Vol. 23, No. 1
http://hdl.handle.net/11012/179189
2021-10-18T11:50:31ZEmpirical Casestudy on Approaches to Sampled Control
http://hdl.handle.net/11012/179214
Empirical Casestudy on Approaches to Sampled Control
Burger, Alexandra; Wellhausen, Jens
Two main approaches on sampled control are compared, the descrete equivalent design and the direct discrete design. Both design methods are used to implement a control system for a real plant, DC-motor. It is shown that discrete equivalent design works well with su cient high sampling rates and that direct discrete design works nearly independent of sampling time.
2019-06-26T10:18:10ZHeuristic Methodology for Forecasting of Production in Waste Management
http://hdl.handle.net/11012/179216
Heuristic Methodology for Forecasting of Production in Waste Management
Smejkalova, Veronika; Somplak, Radovan; Nevrly, Vlastimir
The forecast of waste production and disposal is an important requirement for a future waste management planning. The problem is very often a short time series of the database. This paper suggests an approach to forecast the production of multiple waste types in micro-regions taking into account this challenge by combining many techniques. The heuristic methodology consisting of few steps is formulated. First, the input data are transformed and the methods from cluster analysis are repetitively applied. The second step is about a determination of quality for trend functions based on historical data. In the last step is performed the testing. The di erent type of representatives from cluster analysis is used to calculate indices of determination which are compared. This procedure is repeated until the criteria hit. The proposed approach reduced the computational time and managed to aggregate micro-regions with a similar trend. The forecast should have contributions in terms of building new facilities or adaptations to the existing ones, where it is necessary to estimate the production of waste for several years in advance. The article includes a case study of production forecast for several waste types in territorial units of the Czech Republic. The forecast is based on data in years 2009{2014 and following year 2015 was used to assess the quality of the nal models. In the future, the database will expand and thus it will be possible to make more precise estimates and to develop statistical methods to measure this prognostic tool.
2019-06-26T10:18:10ZOn Testing Reduction of Left-Censored Weibull Distribution to Exponential Submodel
http://hdl.handle.net/11012/179215
On Testing Reduction of Left-Censored Weibull Distribution to Exponential Submodel
Fusek, Michal
When analyzing environmental or chemical data, it is often necessary to deal with left-censoredobservations. Since the distribution of the observed variable is often asymmetric, the exponential or the Weibulldistribution can be used. This paper summarizes statistical model of a multiply left-censored Weibull distribution,and estimation of its parameters and their variances using the expected Fisher information matrix. Since inmany situations the Weibull distribution is unnecessarily complicated for data modelling, statistical tests (theLagrange multiplier test, the likelihood ratio test, the Wald test) for assessing suitability of replacement ofthe censored Weibull distribution with the exponential submodel are introduced and their power functions areanalyzed using simulations.
2019-06-26T10:18:10ZDesign of Linear Quadratic Regulator (LQR) Based on Genetic Algorithm for Inverted Pendulum
http://hdl.handle.net/11012/179211
Design of Linear Quadratic Regulator (LQR) Based on Genetic Algorithm for Inverted Pendulum
Marada, Tomas; Matousek, Radomil; Zuth, Daniel
One of the crucial problems in the dynamics and automatic control theory is balancing of an invertedpendulum robot by moving a cart along a horizontal path. This task is often used as a benchmark for di erentmethod comparison. In the practical use of the LQR method, the key problem is how to choose weight matricesQ and R correctly. To obtain satisfying results the experiments should be repeated many times with di erentparameters of weight matrices. These LQR parameters can be tuned by a Genetic Algorithm (GA) techniquefor getting better results. In our paper, the LQR parameters weight matrices Q and R which were tuned usingthe Genetic Algorithm. The simulations of the control problem are designed using MATLAB script code andMATLAB Simulink on an inverted pendulum model. The results show that the Genetic Algorithm is suitablefor tuning the parameters to give an optimal response. The control problem of the inverted pendulum was solvedsuccessfully.
