Science Bulletin of the Novosibirsk State Technical University最新文献_第10页

A neuro-fuzzy classification of objects and their states 物体及其状态的神经模糊分类

Science Bulletin of the Novosibirsk State Technical University Pub Date : 2018-09-17 DOI: 10.17212/1814-1196-2018-3-73-86

A. Korikov, Russian Federation Radioelectronics, A. T. Nguyen

{"title":"A neuro-fuzzy classification of objects and their states","authors":"A. Korikov, Russian Federation Radioelectronics, A. T. Nguyen","doi":"10.17212/1814-1196-2018-3-73-86","DOIUrl":"https://doi.org/10.17212/1814-1196-2018-3-73-86","url":null,"abstract":"In this paper, we present the development and some investigation results of a neural fuzzy network (NFN) to solve classification problems of biological objects and their states. A general approach to the problem of automatic classification of objects is studied. In this work, we use a Fisher's Iris data set for the object classification problem, and the assessment of the emotional state of objects is given on the data set of the balance scale. The NFN is a combina-tion of fuzzy inference systems and a neural network. The structure of the NFN takes the form of a multilayer unidirectional network consisting of an input layer, a fuzzy activation function (AF) layer, a fuzzification layer, a defuzzification layer, a normalization layer and an output layer. In the fuzzy AF layer, we use fuzzy AFs that were developed and investigated in our previous studies. Herein, four fuzzy AFs corresponding to four types of NFN are investigated. The training process of the NFN is conducted by the use of the K-mean cluster method to determine antecedent network parameters. During the training process, a scaled conjugate gradient (SCG) algorithm is used to reduce the computational effort of every learning iteration, and therefore, enhance the learning speed. The testing process is carried out with Fisher's Iris data sets and the balance scale. These data sets are classical and often used to illustrate the performance of various statistical classification algorithms. The results obtained by various types of fuzzy AF have confirmed the validity of the current NFN to solve classification problems. A comparative analysis of the current four NFN types and the NFN using the Gaussian AF from a previous pa-per in terms of efficiency and accuracy is performed in this work.","PeriodicalId":214095,"journal":{"name":"Science Bulletin of the Novosibirsk State Technical University","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123074915","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 2

Applying information methods, neural networks and genetic algorithms for solving the problem of selecting a scheme of treatment 应用信息方法、神经网络和遗传算法解决治疗方案的选择问题

Science Bulletin of the Novosibirsk State Technical University Pub Date : 2018-09-17 DOI: 10.17212/1814-1196-2018-3-7-20

O. Gerget, R. Meshcheryakov

{"title":"Applying information methods, neural networks and genetic algorithms for solving the problem of selecting a scheme of treatment","authors":"O. Gerget, R. Meshcheryakov","doi":"10.17212/1814-1196-2018-3-7-20","DOIUrl":"https://doi.org/10.17212/1814-1196-2018-3-7-20","url":null,"abstract":"Recently, bionic-based IT solutions for monitoring developing biosystems have become a promising scientific trend. Biosystems evolving over millions of years have formed struc-tures, such as genetic, immune and neural systems that ensure their balanced development and the availability of the necessary information means to control and adaptively manage them in a changing environment. At the present time, attempts are made to use artificial information pro-cessing systems that structurally reflect the functioning of biosystems. Particular attention is paid to the development of models and methods that fully consider the specific nature of each research object. The present study is aimed to consider ways to minimize the possibility of a human or-ganism transition to an unfavorable state through the selection of control activities sequence. To solve the problem, we developed a bionic model based on combining information approach-es, neural networks, and a genetic algorithm. The functions of the model elements and their in-teraction are considered in the paper. A special focus is on the neuro-evolutionary interaction. The description of the software implemented in the programming language Python is described. Test-control groups and cross-validations weres used to evaluate the effectiveness of solutions based on bionic modeling. It was experimentally proved that the proposed method is effective for selecting and applying control activities.","PeriodicalId":214095,"journal":{"name":"Science Bulletin of the Novosibirsk State Technical University","volume":"306 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116226915","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 2

An efficient parametric algorithm for optimal scheduling of unrelated parallel machines with release dates 具有出厂日期的不相关并行机最优调度的有效参数化算法

