Skip to main content
main-content

Über dieses Buch

This book gathers extended versions of the best papers presented at the 8th IEEE conference on Intelligent Systems, held in Sofia, Bulgaria on September 4–6, 2016, which are mainly related to theoretical research in the area of intelligent systems. The main focus is on novel developments in fuzzy and intuitionistic fuzzy sets, the mathematical modelling tool of generalized nets and the newly defined method of intercriteria analysis. The papers reflect a broad and diverse team of authors, including many young researchers from Australia, Bulgaria, China, the Czech Republic, Iran, Mexico, Poland, Portugal, Slovakia, South Korea and the UK.

Inhaltsverzeichnis

Frontmatter

Intercriteria Analysis and Arithmetic Functions

Abstract
The possibility to apply the intercriteria analysis over normalized data is discussed. An example, related to evaluation of six arithmetic functions, is given.
K T Atanassov, Vassia Atanassova, Panagiotis Chountas

Fuzzy Harmony Search Algorithm Using an Interval Type-2 Fuzzy Logic Applied to Benchmark Mathematical Functions

Abstract
This paper presents a fuzzy harmony search algorithm (FHS) based on an interval type-2 fuzzy logic system for dynamic parameter adaptation. The harmony memory accepting (HMR) and pitch adjustment (PArate) parameters are changing during the iterations in the improvisation process of this algorithm using the fuzzy system. The FHS has been successfully applied to various benchmark optimization problems. Numerical results reveal that the proposed algorithm can find better solutions when compared to a type-1 FHS and other heuristic methods and is a powerful search algorithm for various benchmark optimization problems.
Cinthia Peraza, Fevrier Valdez, Oscar Castillo

Mixture Initialization Based on Prior Data Visual Analysis

Abstract
The initialization is known to be a critical task for running a mixture estimation algorithm. A majority of approaches existing in the literature are related to initialization of the expectation-maximization algorithm widely used in this area. This study focuses on the initialization of the recursive mixture estimation for the case of normal components, where the mentioned methods are not applicable. Its key part is a choice of the initial statistics of normal components. Several initialization techniques based on visual analysis of prior data are discussed. Validation experiments are presented.
Evgenia Suzdaleva, Ivan Nagy

Spatiotemporal Parameter Estimation of Thermal Treatment Process via Initial Condition Reconstruction Using Neural Networks

Abstract
In this paper the design of control systems of periodical thermal treatment processes (TTP) with distributed parameters modeled by partial differential equations (PDEs) is considered. The main problem to decide is the estimation of the initial charge parameters—size, humidity, temperature and the relative load, which are all immeasurable. The investigation is based on first-principle models of the internal and external heat-exchange. Initially, after deriving the PDEs is created a representative TTP set by simulation using relevant combinations of charging parameters. To obtain their real estimates, the only measurable heating medium temperature, informative only during the first TTP stage, is applied. A cluster of N-nearest neighborhoods is found around the charge experimental temperature curve. A local situation-based dynamic neural network is learned to assess the charging parameters. They are implemented to define the optimal heating time of the current charge using another static neural network. Finally some aspects of industrial application of the proposed approaches are discussed.
M Hadjiski, Nencho Deliiski, Aleksandra Grancharova

Interval Type-2 Fuzzy Logic Dynamic Mutation and Crossover Parameter Adaptation in a Fuzzy Differential Evolution Method

Abstract
In this paper we consider the Differential Evolution (DE) algorithm by using fuzzy logic to make dynamic changes in the mutation (F) and crossover (Cr) parameters separately, and this modification of the algorithm we can call it the Fuzzy Differential Evolution algorithm (FDE). A comparison of the FDE algorithm using type-1 fuzzy logic and interval type-2 fuzzy logic is performed for a set of Benchmark functions.
Patricia Ochoa, Oscar Castillo, José Soria

Intuitionistic Fuzzy Evaluations for the Analysis of a Student’s Knowledge in University e-Learning Courses

Abstract
In the paper is proposed a method for evaluation of the student’s knowledge obtained in the university e-learning courses. For the assessment of the student’s solution of the respective assessment units the theory of intuitionistic fuzzy sets is used. The obtained intuitionistic fuzzy estimations reflect the degree of each student’s good performances, or poor performances, for each assessment unit. We also consider a degree of uncertainty that represents such cases where the student is currently unable to solve the problem. The method presented here provides the possibility for the algorithmization of the process of forming the student’s evaluations.
Evdokia Sotirova, Anthony Shannon, Taekyun Kim, Maciej Krawczak, Pedro Melo-Pinto, Beloslav Riečan

