Scientific Journal of KubSAU

Polythematic online scientific journal
of Kuban State Agrarian University
ISSN 1990-4665
AGRIS logo UlrichsWeb logo DOAJ logo

Name

Lutsenko Yevgeniy Veniaminovich

Scholastic degree


Academic rank

professor

Honorary rank

Organization, job position

Kuban State Agrarian University
   

Web site url

lc.kubagro.ru

Email

prof.lutsenko@gmail.com


Articles count: 266

Sort by: Date Title Views
9686 kb

FORECASTING THE NUMBER AND THE CLASSES OF SOLAR FLARES BASED ON THEIR BACKGROUND ACCORDING TO THE UCI REPOSITORY USING ASCANALYSIS AND "EIDOS" INTELLIGENT SYSTEMS

abstract 1041410099 issue 104 pp. 1328 – 1389 30.12.2014 ru 566
The article describes a numerical example of creating intellectual application designed to predict solar flares of different classes on the basis of the history of their development in the environment of "Eidos" system. As the source data, we used the database of UCI repository
3166 kb

MODIFICATION OF THE METHOD OF WEIGHTED LEAST SQUARES BY USING THE OBSERVATIONS OF THE AMOUNT OF INFORMATION IN THE ARGUMENT OF THE VALUE FUNCTION AS WEIGHTS (ALGORITHM AND SOFTWARE IMPLEMENTATION)

abstract 1041410100 issue 104 pp. 1390 – 1440 30.12.2014 ru 735
The method of ordinary least squares (OLS) is widely known and deservedly popular. However, some attempts to improve this method. The result of one of such attempts is the weighted least squares (WMNC), the essence of which is to give the observation a weight which is inversely proportional to the errors of their approximation. Thereby, in fact, monitoring is ignored the more the difficult to approximate it. The result of this approach, formally, is the approximation error decreasing, but in fact, this occurs by partial refusal to consider the "problem" of observations, making a big mistake. If the idea underlying WMNC to bring to the extreme (and absurd), then in the limit, this approach will lead to the fact that from the entire set of observations there will be only those that lie almost exactly on the trend obtained by the method of least squares, and the rest will simply be ignored. However, according to the author, it's not a problem, and the failure of its decision, though it might look like a solution. In the work we have proposed a solution, based on the theory of information: to consider the weight of observations, the number of the argument of the value function. This approach was validated in the framework of a new innovative method of artificial intelligence: methods for automated system-cognitive analysis (ASA-analysis) and implemented 30 years ago in its software toolkit, which is "Eidos" intelligent system in the form of so-called "cognitive functions". This article presents an algorithm and software implementation of this approach, illustrated in detailed numerical example. In the future it is planned to give a detailed mathematical basis of the method of weighted least squares, which is modified by the application of information theory to calculate the weights of the observations, and investigate its properties
615 kb

MODIFICATION OF WEIGHTED LEAST SQUARES BY USING THE OBSERVATIONS OF THE AMOUNT OF INFORMATION IN THEM AS WEIGHTS (MATHEMATICAL ASPECTS)

abstract 1051501050 issue 105 pp. 813 – 844 30.01.2015 ru 752
This article briefly discusses the mathematical nature of the author's proposed modification of the weighted least squares, in which the amount of the data is used as the weights of observations. There are two variants of this modification. In the first one, the weighting of the observations was made by replacing one observation with a certain amount of the information in it by the corresponding number of observations for unit weight, and then we applied the standard method of least squares. In the second method, the weighting of the observations was performed for each value of the argument by replacing all observations with a certain amount of information in one observation of unit weight which had been obtained as a weighted average of them, and then we applied the standard method of least squares. We have described in detail the technique of numerical calculations of the amount of information in the observations, based on the theory of automated system-cognitive analysis (ASC-analysis) and implemented it with a help of software tools - intelligent system called "Eidos". The article provides an illustration of the proposed approach on a simple numerical example. In the future, we are planning to give more detailed mathematical basis of the method of weighted least squares, which is modified by using the amount of information as weights, but also to explore its properties
3033 kb

SOLVING PROBLEMS OF STATISTICS WITH THE METHODS OF INFORMATION THEORY

abstract 1061502001 issue 106 pp. 1 – 47 28.02.2015 ru 842
The article presents a theoretical substantiation, methods of numerical calculations and software implementation of the decision of problems of statistics, in particular the study of statistical distributions, methods of information theory. On the basis of empirical data by calculation we have determined the number of observations used for the analysis of statistical distributions. The proposed method of calculating the amount of information is not based on assumptions about the independence of observations and the normal distribution, i.e., is non-parametric and ensures the correct modeling of nonlinear systems, and also allows comparable to process heterogeneous (measured in scales of different types) data numeric and non-numeric nature that are measured in different units. Thus, ASC-analysis and "Eidos" system is a modern innovation (ready for implementation) technology solving problems of statistical methods of information theory. This article can be used as a description of the laboratory work in the disciplines of: intelligent systems; knowledge engineering and intelligent systems; intelligent technologies and knowledge representation; knowledge representation in intelligent systems; foundations of intelligent systems; introduction to neuromaturation and methods neural networks; fundamentals of artificial intelligence; intelligent technologies in science and education; knowledge management; automated system-cognitive analysis and "Eidos" intelligent system which the author is developing currently, but also in other disciplines associated with the transformation of data into information, and its transformation into knowledge and application of this knowledge to solve problems of identification, forecasting, decision making and research of the simulated subject area (which is virtually all subjects in all fields of science)
18043 kb