Science Bulletin of the Novosibirsk State Technical University Pub Date : 2018-09-17 DOI: 10.17212/1814-1196-2018-3-87-106

Yu. A. Mesentsev, I. Estraykh

引用次数: 0

Study of classification methods for automatic truth and deception detection in speech 语音自动真伪检测的分类方法研究

Science Bulletin of the Novosibirsk State Technical University Pub Date : 2018-09-17 DOI: 10.17212/1814-1196-2018-3-21-32

A. Velichko, V. Budkov, Alexey Karpov

引用次数: 1

Research on optimal modes of intelligent networks with a two-way energy flow 具有双向能量流的智能网络最优模式研究

Science Bulletin of the Novosibirsk State Technical University Pub Date : 2018-09-17 DOI: 10.17212/1814-1196-2018-3-175-190

V. Manusov, Nasrullo Khasanzoda, S. Boboev

引用次数: 1

Elimination of unstable single-phase short-circuits in transposed lines in the SPAR cycle using automatic phase shunting 利用自动相位分流消除SPAR循环中转置线路的不稳定单相短路

Science Bulletin of the Novosibirsk State Technical University Pub Date : 2018-09-17 DOI: 10.17212/1814-1196-2018-3-157-174

Saiyod Jononaev

引用次数: 1

Compensation of disturbances in the problem of robust control of a plant network with distributed and discrete delays 具有分布和离散延迟的植物网络鲁棒控制问题中的扰动补偿

Science Bulletin of the Novosibirsk State Technical University Pub Date : 2018-09-17 DOI: 10.17212/1814-1196-2018-3-59-72

A. Imangazieva

引用次数: 0

Energy processes in the simplest electrical circuit 能量在最简单的电路中产生

Science Bulletin of the Novosibirsk State Technical University Pub Date : 2018-09-17 DOI: 10.17212/1814-1196-2018-3-191-202

A. Mashkin, S. Mashkina

引用次数: 0

A system for simulation modeling of dynamic automata networks 动态自动机网络的仿真建模系统

Science Bulletin of the Novosibirsk State Technical University Pub Date : 2018-09-17 DOI: 10.17212/1814-1196-2018-3-107-120

I. Tsarev

{"title":"A system for simulation modeling of dynamic automata networks","authors":"I. Tsarev","doi":"10.17212/1814-1196-2018-3-107-120","DOIUrl":"https://doi.org/10.17212/1814-1196-2018-3-107-120","url":null,"abstract":"This paper is devoted to some aspects of creation a program system for simulation modeling of complex computer systems with a network structure based on the model of parallel computations known as “dynamic automata networks”. Such systems as multiprocessors and supercomputers with a dynamic architecture belong to these systems. The system of simulation modeling of dynamic automata networks makes it possible to investigate the operability of the structure of such systems as well as structures, functions and algorithms of individual compo-nents of the system on the stage of development, and to assess the main characteristics of the system designed. This research resulted in developing the structure of the system for simulation modeling. In addition, a non-algorithmic programming language to describe simulated systems and problems being solved on such systems is proposed. A compiler to convert the description of the system into an internal representation used as input information both for the program of simulation modeling and for real devices being developed. Algorithms of the compiler permit-ting us to perform tuning of the compiler for a concrete version of a programming language with a possibility of changing the syntax and semantics of the language are developed. The main algorithms and data structures of the simulation system are developed. General problems of creation of such a system are noted and some possible ways for solving these problems are defined. The practical value of the solutions proposed consists in a possibility of examining the operability and characteristics of complex computing systems with the network structure creat-ed on the basis of dynamic automata networks. The developed system of simulation modeling permits us both to define the operability of the system being designed and to evaluate its basic characteristics.","PeriodicalId":214095,"journal":{"name":"Science Bulletin of the Novosibirsk State Technical University","volume":"7 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125024310","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Synthesis of a neural network for the implementation of the recursive least-squares method 综合了一个神经网络，实现了递归最小二乘法

Science Bulletin of the Novosibirsk State Technical University Pub Date : 2018-09-17 DOI: 10.17212/1814-1196-2018-3-33-42

A. Voevoda, D. Romannikov

引用次数: 0