Dušan Fister

University of Alcalá | UAH

Numerical association rule mining offers a very efficient way of mining association rules, where algorithms can operate directly with categorical and numerical attributes. These methods are suitable for mining different transaction databases, where data are entered sequentially. However, little attention has been paid to the time series numerical a...

A novel method for prediction of the extreme wind speed events based on a Hierarchical Classification/Regression (HCR) approach is proposed. The idea is to improve the prediction skills of different Machine Learning approaches on extreme wind speed events, while preserving the prediction performance for steady events. The proposed HCR architecture...

Population health and wellbeing is both a result, as well as a driver, of economic development and prosperity on global, European, national and sub-national (local) levels. Wales, one of the four United Kingdom (UK) nations, has shown a long-term commitment to sustainable development and achieving prosperity for all, providing a good example of bot...

In this paper three customised Artificial Intelligence (AI) frameworks, considering Deep Learning (convolutional neural networks), Machine Learning algorithms and data reduction techniques are proposed, for a problem of long-term summer air temperature prediction. Specifically, the prediction of average air temperature in the first and second Augus...

Atmospheric Extreme Events (EEs) cause severe damages to human societies and ecosystems. The frequency and intensity of EEs and other associated events are increasing in the current climate change and global warming risk. The accurate prediction, characterization, and attribution of atmospheric EEs is therefore a key research field, in which many g...

Large information datasets often impose an immense number of features where many are found redundant and thus inessential for statistical analysis. In the past, a data preprocessing phase was formalized to cope with the problem and take appropriate remedial measures. Traditionally, this was a fixed and stationary process that suffered from a lack o...

The results of evolutionary algorithms depend on population diversity that normally decreases by increasing the selection pressure from generation to generation. Usually, this can lead the evolution process to get stuck in local optima. This study is focused on mechanisms to avoid this undesired phenomenon by introducing parallel self-adapted diffe...

Governments around the world are looking for ways to manage economic consequences of COVID-19 and promote economic development. The aim of this study is to identify the areas where the application of economic policy measures would enhance the resilience of societies on epidemic risks. We use data on the COVID-19 pandemic outcome in a large number o...

Reinforcement learning is a computational approach that mimics learning from interaction and supplements the existing supervised and unsupervised learning methods within the machine learning field. It bases on the mapping of a given situation to the action, and each action is evaluated by a reward. Of crucial concern, here is that the mapping is pe...

The rapid development of computer science and telecommunications has brought new ways and practices to sport training. The artificial sport trainer, founded on computational intelligence algorithms, has gained momentum in the last years. However, artificial sport trainer usually suffers from a lack of automatisation in realization and control phase...

This paper aims to find a superior strategy for the daily trading on a portfolio of stocks for which traditional trading strategies perform poorly due to the low frequency of new information. The experimental work is divided into a set of traditional trading strategies and a set of long short-term memory networks. The networks incorporate general a...

Computational Intelligence methods for automatic generation of sport training plans in individual sport disciplines have achieved a mature phase. In order to confirm their added value, they have been deployed into practice. As a result, several methods have been developed for generating well formulated training plans on computers automatically that...

Recently, sports training sessions have been generated automatically according to the TRIMP load quantifier that can be calculated easily using data obtained from mobile devices worn by an athlete during the session. This paper focuses on generating a sport training session in cycling, and bases on data obtained from power-meters that, nowadays, pr...

The results of evolutionary algorithms depends on population diversity that normally decreases by increasing the selection pressure from generation to generation. Usually, this can lead evolution process to get stuck in local optima. The study is focused on mechanisms to avoid this undesired phenomenon by introducing parallel differential evolution...

In recent years, some sport clubs have adopted web forums for online discussions about planning training sessions, races, club problems, sponsors and supporters, equipment and so on. Mostly, these forums are closed, because some discussions about critical information must be permitted only to registered club members. Indeed, various members are con...

