Petr Šaloun

Chaos Powered Selected Evolutionary Algorithms

It is well known that the evolution algorithms use pseudo-random numbers generators for example to generate random individuals in the space of possible solutions, crossing etc. In this paper we are dealing with the effect of different pseudo-random numbers generators on the course of evolution and the speed of their convergence to the global minimum. From evolution algorithms the differential evolution and self organizing migrating algorithm have been chosen because they have different strategies. As the random generators Mersenne Twister and chaotic system - logistic map have been used.

Hidden Complexity of Evolutionary Dynamics: Analysis

This chapter presents a method for visualization of the dynamics of evolutionary algorithms in the form of complex networks and is continuation of our previous research. The analogy between individuals of populations in an arbitrary evolutionary algorithm and vertices of a complex network is mentioned, as well as between edges in a complex network and communication between individuals in a population. Visualization of various attributes of network based on differential algorithm is presented here.

Towards Automated Navigation over Multilingual Content

Web-based adaptive and personalized systems are becoming standards on the web. The chapter describes the main features of selected webbased adaptive systems in the domain of education, and summarizes the architecture of such systems. One of the representatives of web-based adaptive personalized systems is described in more detail, XAPOS, designed and developed by the authors. XAPOS’s design is independent of the language of closed content and was used in a multilingual environment. XAPOS’s language-independent navigation feature is based on the domain ontology and principles of semantic web. We conducted a controlled experiment in an educational course of programming language Lisp in universities in Ostrava, Czech Republic, and Ankara, Turkey; results of two groups of users navigated using our algorithm as well as the standard one, and are compared and discussed in detail.

Big Data Movement: A Challenge in Data Processing

This chapter discusses modern methods of data processing, especially data parallelization and data processing by bio-inspired methods. The synthesis of novel methods is performed by selected evolutionary algorithms and demonstrated on the astrophysical data sets. Such approach is now characteristic for so called Big Data and Big Analytics. First, we describe some new database architectures that support Big Data storage and processing. We also discuss selected Big Data issues, specifically the data sources, characteristics, processing, and analysis. Particular interest is devoted to parallelism in the service of data processing and we discuss this topic in detail. We show how new technologies encourage programmers to consider parallel processing not only in a distributive way (horizontal scaling), but also within each server (vertical scaling). The chapter also intensively discusses interdisciplinary intersection between astrophysics and computer science, which has been denoted astroinformatics, including a variety of data sources and examples. The last part of the chapter is devoted to selected bio-inspired methods and their application on simple model synthesis from astrophysical Big Data collections. We suggest a method how new algorithms can be synthesized by bio-inspired approach and demonstrate its application on an astronomy Big Data collection. The usability of these algorithms along with general remarks on the limits of computing are discussed at the conclusion of this chapter.

Fractal Analysis of Fitness Landscapes

Complex optimization problems may have fitness landscapes with fractal characteristics. This chapter reviews landscapes obtained from basic artificial test functions as well as cost functions of real application problems which have the property to be fractal. We will discuss the description, structure and complexity of these fractal fitness landscapes. A major topic of this chapter is to use elements from fractal geometry to measure attributes of fractal landscapes. Also, structural as well as functional properties of the landscape are discussed. The examples used in this chapter are two-dimensional, however it is possible to extend the proposed analysis to n dimensions.

Chaos Powered Symbolic Regression in Be Stars Spectra Modeling

Be stars are characterized by prominent emission lines in their spectrum. In the past research has attention been given to creation a feature extraction method for classification of Be stars with focusing on the automated classification of Be stars based on typical shapes of their emission lines. The aim was to design a reduced, specific set of features characterizing and discriminating the shapes of Be lines. In this chapter we discuss possibility to create in an evolutionary way the model of spectra of Be stars. We focus on the evolutionary synthesis of the mathematical models of Be stars based on typical shapes of their emission lines. Analytical programming powered by classical random as well as chaotic random-like number generator is used here. Experimental data are used from the archive of the Astronomical Institute of the Academy of Sciences of the Czech Republic. Interpretation and explanation of analysis is given and discussed in this chapter.

Evolutionary synthesis of automatic classification on astroinformatic big data

Abstract This article describes using of new approach to automatic classification of big data records in Be and B[e] stars spectra in large astrophysical archives. With enormous amount of these data it is no longer feasible to analyse it using classical approaches. We introduce evolutionary synthesis of the classification by means of so called analytic programming (AP), one of methods of symbolic regression. By using this method, we synthesise the most suitable mathematical models that approximate chosen samples of the stellar spectra. As a result is then selected the class whose synthesised formula has the lowest difference (i.e. the most similar) compared to the particular spectrum. The results show us that classification of stellar spectra by means of AP is able to identify different shapes of the spectra and classify them. Graphical Abstract

Personality Disorders Identification in Written Texts

Greater or lesser personality disorders are due to a stressful and time-tight way of life today, and quite frequent in the case of restrictions or complications in the life of an individual is suffering from early identification and problem solving is more than desirable. But some people consider visiting a specialist as a personal failure and shame due not solve the problem, even if it suspects themselves. Psychologists and psychiatrists on the other hand today use several methods to detect personality disorders, either by observation during the interview, a questionnaire and a written description of their own people. This article describes a method for dealing with the first detection of possible personality disorder without the necessary presence of specialists and using the patient himself essays. Written text analytes using NLP techniques and categorize it into one of the three main groups of personality disorders—fear, procrastination and intolerance of uncertainty. We customized approach based on the method of support vector machine and the first experiential, based on real data consulted with specialists, have shown promising results.

Thematic labeling of Twitter accounts using DBpedia properties

In this paper, we propose an iterative algorithm towards the automatic labeling of Twitter accounts in respect to thematic categories derived from DBpedia properties. We describe the rationale behind the selection of these thematic categories, and discuss their evaluation assessment. Finally, we propose and analyze two generic and adaptable methodologies for discovering the necessary linked data resources for further enhancing the thematic description of Twitter accounts.

Chaos Powered Grammatical Evolution

In this paper we discuss alternative nonrandom generators for symbolic regression algorithms and compare its variants powered by classical pseudo-random number generator and chaotic systems. Experimental data from previous experiments reported for genetic programming and analytical programming is used. The selected algorithms are differential evolution and SOMA. Particle swarm, simulated annealing and evolutionary strategies are in process of investigation. All of them are mutually used in scheme Master-Slave meta-evolution for final complex structure fitting and its parameter estimation.