ideas about ongoing and future research related to how learning theory can be applied to control problems.


• Matilde Santos, University Complutense of Madrid,
• Juan G. Victores, University Carlos III of Madrid,

Special Session 6: Finance and Data Mining


This special session aims at encompassing the latest innovations about data mining in the context of Finance. Therefore, the presentation of works tackling theoretical issues and applications, from industry or academia, on data mining is welcome. The finance may be focused on short, medium or long-term period; all of them have room in this session. The problem complexity is increasing due to the huge amount of transactions that are happening at a specific instant all over the world and is not very easy to detect the causality among all of them independently of the very distant point wherever the operations could take place.


Topics of interest for this session include but are not limited to:

• Finance
• Data mining
• Economical prediction
• Stock-market analysis
• Markets on the rise
• Reputation on finance
• Economist intelligence
• Money laundering avoidance
• Early detection of tax evasion
• Technological issues affecting the economy of a company
• Big numbers in the economy of a country
• Contemporary trade
• Business intelligence


• Fernando Núñez Hernández, University of Seville, Spain (
• Antonio J. Tallón-Ballesteros, University of Huelva, Spain (
• Paulo Vasconcelos, University of Porto, Portugal (
• Ángel Arcos-Vargas, University of Seville, Spain (

Special Session 7: Knowledge Discovery from Data


This special session aims to join the contemporary innovations about knowledge discovery and data mining. Introductory work for coping with theoretical matters as well as applications, from industry or academia, on machine learning and more broadly on data mining is welcome. The complexity of the data at hand is increasing and there is a good number of approaches to deal with it. A number of factors may also be considered for the suitable choice of a concrete learner.


The topics of interest for this session include, but are not limited to:
- Data mining
- Machine learning
- Computational intelligence
- Rough sets
- Big data
- Non-stationary data
- Data fusion
- Multi-label learning
- Deep learning
- Real-world intelligent applications
- Bioinformatics
- Signal mining
- Mining databases
- Granular computing
- Cloud computing
- Swarm intelligence
- Web intelligence
- Hybrid intelligent systems
- Complex and biological inspired systems
- Bio-inspired computation, evolutionary computing, evolutionary programming and evolutionary strategies
- Meta-Heuristics: theory and foundations
- Time series
- Emerging approaches: edge computing, fog computing

- Barbara Pes, University of Cagliari, Italy (
- Antonio J. Tallón-Ballesteros, University of Huelva, Spain (
- Julia Handl, The University of Manchester, UK (

Special Session 8: Machine Learning Algorithms for Hard Problems


Difficulties associated with real data pose many challenges for classical pattern recognition algorithms like Gaussian Naive Bayes or Support Vector Machines. The quality of the models built from the real data may strongly depend on the scale of the class imbalance, the variability of the concepts in time, large number of problem classes, high spatiality or a huge number of or very few patterns available for training. We often encounter datasets that are characterized by data incompleteness or concept drift, to enumerate a few.

The main aim of this Special Session is to bring together researchers and scientists who are pioneering difficult data analysis methods, as well as in wider humanitarian fields, to discuss problems and solutions in this area, to identify new issues, and to shape future directions for research.


Topics of interest for this Special Session include but are not limited to:

  • class imbalanced learning
  • learning from data streams
  • learning in the presence of concept drift
  • learning with limited ground truth access
  • learning from high dimensional data
  • learning on the basis of limited data sets, including one-shot learning
  • instance and prototype selection
  • data imputation methods
  • case studies and real-world applications affected by data difficulties

Workshop Organizers:

Special Session 9: Reconfigurable Computing for Intelligent Data Engineering


Many modern applications in data engineering and machine learning demand huge computational resources and require substantial amounts of energy. Reconfigurable hardware technology allows for the implementation of application-specific accelerators that can achieve great improvements in performance and energy efficiency through customized operators and interconnections, applications-specific wide data paths and memory architectures, as well as massively parallel execution and deep pipelines.

The aim of this session is to bring together researchers and engineers working on reconfigurable accelerators for data engineering and machine learning with the goal to discuss and exchange latest ideas and results.


The topics of interest for this session include, but are not limited to:
• The use of FPGAs for Deep Learning
• FPGA-based Machine Learning Accelerators
• Reconfigurable Architectures for Machine Learning
• Reconfigurable Architectures for Data Mining and Data Analytics
• Reconfigurable Architectures for Information Retrieval
• Bio- And Neuro-inspired Reconfigurable Computing
• Programming and Compiling Machine Learning techniques to FPGAs
• Reconfigurable Computing for Intelligent Techniques and AI
• Use of Machine Learning in FPGA Programming and Design Tasks
• Energy-Efficient Reconfigurable Computing for Intelligent Systems

We are interested in accelerated data engineering and machine learning at all levels of the performance spectrum, from HPC/data center down to edge/IoT systems.


- Marco Platzner, Paderborn University, Germany (
- João MP Cardoso, University of Porto, Portugal (
- Antonio J. Tallón-Ballesteros, University of Huelva, Spain (
- Deming Chen, University of Illinois, US (

Contact us