Mark Farkas

36238002900

Publications - 7

Fuzzy and Kohonen SOM based classification of different 0D nanostructures

Mark Farkas Ákos Nemcsics Antal Ürmös Zoltán Farkas Tamás Sándor L. T. Koczy

Publication Name: Sami 2017 IEEE 15th International Symposium on Applied Machine Intelligence and Informatics Proceedings

Publication Date: 2017-03-16

Volume: Unknown

Issue: Unknown

Page Range: 365-370

Description:

In this paper, the clustering of the GaAs-based droplet epitaxially grown self-assembled nanostructures was investigated by soft-computing methods. The properties and the operation of these devices, depend on the type, the shape, the size, and their distribution of these 0 dimensional nanostructures. Because of this, it is very important to know, how and what kind of nanostructures can form, at the given technological parameters. Our goal is the classification of these nanostructures, in order to support the research and the production of these devices. Our solution is based on the shape factor calculation of the given nanostructure. In this work, two possible classification methods of nanostructures were introduced as well. First, the classification potential of the Kohonen Self-Organizing Mapping (SOM) was investigated. Second, the fuzzy inference system based classification was studied. In this case, the shape factor was determined by geometrical sizes of the nanostructures. In this paper the clustering was introduced, which supports many kinds of technology as well.

Open Access: Yes

DOI: 10.1109/SAMI.2017.7880335

Application of self-organizing maps for technological support of droplet epitaxy

Mark Farkas Ákos Nemcsics Antal Ürmös Zoltán Farkas Tamás Sándor L. T. Koczy

Publication Name: Acta Polytechnica Hungarica

Publication Date: 2017-01-01

Volume: 14

Issue: 4

Page Range: 207-224

Description:

The subject of this paper is the self-organized grouping of droplet epitaxial III-V-based nano-structures. For the nano-structure grouping, our developed algorithm - called Quantum Structure Analyzer 1.0 - is used. The operation of this software is based on the principles of the Kohonen Self-Organizing Network. Here, three possibilities for nano-structured groupings are shown. On one hand, we examine the classification of nanostructures with Kohonen Self-Organizing Maps, on the other hand, fuzzy inference systems are applied for the same goal. In the case of the fuzzy methods two approaches are examined in detail. According to the first fuzzy inference approach, the shape factor is calculated from the size of nanostructures. According to the second fuzzy inference approach, the shape factor calculation is based on the controllable parameters of the growth process (eg. pressure and the temperature of the substrate).

Open Access: Yes

DOI: 10.12700/APH.14.4.2017.4.12

Quantum structure classification by Kohonen Self-Organizing Map and by Fuzzy C-Means algorithm

Mark Farkas Ákos Nemcsics Antal Ürmös L. T. Koczy

Publication Name: Icsse 2013 IEEE International Conference on System Science and Engineering Proceedings

Publication Date: 2013-11-18

Volume: Unknown

Issue: Unknown

Page Range: 313-318

Description:

Nowadays the nanostructures, formed on the way of self assembly are intensively investigated both in the basic and the applied sciences. In our paper, we investigate the structures on III-V compund semiconductor based materials, which are grown by epitaxial process. This process is analized by the beta version of Quantum Structure Analyzer 1.0, which is developed in C# langague, in the Microsoft© Visual Studio 2008 development environment. This software operates with the help of the Kohonen Self-Organizing Maps (SOM) algorithm and with the help of the Fuzzy C-Means algorithm. In present work, in the preface we give a short introduction of Molecular Beam Epitaxy (MBE), after this we introduce the algorithms, applied in this software. Finally, we demonstrate the results of the program. © 2013 IEEE.

