Scientific Journal of KubSAU

Agronomy systems with good reason can be considered as complex multiparameter natural and technical systems. In these systems, there are numerous and diverse physical, chemical and biological processes. On the one hand, these processes have a significant impact on the performance of these systems. On the other hand, they are extremely difficult to be described in the form of meaningful analytical models based on equations. As a result, the development of meaningful analytical models is associated with a large number of simplifying assumptions that reduce the validity of these models. Usually we consider linear univariate models for agronomic systems, whereas practices are necessary for nonlinear multiparameter models. Thus, we face the problem proposed to be solved by the application of a phenomenological meaningful systemic cognitive models. These models are created using automated system-cognitive analysis (ASC-analysis) using the intellectual system called "Eidos" directly based on empirical data and used for the decision of tasks of forecasting, decision support and research of the modeled subject area. In this case, empirical data can be large, incomplete (fragmented), noisy, presented in different types of measuring scales (nominal, ordinal and numerical) and in different units of measurement. The comparability of the processing of heterogeneous data is ensured by the fact that they are all converted into units of measurement of the amount of information. A numerical example has been given

In the article, on a small numerical example, we consider the similarity and difference of symptoms and syndromes according to their diagnostic meaning, i.e. according to the information they contain about the belonging of conditionals of animals to different nosological images. This problem can be solved for veterinary with the use of a new method of agglomerative cognitive clustering, implemented in Automated System-Cognitive analysis (ASC-analysis). This method of clustering differs from the known traditional methods in: a) in this method, the parameters of the generalized image of the cluster are calculated not as averages from the original objects (symptoms) or their center of gravity, but are determined using the same basic cognitive operation of ASC-analysis, which is used to form generalized images of the classes based on examples of objects and which really correctly provides a generalization; b) the similarity criterion is not the Euclidean distance or its variants, but the integral criterion of non-metric nature: "the total amount of information", the application of which is theoretically correct and gives good results in unortonormated spaces, which are usually found in practice; c) cluster analysis is carried out not on the basis of initial variables, frequency matrices or matrix of similarity (differences), depending on the units of measurement on the axes (measurement scales), but in cognitive space, in which one unit of measurement is used for all axes: the amount of information, and therefore the results of clustering do not depend on the initial units of measurement of features of objects. All this allows us to get the results of clustering, understandable to specialists and amenable to meaningful interpretation, well-consistent with the experts ' assessments, their experience and intuitive expectations, which is often a problem for classical clustering methods

Techniques of value analysis and "Direct-costing" are well-known and popular. The ideas and principles of value analysis and the method of "Direct costing" are very similar, if not identical. On the one hand, these ideas are very reasonable, well grounded theoretically and proved its effectiveness in practice. On the other hand, the wide use of these methods is hampered by the difficulty of obtaining large amounts of detailed technological and financial-economic information, as well as the need for careful research by competent professionals, well-versed in substantive subject area. This is the contradiction between the desire to apply the methods of the value analysis and "Direct costing" and difficulty to perform it in practice. This contradiction constitutes a real problem and may often be discouraging and frustrating. In this work, we propose a simple and effective solution to this problem, theoretically well-informed with all the necessary methodological and software tools and widely and successfully tested in practice. The proposed solution is based on two simple ideas: 1) instead of collecting and holding a meaningful large amount of technological and financial-economic information we might apply approaches, pleasant management theory; 2) to create systems for automated control of natural and financial-economic efficiency of expenses we might use the automated system-cognitive analysis and its software tool – an intellectual system called "Eidos". In the name of the specialty 08.00.05 – Economics and national economy management, there are such words: "management of enterprises, branches, complexes, innovation." The use of the term "Management" implies that there is a model that reflects the influence of factors on the object of control, and there is the management system making decisions based on this model. However, as a rule, the dissertations in this field have nothing of this, except only financial and economic calculations. The article proposes an approach based on the control theory, removing this disadvantage