2019-06-26T10:18:10ZDetection of Emergent Situations in Complex Systems by Structural Invariant (MB, M)
http://hdl.handle.net/11012/179213
Detection of Emergent Situations in Complex Systems by Structural Invariant (MB, M)
Bila, Jiri; Novak, Martin
The paper introduces complete description of the detection method that uses structural invariant Matroid and its Bases (MB, M). There are recapitulated essential concepts from the used knowledge field as “complex system, emergent situations (A, B, C)”, Ramsey theorem and principal computation variables “power” and “complexity” of emergence phenomenon. The method is explained in details and the demonstration of its application is done by the detection of emergent situation – violation of Short Water Cycle in an ecosystem.
2019-06-26T10:18:10ZLocal Control of (4,5,7,8-10)-Filtration Snake Robot via CGA
http://hdl.handle.net/11012/179212
Local Control of (4,5,7,8-10)-Filtration Snake Robot via CGA
Hrdina, Jaroslav; Navrat, Ales; Vasik, Petr; Matousek, Radomil
We describe the local control of a (6{8){link snake like robot endowed with omnidirectional wheels on two links (head and tail). All calculations including the position, direct kinematics, di erential kinematics and inverse kinematics are described in the terms of CGA only.
2019-06-26T10:18:10ZDynamic of Firework Algorithm Analyzed with Complex Network
http://hdl.handle.net/11012/179202
Dynamic of Firework Algorithm Analyzed with Complex Network
Kadavy, Tomas; Pluhacek, Michal; Viktorin, Adam; Senkerik, Roman
In this paper, a visualization of Firework Algorithm (FWA) inner dynamics as an evolving complex network is presented. Recent research in unconventional controlling and simulation of metaheuristic dynamics shows that this kind of visualization technique has been utilized only for algorithms with some social communication or behavior leading to sharing information across the population. However, provided analysis suggests that the network can identify some types of surface of tested functions.
2019-06-26T10:18:09ZThe Computational Power of Neural Networks and Representations of Numbers in Non-Integer Bases
http://hdl.handle.net/11012/179205
The Computational Power of Neural Networks and Representations of Numbers in Non-Integer Bases
Sima, Jiri
We briefly survey the basic concepts and results concerning the computational power of neural net-orks which basically depends on the information content of eight parameters. In particular, recurrent neural networks with integer, rational, and arbitrary real weights are classi ed within the Chomsky and finer complexity hierarchies. Then we re ne the analysis between integer and rational weights by investigating an intermediate model of integer-weight neural networks with an extra analog rational-weight neuron (1ANN). We show a representation theorem which characterizes the classification problems solvable by 1ANNs, by using so-called cut languages. Our analysis reveals an interesting link to an active research field on non-standard positional numeral systems with non-integer bases. Within this framework, we introduce a new concept of quasi-periodic numbers which is used to classify the computational power of 1ANNs within the Chomsky hierarchy.
2019-06-26T10:18:09ZInformation Security Risk Assessment Model Based on Computing with Words
http://hdl.handle.net/11012/179207
Information Security Risk Assessment Model Based on Computing with Words
Tymchuk, Oleg; Iepik, Maryna; Sivyakov, Artyom
The basis for company IT infrastructure security is information security risks assessment of IT services. The increased complexity, connectivity and rapid changes occurring in IT services make it impossible to apply traditional models of quantitative/qualitative risk assessment. Existing quantitative assessment models are time-consuming, at the same time, qualitative assessment models do not take into account the subjective expert assessments and the uncertainty of risk factors. This paper presents the new information security risk assessment model for IT services based on computing with words. The model methodology is based on OWASP risk rating methodology for web applications. To evaluate risk factors, it is proposed to use dictionary consisting of 16/32 granular terms (words). Problems of uncertainty in perceptual assessments of risk factors are taken into account using methods of the theory of discrete interval type-2 fuzzy sets and systems.
