International Journal of Neural Networks and Advanced Applications

E-ISSN: 2313-0563
Volume 2, 2015

Notice: As of 2014 and for the forthcoming years, the publication frequency/periodicity of NAUN Journals is adapted to the 'continuously updated' model. What this means is that instead of being separated into issues, new papers will be added on a continuous basis, allowing a more regular flow and shorter publication times. The papers will appear in reverse order, therefore the most recent one will be on top.

Main Page

Submit a paper | Submission terms | Paper format


Volume 2, 2015

Title of the Paper: Enhanced MVDR Beamforming for MEMS Microphone Array


Authors: Martin Papež, Karel Vlček

Pages: 42-46

Abstract: Microphone array technology has been widely used for the localization of sound sources. In particular, beamforming is a well-established signal processing method that maps the position of acoustic sources by steering the array transducers toward different directions. In this paper an implementation of DAS and MVDR beamforming were chosen for further simulations and testing, based on the terms of capabilities of MEMS microphone array. Both beamformers were extensively simulated and tested in Matlab. The results were used to compare the two beamformers based on their noise cancelling, frequency range and spatial filtration. The goal of this paper is comparing the accuracy of the MEMS embedded localization system when the different beamforming algorithms are used.

Title of the Paper: Mesh Refinement with Finite Elements and Artificial Neural Networks


Authors: Fatima Belhabib, Mohamed Ettaouil

Pages: 34-41

Abstract: In this paper, we present a new modeling for Mesh-size refinement with finite elements and artificial neural networks adopted by standards actual videos based on the SOM for image one domain, in the form of a structure. We developed in this study a mesh based on the object of interest by finite elements method and reduce the effort required to apply finite element analysis to image, this presentation that allows the identification of edges is a good representation of the movement of network nodes, and then we approach the follow-up of objects on sequences of Mesh-size refinement images. The algorithm of SOM the Kohonen is one of the important methods; it is a biologically inspired data clustering technique. It is a question of determining the Mesh adapte of an object nets, from one image to another. For that we used the algorithm allowing following a deformable plane object. On the one hand, we improve its performance, and then we study the optimization of the error function by error the Mesh-size refinement object simplification of our model, among the different meshes associated with images references. At the end of this work, we present simulation results.

Title of the Paper: An Algorithm to Transform an Artificial Neural Network into its Open Equation Form and Its Potential Applications


Authors: Wolfram C. Rinke

Pages: 28-33

Abstract: During the last decades artificial neural networks have evolved to an accepted and proven technology for modelling and function approximation. Different kinds of network architectures exist to support certain domains and applications in an efficient way.

Title of the Paper: The Application of Neural Networks on Analysis of Optical Glass Properties


Authors: Zora Jančíková, Pavel Koštial, Ondřej Zimný, Ondrej Bošák, Marcel Poulain

Pages: 24-27

Abstract: In the paper we present application of artificial neural network (ANN) on relation between glass composition versus optical transmittance of the chosen glass systems. The excellent prediction ability of ANN program shows a possibility to influence the glass composition to obtain required optical properties.

Title of the Paper: Handling Sparse Data Sets by Applying Contrast Set Mining in Feature Selection


Authors: Dijana Oreski, Bozidar Klicek

Pages: 12-23

Abstract: A data set is sparse if the number of samples in a data set is not sufficient to model the data accurately. Recent research emphasized interest in applying data mining and feature selection techniques to real world problems, many of which are characterized as sparse data sets. The purpose of this research is to define new techniques for feature selection in order to improve classification accuracy and reduce the time required for feature selection on sparse data sets. The extensive comparison with benchmarking feature selection techniques conducted on 128 data sets was conducted. Results of the 1792 analysis showed that in the more than 80% of the 128 analyzed data sets contrast set mining techniques are superior to benchmarking feature selection techniques. This paper provides a study on the new methodologies that have tried to handle the sparse datasets and showed superiority in handling data sparsity.

Title of the Paper: Toward IoT System Project: BRICS Mosaic Model and System Engineering Management


Authors: Marcel J. Simonette, Rodrigo F. Maia, Jose R. A. Amazonas, Edison Spina

Pages: 1-11

Abstract: One of the targets of Internet of Things (IoT) systems is to provide access to any service, to any user, anytime, anywhere, regardless the access network technology or the type of user device. The development of these systems demand the evaluation of several dimensions that are present in an IoT solution, including the integrated management of the engineering effort throughout system life cycle. As our knowledge about IoT systems grows and evolves, so has our understanding about the need of a management process to conduct the system life cycle. The diversity of dimensions present in IoT systems demands a systemic management process to promote the system vision as a whole. This work presents the BRICS Mosaic Model and the Feasibility Barriers Factors to evaluate IoT solutions. It is a model that quantifies, through the developers’ experience and analysis of application scenarios, a numerical relationship that allows identifying barriers to IoT solution development. This model is integrated with systems engineering management concepts, both to reduce a failure point, the managerial error, and to framework and guidance all engineering activities within the IoT system life cycle, from lust-to-dust.