From a huge number of the organisms inhabiting our planet, insects make 70%, being the most numerous of the invertebrate animal classes numbering more than 2 million types. It is difficult to find such place where it would be impossible to meet representatives of this huge class. They completely took over the entire environment - water, the land, air. For them, it is the common characteristic: complex instincts, omnivorous, high fecundity, and for some of them – a public way of life. Insects can be found at tremendous heights, reaching the level of 5000 meters, and they inhabit the desert where it practically never rains, not to mention the absence of any vegetation. Deep caves where no sunlight, nor the conditions for food and existence of living organisms — it is also the habitat of insects, they can be found far beyond the Arctic circle, and even on many Islands of Antarctica, where in addition to lifeless rock, it would seem that there is nothing else. Among insects, one of the largest and most numerous families are the ground beetles (Carabidae). They subtly respond to changes in soil and vegetation, hydrothermal and micro-climatic conditions of the environment, which makes them a convenient model subject to various environmental and Zoological researches. Ground beetles belong to a large number of genera and species, often difficult to see, in this regard, we use many different signs to diagnose. We have taken into consideration the coloration, body shape, external structure, surface structure, size, and arrangement of the genitals and chaetotaxy. Due to the fact, that the number of ground beetles is enormous, and, using their appearance, it is very difficult to determine their generic identity, there is a need of automation of the identification process, due to which we require a special mechanism that would increase the accuracy of these insects. In the previous work of the authors (http://ej.kubagro.ru/2016/05/pdf/01.pdf) we considered the further possibility of using the method of ASC- analysis to classify insects, not only in species but also in genera, orders, thereby increasing the reliability of determination of ground beetles, which will be done in this article. A numerical example is given. We also have gained a successful experience of solving such problems in other subject areas. This article can be considered as a continuation of the series of works dedicated to governmental use of the automated system-cognitive analysis (ASC-analysis) and its software tools – the system of "Eidos"

The creation of artificial intelligence systems is one of important and perspective directions of development of modern information technology. As there are many alternatives to artificial intelligence systems, there is a need to evaluate mathematical models of these systems. In this work, we consider a solution of the problem of identifying classes of levels of pay of employees on their characteristics. To achieve this goal, it requires free access to test the source data and methodology, which will help to convert the data into the form needed for work in artificial intelligence systems. A good choice is the databases from the site: http://allexcel.ru/gotovyetablitsy-excel-besplatno. In this work, we have used the database called "The database table of employees, payments calculation". The most reliable in this application was the model of the INF4 based on semantic appropriate measure of information of A. Kharkevich with integral criteria of "Amount of knowledge". The accuracy of the model is 0.960, which is much higher than the reliability of expert evaluations, which is equal to about 70%. To assess the reliability of the models in the ACS-analysis and the system called "Eidos" we have used F-criterion of van Ritbergen and fuzzy multiclass generalization proposed by Professor E. V. Lutsenko

The creation of artificial intelligence systems is one of important and perspective directions of development of modern information technology. As there are many alternatives to artificial intelligence systems, there is a need to evaluate mathematical models of these systems. In this article, we consider a solution of the problem of identifying classes of levels of pay to employees on their characteristics. To achieve this goal it requires free access to test the source data and methodology, which will help to convert the data into the form needed for work in artificial intelligence systems. A good choice is a database of test problems for systems of UCI artificial intelligence repository. In this work we have used data base on teaching effectiveness for three regular semesters and two summer semesters of 151 teaching assistant (TA) assignments at the statistics Department of the University of Wisconsin-Madison. The most reliable in this application was the model of the INF4. The accuracy of the model in accordance with Lmeasure made up 0,809, which is much higher than the reliability of expert evaluations, which is equal to about 70%. To assess the reliability of the models in the ASC-analysis and in the system of "Eidos" we use F-criterion of van Ritbergen and its fuzzy multiclass generalization proposed by Professor E. V. Lutsenko