2019-06-26T10:18:09ZFinding Multiple Solutions of Multimodal Optimization Using Spiral Optimization Algorithm with Clustering
http://hdl.handle.net/11012/179204
Finding Multiple Solutions of Multimodal Optimization Using Spiral Optimization Algorithm with Clustering
Sidarto, Kuntjoro Adji; Kania, Adhe; Sumarti, Novriana
Multimodal optimization is one of the interesting problems in optimization which arises frequently in a widerange of engineering and practical applications. The goal of this problem is to find all of optimum solutions in a single run. Some algorithms fail to find all solutions that have been proven their existence analytically. In our paper [1], a method is proposed to find the roots of a system of non-linear equations using a clustering technique that combine with Spiral Optimization algorithm and Sobol sequence of points. An interesting benefit using this method is that the same inputs will give the same results. Most of the time this does not happen in meta-heuristic algorithms using random factors. Now the method is modified to find solutions of multimodal optimization problems. Generally in an optimization problem, the differential form of the objective function is needed. In this paper, the proposed method is to find optimum points of general multimodal functions that its differential form is not required. Several problems with benchmark functions have been examined using our method and they give good result.
2019-06-26T10:18:09ZOptimization of Personnel Cost in Aircrew Assignment Problem using a Simple Fuzzy Logic Approach
http://hdl.handle.net/11012/179209
Optimization of Personnel Cost in Aircrew Assignment Problem using a Simple Fuzzy Logic Approach
Sumarti, Novriana; Chandra, Ferdyanto; Minardi, Jeremy
In aviation industries, the aircrew assignment problem is one of the most important factors in total operational cost optimization. This problem will be solved in two steps: flight pairing and aircrew scheduling. The constraints to be satisfied in flight pairing include having the same airport for first departure and final destination, and the limitations of flying time, duty time and transit time. The optimization process results in optimal flight pairings that minimize the number of personnel needed to serve a flight schedule over a given period of time. Further optimization is needed to obtain a schedule in which an aircrew team can serve a rotation with the largest possible number of pairings on the condition that all constraints are fulfilled. For aircrew scheduling, there are constraints on flying time, resting time, total number of takeoffs, and number of holidays and workdays. The investigated optimization process was designed to get optimal rotations along with maximum total personnel cost reduction. The data set used in this research is a one-month full flight schedule from a big airline in Indonesia. A simple fuzzy logic approach was used to find a new flying time constraint in order to optimize personnel cost and evenly distribute the assignments. The results show that the new optimal flying time constraint can reduce personnel cost up to 5.07% per month, so it can yield significant savings on a yearly basis.
2019-06-26T10:18:09ZA Numerical Characterization of the Nanoparticles Distribution on the Surface of a Semiconductor
http://hdl.handle.net/11012/179208
A Numerical Characterization of the Nanoparticles Distribution on the Surface of a Semiconductor
Rudolfova, Zdena; Hoderova, Jana
The motivation for this work was to qualitatively describe the distribution of Au nanoparticles on the surface of a semiconductor. We discuss suitable mathematical characteristics which allow the uniform distribution to be distinguished from the distribution a ected by any physical phenomenon, i.e. by the repulsive force between electrically charged particles or by the influence of properties of the surface. We identify Voronoi decomposition and a statistical analysis of Voronoi cell properties as a suitable tool for this purpose.
2019-06-26T10:18:09ZAn Approach to Customer Behavior Modeling using Markov Decision Process
http://hdl.handle.net/11012/179210
An Approach to Customer Behavior Modeling using Markov Decision Process
Grunt, Ondrej; Plucar, Jan; Stakova, Marketa; Janecko, Tomas; Zelinka, Ivan
This paper presents an application of Markov Decision Process method for modeling of selected marketing processes. Based on available realistic data, MDP model is constructed. Customer behavior is represented by a set of states of the model with assigned rewards corresponding to the expected return value. Outcoming arcs then represent actions available to the customer in current state. Favourable outcome rate of available actions is then analyzed, with emphasis on suitability of the model for future predictions of customer behavior.