Association Rule Mining belongs to one of the more prominent methods in Data Mining, where relations are looked for among features in a transaction database. Normally, algorithms for Association Rule Mining mine a lot of association rules, from which it is hard to extract knowledge. This paper proposes a new visualization method capable of extracti...

Advising athletes how to improve their performance after a race is a very important aspect of sport training. It can also be called a post-hoc analysis, which often includes a deep analysis of an athlete's performance, behavior and body characteristics after a race. These analyses help trainers to adapt their training plan according to the athlete'...

Preference time in a triathlon denotes the time that is planned to be achieved by an athlete in a particular competition. Usually, the preference time is calculated some days, weeks, or even months before the competition. Mostly, trainers calculate the proposed preference time according to the current form, body performances of athletes, psychologi...

Stochastic nature-inspired population-based algorithms are very powerful tools for solving stationary and deterministic, NP-hard optimization problems. These algorithms have rarely been applied to real-world dynamic and uncertain optimization due to their complexity. In this paper, this kind of algorithms were ported onto real hardware (i.e., the v...

With the advent of big data, interest for new data mining methods has increased dramatically. The main drawback of traditional data mining methods is the lack of comprehensibility. In this paper, the firefly algorithm was employed for standalone binary classification, where each solution is represented by two classification rules that are easy unde...

Nowadays, a big pool of different machine learning components (i.e., algorithms and tools) exists that are capable of predicting various decisions in different problem domains successfully. Unfortunately, a problem has emerged in this respect that we cannot estimate safely which component behaves well on a particular dataset without huge experiment...

Knowledge discovery in databases is a comprehensive procedure which enables researchers to explore knowledge and information from raw sample data usefully. Some problems may arise during this procedure, for example the Curse of Dimensionality, where the reduction of database is desired to avoid feature redundancy or irrelevancy. In this paper, we p...

Financial cycles have an important impact on the real economy and should be closely observed by the monetary policy makers. This can be done by monitoring individual indicators, using composite indicators, or as we propose in this paper, by developing prediction models for cyclical components of GDP and phase detection. This way, we have tools whic...

This paper estimates the size of the economic impact generated by the Slovenian health care sector on the national economy in the 2009–2014 period. The study separately calculates output, income, employment, value-added and import multipliers for the Slovenian health care sector based on input-output analysis covering 49 sectors. Initially, values...

Stochastic population-based nature-inspired metaheuristics have recently revealed that they are a very robust tool for planning sport training sessions in various sports, e.g. running, cycling, triathlon. Most of the existing solutions in literature are focused on planning training sessions for a particular training cycle. Until recently, no specia...

Razvita ogrodja za razvoj spletnih aplikacij lahko domala najdemo za vsak programski jezik. Dandanes prevladujejo ogrodja v programskih jezikih Python, Java, Go, PHP, Ruby. Yesod je spletno ogrodje, razvito v programskem jeziku Haskell, namenjeno za razvoj tako enostavnih kot tudi zahtevnih spletnih aplikacij. Glavni značilnosti, ki opisujeta to og...

Studying the lifestyle of various groups of athletes has been a very interesting research direction of many social sport scientists. Following the behavior of these athletes' groups might reveal how they work, yet function in the real-world. Triathlon is basically depicted as one of the hardest sports in the world (especially long-distance triathlo...

In the last chapter, the automatic generation of sports training plans using CI algorithms was presented, where it was expected that the their realization would be proceeded smoothly. However, realization of the proposed training plan can be disrupted when an athlete in training process ill or is injured. These unpredictable events can interrupt th...

Creating training plans is the more important task for real trainers, in which specific training sessions are prescribed to trainees according to intensity, duration, type, and repetition, for a specific training period. After realization of the plan, it is expected that the athlete in training would acquire the proper performance level needed for...

The process of sports training consists of four phases: planning, realization, control and evaluation. In the planning phase, trainers prescribe a training load that must be overcome by the athlete during the realization. The response of the athlete on the load represents physical stress. This can be measured indirectly by several physical measures...