SYNTHESIS AND VERIFICATION OF MULTICRITERIA SYSTEM-COGNITIVE MODEL OF THE GUARDIAN UNIVERSITY RANKING AND ITS APPLICATION FOR THE PROPER EVALUATION OF THE EFFECTIVENESS OF RUSSIAN UNIVERSITIES WITH RESPECT TO THE DIRECTION OF PREPARATION

abstract 1071503001 issue 107 pp. 1 – 62 31.03.2015 ru 365
The article is devoted to the solution of the problem which is the fact that on the one hand, the rating of Russian universities is in demand and on the other hand it hasn’t been created yet. The proposed idea of solving the problem consists in the application of domestic licensing of innovative intelligent technologies for these purposes: we have suggested using an automated system-cognitive analysis (ASC-analysis) and its software tools – the intelligent system called "Eidos". These methods are described in detail in this context. It is proposed to consider the possibility of applying these tools on the example of the Guardian University ranking. The article discusses its private criteria (indicators of universities). We specify the sources of data and the methods of their preparation for processing in "Eidos" system. In accordance with ASC-analysis methodology the article describes the installation of "Eidos", the data input into it, and the formalization of the subject area, synthesis and verification of models, their display and use to solve problems of assessment of the Guardian rating for Russian universities and research object modeling. It also discusses the prospects and ways of development of the integrated rating of Russian universities and operation of rating in adaptive mode. We have also specified the limitations of the proposed approach and the prospects of its development
561 kb

HIRSCH-MANIA WHEN EVALUATING THE RESULTS OF SCIENTIFIC ACTIVITY, ITS NEGATIVE EFFECTS AND AN ATTEMPT TO OVERCOME THEM WITH THE USE OF A MULTI-CRITERIA APPROACH AND INFORMATION THEORY

abstract 1081504001 issue 108 pp. 1 – 29 30.04.2015 ru 786
Recently, the process of monetization of the evaluation of scientific activity was initiated, and there is a need for quantitative methods and comparable assessment of the effectiveness and quality of work of a scientist. There are numerous methods to reward for these results. What is common to all these techniques covered is the role of the Hirsch index or h-index. By itself, this index is well founded. However, in connection with the practice of application of h-index in our environment in the minds of the scientific community it has started some kind of mania, which the author proposes to call "Hirsch-mania". This mania is characterized by elevated unhealthy interest to the value of the Hirsch index, especially inadequate artificial exaggeration of this value, as well as a number of negative implications of this interest. In this article we have made an attempt to briefly describe some of the negative effects of this new mental infection that hit the public consciousness of the scientific community. And also we want to identify ways of overcoming at least some of their causes. This is the problem solved in this work. To solve the formulated problem, we propose to apply multi-criteria approach based on information theory, namely those options, which are implemented in an automated system-cognitive analysis (ASC-analysis) and its software tools - intelligent system called "Eidos
5831 kb

INTELLIGENT MANAGEMENT OF THE QUALITY OF SYSTEMS BY SOLVING A GENERALIZED ASSIGNMENT PROBLEM WITH THE USE OF ASC-ANALYSIS AND "EIDOS-X++" SYSTEM

abstract 1091505001 issue 109 pp. 1 – 51 29.05.2015 ru 752
The quality of a system is seen as an emergent property of systems, due to their composition and structure, and it reflects their functionality, reliability and cost. Therefore, when we speak about quality management, the purpose of management is the formation of pre-defined system properties of the object of management. The stronger the object of the control expresses its system properties, the stronger the nonlinearity manifests of the object: both the dependence of the management factors from each other, and the dependence of the results of the action of some factors from the actions of others. Therefore, the problem of quality management is that in the management process the management object itself changes qualitatively, i.e. it changes its level of consistency, the degree of determinism and the transfer function itself. This problem can be viewed as several tasks: First is the system identification of the condition of the object of management, 2nd – making decisions about controlling influence that changes the composition of the control object in a way its quality maximally increases at minimum costs. To solve the 2nd problem we have proposed an application of the component selection of the object by functions based on the resources allocated for the implementation of different functions; costs associated with the choice of the components and the degree of compliance of various components to their functional purpose. In fact, we have proposed a formulation and a solution of the new generalization of a variant of the assignment problem: "multi backpack", which differs from the known with the fact that the selection has been based not only on the resources and costs, but also with taking into account the degree of compliance of the components to their functional purpose. A mathematical model, which provides a solution to the 1st problem, and reflecting the degree of compliance of the components to their functionality, as well as the entire decision-making process for selections, i.e. 2nd task, has been implemented in the ASC-analysis and in the system called "Eidos" X++". The article also provides a simplified numerical example of the proposed approach with the selection of staff members
6639 kb