Open Access: Yes

DOI: 10.1109/ICSSE.2013.6614682

Determining an optimal subdivision of gene transfer partitions

J. Botzheim P. Földesi Mark Farkas L. T. Koczy

Publication Name: Proceedings of the 9th Wseas International Conference on Applied Computer and Applied Computational Science Acacos 10

Publication Date: 2010-12-01

Volume: Unknown

Issue: Unknown

Page Range: 202-207

Description:

Bacterial memetic algorithms are widely used on discrete combinatorial problems, which are essential in the field of logistics and forwarding, such as the well known Traveling Salesman Problem. The original Bacterial Evolutionary Algorithm proposed by Nawa and Furuhashi [5] has a predefined set of operators such as bacterial mutation and gene transfer also known as infection. The traditional bacterial infection operator is proven to be far from optimal. The authors suggest an alternative gene transfer operator that is applied on the metric Traveling Salesman Problem [9]. This alternative infection algorithm has superior rate of convergence while reducing the risk of getting stuck in a local optima.

Open Access: Yes

DOI: DOI not available

Approximation of a modified traveling salesman problem using bacterial memetic algorithms

J. Botzheim P. Földesi Mark Farkas L. T. Koczy

Publication Name: Studies in Computational Intelligence

Publication Date: 2009-12-01

Volume: 243

Issue: Unknown

Page Range: 607-625

Description:

The goal of this paper is to develop an algorithm that is capable to handle a slightly modified version of the minimal Traveling Salesman Problem in an efficient and robust way and produces high-quality solutions within a reasonable amount of time. The requirements of practical logistical applications, such as road transportation and supply chains, are also taken into consideration in this novel approach of the TSP. This well-known combinatorial optimization task is solved by a bacterial memetic algorithm, which is an evolutionary algorithm inspired by bacterial transduction. A new method is also proposed to deal with the time dependency in the cost matrix. The efficiency of the implementation, including time and space constraints, is investigated on a real life problem. © 2009 Springer-Verlag Berlin Heidelberg.

Open Access: Yes

DOI: 10.1007/978-3-642-03737-5_44

Delay propagation in a real life railway network controlled by a fuzzy logic rule base

Tibor Héray Gábor Rózsa Mark Farkas L. T. Koczy

Publication Name: 10th International Symposium of Hungarian Researchers on Computational Intelligence and Informatics Cinti 2009

Publication Date: 2009-12-01

Volume: Unknown

Issue: Unknown

Page Range: 423-433

Description:

This paper considers a real-life railway timetable related problem, where a set of interconnected railway junction points form a railway network, which is essentially a directed graph with corresponding vertex and edge capacities representing railway tracks and railway platforms.The dynamic behavior of this model is driven by a timetable. Unforeseen weather and other external effects may contribute to delays in the timetable, which alters the behavior of the whole system. In this scenario a hierarchical fuzzy system is proposed that can suggest a possible outgoing delay for each train by evaluating a set of fuzzy rule bases using two input data. The first proposal does not take into account the possible propagation of delay in the whole railway network. In this article a negative feedback is applied on the hierarchical fuzzy system.The fuzzy sets are optimized by an evolution based global search metaheuristics.

Open Access: Yes

DOI: DOI not available

The improvement of an existing fuzzy logic rule base for the treatment and simulation of conflicts in the dispositional tasks of a railway traffic control center

Tibor Héray Mark Farkas L. T. Koczy

Publication Name: Proceedings 2009 3rd International Workshop on Soft Computing Applications Sofa 2009

Publication Date: 2009-11-25

Volume: Unknown

Issue: Unknown

Page Range: 203-207

Description:

The paper considers a railway timetable related problem in a simplified form, generated by the delay of one or several incoming trains at a given station. These incoming train delays are either automatically generated or manually entered. Usually there are connecting trains in the timetable, especially in up to date periodic timetables and thus incoming delays might indicate the necessity of introducing a delay with connecting outgoing trains. A hierarchical fuzzy rule base is applied in order to determine the optimal outgoing delay, taking also usual restrictions into consideration. The delays of the incoming trains are modeled by independent exponential distributions and a software simulation is built around the existing rule base. The behavior of the hierarchical fuzzy rule base and the pertinence of the outgoing delay is a subject of further investigation. The delays are also updated periodically that makes the recalculation of outgoing delays necessary from time to time. 1 © 2009 IEEE.

Open Access: Yes

DOI: 10.1109/SOFA.2009.5254850