S-Logic with First and Second Imaginary States

Abstract
An imaginary logic, i-logic was introduced for solving of some unsolvable problems in the framework of the classical propositional logic. On the other hand similar unsolvable problems arise in the imaginary logic itself. Introduction of a second imaginary j-logic is suggested in the present work through which this insolvability in the i-logic is surmounted. For this purpose constraints are ushered in between the variables classical r-logic and those of the i-logic and j-logic. This gives rise to complex logics—s, s 1 , s 2 , respectively, in which functioning of the r, i, and j-logics is being interpreted. It is shown that all these logics are based on the algebraic structures Boolean algebra and lattice. A rule is proposed through which the contradictions may be avoided at the realization of associativity of disjunction and conjunction. A number of results are received for the behavior of conjunction and disjunction in the complex logics being concerned. On their base two truth tables are filled in and shown for the variables of indices r, i, j, s, s 1 —for conjunction and disjunction separately. It is pointed out that the logical structures being proposed may be considered as a complex multiple-valued logic with 12 states in which three two valued logics with indices r, i, j are interpreted in an appropriate manner. The field of application of the logical structures being investigated are shown.
Vassil Sgurev

Generalized Net Model of the Processes in a Center of Transfusion Haematology

Abstract
The proceeses in a center of transfusion haematology—receiving of person’s blood, obtaining of its fresh frozen plasma, erythrocytes and thrombocytes, its testing for transmissible diseases (HIV, HBV, HCV, Wass) and evaluation of blood group and Rh, and antibodies screening—are described by a generalized net.
Nikolay Andreev, Evdokia Sotirova, Anthony Shannon, K T Atanassov

Image to Sound Encryption Using a Self-organizing Map Neural Network

Abstract
This paper describes the process of encrypting image in a sound using artificial neural network. In order to achieve it the process is divided into several steps where each of the steps is described with a generalized net. The main goal is to send an image which is encrypted into a sound between two persons and if a wrong person receives it he will not be permitted to see the image. The neural network is divided into 5 clusters where each cluster responds to areas where the image has to be encrypted. When the procedure ends a random sound is applied to the network for testing and depending on which cluster it enters the necessarily areas are taken and the image is applied on them.
Todor Petkov, Sotir Sotirov

On Different Algorithms for InterCriteria Relations Calculation

Abstract
Contemporary InterCriteria analysis (ICrA) approach for searching of existing or unknown correlations between multiple objects against multiple criteria is applied here. Altogether five different algorithms for InterCriteria relations calculation have been examined to render the influence of the genetic algorithm parameters on the algorithm performance. Two cases, i.e. the model parameter identification of E. coli and S. cerevisiae fed-batch fermentation processes, are considered. In this investigation \(\mu \)-biased, Balanced, \(\nu \)-biased, Unbiased, as well as the newly elaborated and proposed here Weighted algorithm have been consequently applied and thoroughly examined. The obtained results for considered here two Case studies have been compared showing that the most reliable algorithm is the \(\mu \)-biased one.
Olympia Roeva, Peter Vassilev, Nikolay Ikonomov, Maria Angelova, Jun Su, Tania Pencheva

Defining Consonance Thresholds in InterCriteria Analysis: An Overview

Abstract
The present paper aims to provide an overview of the development of the approaches adopted in defining the consonance thresholds in the recently proposed method for decision support named InterCriteria Analysis (ICA). Discussing the rationale of this leg of the ICA research, and the motivation behind each of the subsequent steps, we trace here the gradual progress in defining the thresholds of the membership and non-membership parts of the intuitionistic fuzzy pairs serving as estimations of the pairwise consonances. This progress is based on both our deepening understanding of the ICA method, and the constant observations being made during the application of ICA to a wide range of different real-life problems and datasets.
Lyubka Doukovska, Vassia Atanassova, Evdokia Sotirova, Ivelina Vardeva, Irina Radeva

Design and Comparison of ECG Arrhythmias Classifiers Using Discrete Wavelet Transform, Neural Network and Principal Component Analysis

Abstract
Automatic classification of heartbeat is getting a significant value in today’s medical systems. By implementation of these methods in portable diagnosis devices, many mortal diseases can be realized and cured in primary steps. In this paper two separate classifiers are designed and compared for heartbeat classification. The first strategy profits principal component analysis for feature extraction and neural network for classification whereas the second strategy utilizes discrete wavelet transform (DWT) for feature extraction and neural network (NN) as classifier. The arrhythmias which are investigated here include: normal beats (N), right bundle branch block (RBBB), left bundle branch block (LBBB), ventricular premature contraction (VPC) and paced beat (P). In addition, an output for unspecified signals is considered which devotes to anonymous signals which are not in the above list. The results show that both methods could achieve above 98% accuracy on MIT-BIH database.
Seyed Saleh Mohseni, Vahid Khorsand
Weitere Informationen

Premium Partner

    Bildnachweise