Pervasive computing has emerged with the advent of mobile technology. The aim is to be able to obtain information everywhere, at any time, for any objective. Consequently, the concept of disappearing hardware has been developed, where computers are hidden from the users perceptions. This is made possible through the use of sensors that are capable...

This chapter deals with identifying the characteristics of athletes in training. According to the theory of the sports training, this identification is conducted after an evaluation phase, where goals set prior to the training cycle are compared with the achieved results. The purpose of this process is to discover those characteristics of the athle...

Movement is one of the more complex human functions requiring multiple biological systems in the body to operate in concert. There are five systems that enable the functioning of the organism: skeletal, muscular, nervous, respiratory, and cardiovascular. The first three create a so-called kinetic chain that is responsible for performing the functio...

This chapter presents an automatic construction of sports dietary plans based on the training plan generated by an artificial sports trainer. Differential evolution serves as the core algorithm for this purpose. The goal of this algorithm is to select the suitable foods from a food list dataset according to estimated macro-nutrient requirements. Th...

Nowadays, the use of sport trackers increases from day to day. Athletes from different sports disciplines use them in three ways: (1) to monitor their performance data during training, (2) to analyze data after training sessions, and (3) to use the results of the analysis to improve their performance. Many different tracking technologies have been...

The artificial sports trainer bases on CI algorithms for enriching knowledge from data in databases that are obtained from sports activity datasets. These datasets are generated by wearable mobile devices (e.g., sports watches and smart phones) during sports training sessions. Indeed, the artificial sports trainer serves as an intelligent system ca...

The chapter deals with knowledge discovery from data in sport. In the narrower sense, knowledge discovery from data refers to a data mining that also incorporates methods from other domains, like statistics, pattern recognition, machine learning, visualization, association rule mining and computational intelligence algorithms.

Nowadays, most of databases for classification or regression consists of numerous features that describe the domain of interest. Therefore, they may have a huge influence on the results of classification/regression. A lot of research has shown that some features can be eliminated before the classification/regression in order to obtain better result...

The majority of non-linear systems nowadays are controlled online using rapid PI-controllers with linear characteristics. Evolutionary algorithms are rarely used, especially for online adaptive control, due to their time complexity. This paper proposes an online adaptive controller based on a dynamic evolution strategy and attempts to overcome this...

In recent years, some sport clubs have adopted web forums for online discussions about planning training sessions, races, club problems, sponsors and supporters, equipment and so on. Mostly, these forums are closed, because some discussions about critical information must be permitted only to registered club members. Indeed, various members are con...

Modeling preference time in triathlons means predicting the intermediate times of particular sports disciplines by a given overall finish time in a specific triathlon course for the athlete with the known personal best result. This is a hard task for athletes and sport trainers due to a lot of different factors that need to be taken into account, e...

This technical report presents a collection of sport activity datasets for data mining purposes. It is following the three previously published collections in 2015 and 2016 and covers various sports and different types of athletes (professionals and amateurs). Data were exported from Strava, Garmin Connect and Movescount App profiles (last variant...

To predict the final result of an athlete in a marathon run thoroughly is the eternal desire of each trainer. Usually, the achieved result is weaker than the predicted one due to the objective (e.g., environmental conditions) as well as subjective factors (e.g., athlete's malaise). Therefore, making up for the deficit between predicted and achieved...

The sports training process is an origin of huge amounts of data obtained using mobile devices equipped with various sensors. Heart rate monitors, GPS, step counters, velocity meters, cadence meters and power meters are just some of them. If interpreted carefully, collected data offer an outstanding insight into an athlete's current form. Thus, spo...

This paper presents the automatic generation of optimal eating plans for athletes. The automatic generation of the eating plans is introduced as an optimization problem, where particle swarm optimization is taken as the problem solver. Inputs for the proposed particle swarm optimization algorithm are generated training plan and list of the potentia...

Source: Groove Korea magazine

The presented Technical Report presents a collection of IRONMAN, IRONMAN 70.3 and Ultra-triathlon race results that was scraped from the web. The collection is intended for data mining purposes.