ASC-ANALYSIS OF THE IMPACT OF ENVIRONMENTAL FACTORS ON THE QUALITY OF LIFE OF THE POPULATION OF THE REGION

abstract 1101506001 issue 110 pp. 1 – 37 30.06.2015 ru 913
Without science it would be impossible to form a full environmental consciousness. To increase the validity and weight of the findings on the impact of environment on quality of life, it is necessary to quantify the strength and direction of the influence of diverse environmental factors. However, it appears that this is quite problematic for a number of reasons. First, it is the lack or inaccessibility of source of data which is necessary for such type of research. The same data, which still can be found cover just small periods of observations (small longitudinal research data), and their completion, including performing experiments, is fundamentally impossible. As a result, it is impossible to require such full data replications, which is a necessary condition for correct applying of factor analysis. Secondly, environmental factors are described with heterogeneous indices measured in different types of measurement scales (nominal, ordinal and numerical) and in different measurement units. Mathematical methods of comparable processing of such data, and the right software tools for these methods, generally speaking, do not exist. Third, these tasks are large-scale problems, i.e. they are not talking about 5 or max 7 factors as it was in factor analysis, but about hundreds and thousands. Fourthly, the original data is noisy and require sustainable methods. Fifthly, environmental factors are interrelated and require nonlinear nonparametric approaches. To solve these problems it is proposed to apply a new innovative intelligent technology: automated system-cognitive analysis and its software tool – a system called "Eidos". We have also given a brief numerical example of assessing the impact of environmental factors on life expectancy and causes of death
6248 kb

AUTOMATED SYSTEMIC-COGNITIVE ANALYSIS OF CONTOURS OF IMAGES (generalization, abstraction, classification and identification)

abstract 1101506009 issue 110 pp. 138 – 168 30.06.2015 ru 812
In the article the application of systemic-cognitive analysis, its mathematical model - the system theory of the information and its program toolkit - "Eidos" system for synthesis of the generalized images of classes, their abstraction, classification of the generalized images (clusters and constructs) comparisons of concrete images with the generalized images (identification) are examined. We suggest a new approach to the digitization of images, based on the use of the polar coordinate system, the center of gravity of the image and its contour. Before digitizing images we can use their changes to standardize the position of the picture-frames, their size and rotation. Therefore, if you specify this option, the results of digitization and image ASC-analysis can be invariant (independent) to their position, size and rotation. This means that in the model on the basis of a number of specific examples we will create one image of each class of images, independent of their specific implementations, i.e., the "Eidos" of these images (in the sense of Plato) - a prototype or archetype (in the Jungian sense) images. But the "Eidos" system provides not only the formation of prototype images, which quantitatively reflects the amount of information in the image elements of the prototype, but the removal of all irrelevant to identification (abstraction), and the comparison of specific images with generic (identification) and the generalized images of images together (classification). The article provides a detailed numerical example of ASC- analysis of images
2759 kb

SIMULATING AND PREDICTING GLOBAL CLIMATIC ANOMALIES SUCH AS EL NINO AND LA NINA

abstract 1101506102 issue 110 pp. 1546 – 1578 30.06.2015 ru 847
The paper discusses the modeling and prediction of the climate of our planet with the use of artificial intelligence AIDOS-X. We have developed a number of semantic information models, demonstrating the presence of the elements of similarity between the motion of the lunar orbit and the displacement of the instantaneous pole of the Earth. It was found that the movement of the poles of the Earth leading to the variations in the magnetic field, seismic events, as well as violations of the global atmospheric circulation and water, and particular to the emergence of episodes such as El Niño and La Niña. Through semantic information models studied some equatorial regions of the Pacific Ocean, as well as spatial patterns of temperate latitudes, revealed their relative importance for the prediction of global climatic disturbances in the tropical and temperate latitudes. The reasons of occurrence of El Niño Modoki and their relationship with the movement of elements of the lunar orbit in the long-term cycles are established. Earlier, we had made a forecast of the occurrence of El Niño episode in 2015. Based on the analysis of semantic models concluded that the expected El Niño classical type. On the basis of the prediction block AIDOS-X calculated monthly evolution scenario of global climate anomalies. In this paper, the analysis of the actual implementation forecast of El Niño since its publication in January 2015 - before June 2015. It is shown that the predicted scenario of climatic anomalies actually realized. Calculations of future climate scenarios with system «Aidos-X» recognition module indicate that further possible abnormal excess temperature indicators of surface ocean waters in regions Nino 1,2 and Nino3,4 for 2015 may be comparable with similar abnormalities in the catastrophic El Niño of 1997-1998.
.