Classic quantitative measure of the reliability of the models: F-measure by van Rijsbergen is based on counting the total number of correctly and incorrectly classified and not classified objects in the training sample. In multiclass classification systems, the facility can simultaneously apply to multiple classes. Accordingly, when the synthesis of the model description is used for formation of generalized images of many of the classes it belongs to. When using the model for classification, it is determined by the degree of similarity or divergence of the object with all classes, and a true-positive decision may be the membership of the object to several classes. The result of this classification may be that the object is not just rightly or wrongly relates or does not relate to different classes, both in the classical F-measure, but rightly or wrongly relates or does not relate to them in varying degrees. However, the classic F-measure does not count the fact that the object may in fact simultaneously belongs to multiple classes (multicrossover) and the fact that the classification result can be obtained with a different degree of similarity-differences of object classes (blurring). In the numerical example, the author states that with true-positive and true-negative decisions, the module similarities-differences of the object classes are much higher than for false-positive and false-negative decisions. It would therefore be rational to the extent that the reliability of the model to take into account not just the fact of true or false positive or negative decisions, but also to take into account the degree of confidence of the classifier in these decisions. In classifying big data we have revealed a large number of false-positive decisions with a low level of similarity, which, however, in total, contribute to reducing the reliability of the model. To overcome this problem, we propose a L2-measure, in which instead of the sum of levels of similarity we use the average similarity by different classifications. Thus, this work offers measures of the reliability of the models, called L1-measure and the L2 measure, mitigating and overcoming the shortcomings of the F-measures; these measures are described mathematically and their application is demonstrated on a simple numerical example. In the intellectual system called "Eidos", which is a software toolkit for the automated system-cognitive analysis (ASC-analysis), we have implemented all these measures of the reliability of the models: F, L1 and L2

In forensics there is an urgent need to determine the type of rifle (automatic, rifle, large caliber pistol) depending on its used ammunition found at the scene of the use of weapons. We offer a solution to this problem with the use of new innovative method of artificial intelligence: automated system-cognitive analysis (ASC-analysis) and its program toolkitwhich is a universal cognitive analytical system called "Eidos". In the "Eidos" system we have implemented the software interface that allows posting of images and identifying their outer contours. By multivariable typing, the system creates a systemic-cognitive model, the use of which, if the model is sufficiently accurate, may be helpful in solving problems of system identification, prediction, classification, decision support and research of the modeled object by studying its model. For this task the following stages: 1) input images of ammunitions into the "Eidos" system and creation of their mathematical models; 2) the synthesis and verification of the models of generalized images of ammunition for types of weapons based on the contour images of specific munitions (multiparameter typing); 3) improving the quality of the model by separating classes for typical and atypical parts; 4) quantification of the similarities-the differences between specific types of munitions with generic images of different types of ammunition of the weapon (system identification); 5) quantification of the similarity-differences between types of ammunition, i.e. cluster-constructive analysis of generalized images of ammunition. A numerical example is given. We also possess a successful experience of solving similar problems in other subject areas

The article deals with the similarity and difference of nosological images in veterinary medicine using a new method of agglomerative clustering implemented in Automated system-cognitive analysis (ASC-analysis) on a small numerical example. This method is called Agglomerative cognitive clustering. This method differs from the known traditional facts: a) parameters of a generalized image of the cluster are computed not as averages from the original objects (classes) or their center of gravity, and are defined using the same underlying cognitive operations of ASC-analysis, which is used for the formation of generalized images of the classes on the basis of examples of objects and which is really correct and provides a synthesis; b) as a criterion of similarity we do not use Euclidean distance or its variants, and the integral criterion of non-metric nature: "the total amount of information", the use of which is theoretically correct and gives good results in non-orthonormal spaces, which are usually found in practice; c) cluster analysis is not based on the original variables, matrices of frequency or a matrix of similarities (differences) dependent on the measurement units of the axes, and in the cognitive space in which all the axes (descriptive scales) use the same unit of measurement: the quantity of information, and therefore, the clustering results do not depend on the original units of measurement features. All this makes it possible to obtain clustering results that are understandable to specialists and can be interpreted in a meaningful way that is in line with experts' assessments, their experience and intuitive expectations, which is often a problem for classical clustering methods

The quality of life of the population of the region is an important integral criterion of estimation of efficiency of activity of regional administration. The most important strategic sector of the economy of the Krasnodar region is the agro-industrial complex (AIC). This poses the problem of management of the quality of life of the region through the use of as the control factor of the volume and direction of investment in agriculture