2019-06-26T10:18:09ZDocument Clustering using Self-Organizing Maps
http://hdl.handle.net/11012/179206
Document Clustering using Self-Organizing Maps
Rafi, Muhammad; Waqar, Muhammad; Ajaz, Hareem; Ayub, Umar; Danish, Muhammad
Cluster analysis of textual documents is a common technique for better ltering, navigation, under-standing and comprehension of the large document collection. Document clustering is an autonomous methodthat separate out large heterogeneous document collection into smaller more homogeneous sub-collections calledclusters. Self-organizing maps (SOM) is a type of arti cial neural network (ANN) that can be used to performautonomous self-organization of high dimension feature space into low-dimensional projections called maps. Itis considered a good method to perform clustering as both requires unsupervised processing. In this paper, weproposed a SOM using multi-layer, multi-feature to cluster documents. The paper implements a SOM usingfour layers containing lexical terms, phrases and sequences in bottom layers respectively and combining all atthe top layers. The documents are processed to extract these features to feed the SOM. The internal weightsand interconnections between these layers features(neurons) automatically settle through iterations with a smalllearning rate to discover the actual clusters. We have performed extensive set of experiments on standard textmining datasets like: NEWS20, Reuters and WebKB with evaluation measures F-Measure and Purity. Theevaluation gives encouraging results and outperforms some of the existing approaches. We conclude that SOMwith multi-features (lexical terms, phrases and sequences) and multi-layers can be very e ective in producinghigh quality clusters on large document collections.
2019-06-26T10:18:09ZUsing Complex Network Visualization and Analysis for Uncovering the Inner Dynamics of PSO Algorithm
http://hdl.handle.net/11012/179203
Using Complex Network Visualization and Analysis for Uncovering the Inner Dynamics of PSO Algorithm
Pluhacek, Michal; Senkerik, Roman; Viktorin, Adam; Kadavy, Tomas; Zelinka, Ivan
In this study, we construct a complex network from the inner dynamic of Particle Swarm Optimization algorithm. The subsequent analysis of the network promises to provide useful information for better understanding the dynamic of the swarm that is not acquirable by other means. We present several network visualizations and numerical analysis. We discuss the observations and propose further directions for the research.
2019-06-26T10:18:09ZSOMA Network Model Based on Native Visibility Graph
http://hdl.handle.net/11012/179198
SOMA Network Model Based on Native Visibility Graph
Tomaszek, Lukas; Zelinka, Ivan
In this article, we want to propose a new model of the network for analyzing the evolution algorithms.We focus on the graph called native visibility graph. We show how we can get a time series from the run ofthe self-organizing migrating algorithm and how we can convert these series into a network. At the end of thearticle, we focus on some basic network properties and we propose how can we use these properties for laterinvestigation. All experiments run on well-known CEC 2016 benchmarks.
2019-06-26T10:18:08ZGenetic Algorithm for Independent Job Scheduling in Grid Computing
http://hdl.handle.net/11012/179200
Genetic Algorithm for Independent Job Scheduling in Grid Computing
Younis, Muhanad Tahrir; Yang, Shengxiang
Grid computing refers to the infrastructure which connects geographically distributed computers ownedby various organizations allowing their resources, such as computational power and storage capabilities, to beshared, selected, and aggregated. Job scheduling is the problem of mapping a set of jobs to a set of resources.It is considered one of the main steps to e ciently utilise the maximum capabilities of grid computing systems.The problem under question has been highlighted as an NP-complete problem and hence meta-heuristic methodsrepresent good candidates to address it. In this paper, a genetic algorithm with a new mutation procedure tosolve the problem of independent job scheduling in grid computing is presented. A known static benchmark forthe problem is used to evaluate the proposed method in terms of minimizing the makespan by carrying out anumber of experiments. The obtained results show that the proposed algorithm performs better than some knownalgorithms taken from the literature.
2019-06-26T10:18:08ZDecision Algorithm for Heuristic Donor-Recipient Matching
http://hdl.handle.net/11012/179196
Decision Algorithm for Heuristic Donor-Recipient Matching
Namatevs, Ivars; Aleksejeva, Ludmila
This paper introduces the application of artificial intelligence paradigm towards precision medicine in renal transplantation. The match of the optimal donor-recipient pair in kidney transplantation in Latvian Transplant Centre (LTC) has been constrained by the lack of prediction models and algorithms. Consequently, LTC seeks for practical intelligent computing solution to assist the clinical setting decision-makers during their search for the optimal donor-recipient match. Therefore, by optimizing both the donor and recipient profiles, prioritizing importance of the features, and based on greedy algorithm approach, advanced decision algorithm has been created. The strength of proposed algorithm lies in identification of suitable donors for a specific recipient based on evaluation of criteria by points principle. Experimental study demonstrates that the decision algorithm for heuristic donor-recipient matching integrated in machine learning approach improves the ability of optimal allocation of renal in LTC. It is an important step towards personalized medicine in clinical settings.