A PID controller is an electrical element for reducing the error value between a desired setpoint and an actual measured process variable. The PID controller operates according to its input parameters, which need to be set before its run. The optimal values of these parameters must be found during the so-called tuning process. Today, this process c...

Correct input controller parameter settings are vital and in constant connection with output functions - e.g. robotic positioning. Optimal positioning of robotic arm automatically provides a high level of safety and functionality. The first prevents robot from hurting any people around or even itself, while the second ensures robot advantage. In or...

Endurance sport events have increasingly been gaining the popularity. Every year, more and more amateur athletes decide to participate in such events. During the race, proper eating is one of the most important components for achieving the good finish time and in this respect also the good place. In this paper, we examine possibility to predict wha...

Nowadays, many stochastic metaheuristics have been developed to solve various optimisation problems. The primary characteristics of these heuristics often involve the use of randomness in their search process. Essentially, randomness is useful when determining the next point in the search space and therefore has a crucial impact when exploring new...

Recreational cycling marathons are a kind of attractive sport event where a bunch of riders ride a different courses without competing between each other. In the last fifteen years, sport became popular for people of all ages. Therefore, many recreational cyclists has joint to such events in order to test their abilities, have a good time with frie...

Over fairly recent years the concept of an artificial sport trainer has been proposed in literature. This concept is based on computational intelligence algorithms. In this paper, we try to extend the artificial sports trainer by planning fitness training sessions that are suitable for athletes, especially during idle seasons when no competition ta...

Cuckoo search (CS) was introduced in 2009, and it has attracted great attention due to its promising efficiency in solving many optimization problems and real-world applications. In the last few years, many papers have been published regarding cuckoo search, and the relevant literature has expanded significantly. This chapter summarizes briefly the...

The significant development of the Internet has posed some new challenges and many new programming tools have been developed to address such challenges. Today, semantic web is a modern paradigm for representing and accessing knowledge data on the Internet. This paper tries to use the semantic tools such as resource definition framework (RDF) and RD...

Znanstveniki so vedno iskali vzor za reševanje najtežjih problemov, ki se pojavljajo na področju računalništva, matematike in industrije, v naravi. Narava ponuja različne smeri in poti za razvoj novih algoritmov. Dandanes se pojavlja veliko različnih algoritmov po vzorih iz narave, ki jih v splošnem lahko razdelimo v tri kategorije. Prva skupina al...

Firefly algorithm (FA) was developed by Xin-She Yang in 2008 and it has become an important tool for solving the hardest optimization problems in almost all areas of optimization as well as engineering practice. The literature has expanded significantly in the last few years. Various FA variants have been developed to suit different applications. T...

Today, global object-positioning is accomplished very precisely by GPS satellite technology. Access to this information is provided globally by widespread mobile devices with integrated GPS receivers from everywhere. On the other hand, mobile devices are connected to worldwide networks that ensure anytime access to application service (also web ser...

Pervasive technologies allow people to get information from everywhere and at anytime. The rapid development of smart mobile devices on the one hand and the progress of wireless networks on the other, has led to the widespread growth of pervasive technologies. Pervasive devices are small pieces of hardware that allow users to communicate over the I...

Swarm intelligence and bio-inspired algorithms form a hot topic in the developments of new algorithms inspired by nature. These nature-inspired metaheuristic algorithms can be based on swarm intelligence, biological systems, physical and chemical systems. Therefore, these algorithms can be called swarm-intelligence-based, bio-inspired, physics-base...

In this paper, we present a novel solution for the hybridization of the bat algorithm with differential evolution strategies and a random forests machine learning method. Extensive experiments and tests on standard benchmark functions have shown that these hybridized algorithms improved the original bat algorithm significantly.

Today, global object-positioning is accomplished very precisely by GPS satellite technology. Access to this information is provided globally by widespread mobile devices with integrated GPS receivers from everywhere also from airplane. This paper presents a mobile device using GPS receiver to be used by mobile navigation for sports pilots. In fact,...