2019-06-26T10:18:08ZFunction Set Structure Influence onto GPA Efficiency
http://hdl.handle.net/11012/179195
Function Set Structure Influence onto GPA Efficiency
Brandejsky, Tomas
The paper discusses the influence of function set structure onto efficiency of GPA (Genetic Programming Algorithms), and hierarchical algorithms like GPA-ES (GPA with Evolutionary Strategy to separate parameter optimization) algorithm efficiency. On the foreword, the discussed GPA algorithm is described. Then there is depicted function set and common requirements to its structure. On the end of this contribution, the test examples and environment as well as results of measurement of influence of superfluous functions presence in the used function set is discussed.
2019-06-26T10:18:08ZBubble Captcha - A Start of the New Direction of Text Captcha Scheme Development
http://hdl.handle.net/11012/179199
Bubble Captcha - A Start of the New Direction of Text Captcha Scheme Development
Bostik, Ondrej; Horak, Karel; Klecka, Jan
CAPTCHA, A Completely Automated Public Turing test to tell Computers and Humans Apart, iswell-known system widely used in all sorts of internet services around the world designated to secure the webfrom an automatic malicious activity. For almost two decades almost every system utilize a simple approach tothis problem containing a transcription of distorted letters from image to a text eld. The ground idea is to useimperfection of Optical Character Recognition algorithms against the computers. The development of OpticalCharacter recognition algorithms leads only to state, where the CAPTCHA schemes become more complex andhuman users have a great di culty with the transcription.This paper aims to present a new way of development of CAPTCHA schemes based more a human perception.The goal of this work is to implement new Captcha scheme and assess human capability to read unusual fontsnewer seen before.
2019-06-26T10:18:08ZAn Experimental Study on Competitive Coevolution of MLP Classifiers
http://hdl.handle.net/11012/179197
An Experimental Study on Competitive Coevolution of MLP Classifiers
Castellani, Marco; Lalchandani, Rahul
This paper investigates the effectiveness and efficiency of two competitive (predator-prey) evolutionaryprocedures for training multi-layer perceptron classifiers: Co-Adaptive Neural Network Training, and a modifiedversion of Co-Evolutionary Neural Network Training. The study focused on how the performance of the two procedures varies as the size of the training set increases, and their ability to redress class imbalance problems of increasing severity. Compared to the customary backpropagation algorithm and a standard evolutionary algorithm, the two competitive procedures excelled in terms of quality of the solutions and execution speed. Co-Adaptive Neural Network Training excelled on class imbalance problems, and on classification problems of moderately large training sets. Co-Evolutionary Neural Network Training performed best on the largest data sets. The size of the training set was the most problematic issue for the backpropagation algorithm and the standard evolutionary algorithm, respectively in terms of accuracy of the solutions and execution speed. Backpropagation and the evolutionary algorithm were also not competitive on the class imbalance problems, where data oversampling could only partially remedy their shortcomings.
2019-06-26T10:18:08ZHeuristic Approaches to Stochastic Quadratic Assignment Problem: VaR and CVar Cases
http://hdl.handle.net/11012/179201
Heuristic Approaches to Stochastic Quadratic Assignment Problem: VaR and CVar Cases
Matousek, Radomil; Popela, Pavel; Kudela, Jakub
The goal of this paper is to continue our investigation of the heuristic approaches of solving thestochastic quadratic assignment problem (StoQAP) and provide additional insight into the behavior of di erentformulations that arise through the stochastic nature of the problem. The deterministic Quadratic AssignmentProblem (QAP) belongs to a class of well-known hard combinatorial optimization problems. Working with severalreal-world applications we have found that their QAP parameters can (and should) be considered as stochasticones. Thus, we review the StoQAP as a stochastic program and discuss its suitable deterministic reformulations.The two formulations we are going to investigate include two of the most used risk measures - Value at Risk(VaR) and Conditional Value at Risk (CVaR). The focus is on VaR and CVaR formulations and results of testcomputations for various instances of StoQAP solved by a genetic algorithm, which are presented and discussed.
2019-06-26T10:18:08ZShift Scheduling of Short Time Workers in Large-Scale Home Improvement Center by using Cooperative Evolution
http://hdl.handle.net/11012/179194
Shift Scheduling of Short Time Workers in Large-Scale Home Improvement Center by using Cooperative Evolution
Ohki, Makoto
There are a lot of large-scale Home Improvement Center (HIC) in Japan. In the large-scale HIC,about hundred short time workers are registered. And about forty workers are working every day. A managercreates a monthly shift schedule. The manager selects the workers required for each working day, assigns theworking time and break time for each worker and also work placement. Because there are many requirementsfor the scheduling, the scheduling consumes time costs and efforts. Therefore, we propose the technique to createand optimize the schedule of the short time workers in order to reduce the burden of the manager. A cooperativeevolution is applied for generating and optimizing the shift schedule of short time worker. Several problems hasbeen found in carrying out this research. This paper proposes techniques to automatically create and optimize theshift schedule of workers in large-scale HIC by using a Cooperative Evolution (CE), to solve the situation thatmany workers concentrate in a speci c time zone, and to solve the situation where many breaks are concentratedin a speci c break time zone, and an effective mutation operators.
2019-06-26T10:18:08ZNature-Inspired Algorithms in Real-World Optimization Problems
http://hdl.handle.net/11012/179192
Nature-Inspired Algorithms in Real-World Optimization Problems
Bujok, Petr; Tvrdik, Josef; Polakova, Radka
Eight popular nature inspired algorithms are compared with the blind random search and three advanced adaptive variants of differential evolution (DE) on real-world problems benchmark collected for CEC 2011 algorithms competition. The results show the good performance of the adaptive DE variants and their superiority over the other algorithms in the test problems. Some of the nature-inspired algorithms perform even worse that the blind random search in some problems. This is a strong argument for recommendation for application, where well-verified algorithm successful in competitions should be preferred instead of developing some new algorithms.
2019-06-26T10:18:07ZData Scaling by Differential Evolution for FCA over Data from LMS eLogika
http://hdl.handle.net/11012/179193
Data Scaling by Differential Evolution for FCA over Data from LMS eLogika
Kermaschek, Jakub; Drazdilova, Pavla; Mensik, Marek
The e-learning system eLogika serves for teaching logic. The system collects data about users who arelogged in, e.g. time spent on a particular activity, the number of activities performed by particular students, whatdata is a student interested in, etc. The goal of this paper is to describe the application of many-valued formalconcept analysis (FCA) in order to discover typical patterns of students behavior. Since the data stored in theeLogika system are numerical we need to categorize them in order to be used in the many-valued FCA method.In the paper we describe a way of data categorization by di erential evolution that proved to be applicable in theFCA method with promising results.
2019-06-26T10:18:07ZOpening the Black Box: Alternative Search Drivers for Genetic Programming and Test-based Problems
http://hdl.handle.net/11012/179191
Opening the Black Box: Alternative Search Drivers for Genetic Programming and Test-based Problems
Krawiec, Krzysztof
Test-based problems are search and optimization problems in which candidate solutions interact with multiple tests (examples, fitness cases, environments) in order to be evaluated. The approach conventionally adopted in most search and optimization algorithms involves aggregating the interaction outcomes into a scalar objective. However, passing different tests may require unrelated `skills' that candidate solutions may vary on.Scalar tness is inherently incapable of capturing such di erences and leaves a search algorithm largely uninformed about the diverse qualities of individual candidate solutions. In this paper, we discuss the implications of this fact and present a range of methods that avoid scalarization by turning the outcomes of interactions between programs and tests into 'search drivers' - partial, heuristic, transient pseudo-objectives that form multifaceted characterizations of candidate solutions. We demonstrate the feasibility of this approach by reviewing the experimental evidence from past work, confront it with related research endeavors, and embed it into a broader context of behavioral program synthesis.
2019-06-26T10:18:07Z