   |
  |
||
|
|||
 |
|||
“Informatics and Applications” scientific journalVolume 13, Issue 4, 2019Content Abstract and Keywords About Authors DIGITAL MODEL OF THE AIRCRAFT'S WEIGHT PASSPORT
Abstract: The paper is devoted to the problem of the digital modeling of an aircraft's weight passport. A weight passport is developed at the stage of a new product's design and accompanies it during all other stages of the life cycle. A digital weight passport plays the most important role when the released product is being operated.
Keywords: digital model; design automation; aircraft; weight design; weighting model; design tree; project generator USING THE MODEL OF GAMMA DISTRIBUTION IN THE PROBLEM OF FORMING A TIME-LIMITED TEST IN A DISTANCE LEARNING SYSTEM
Abstract: For the distance learning systems, consideration is given to generation of individual tasks with minimization of execution time. As a criterion, the convolution of two weighted normalized values associated with the deviation of the complexity of the generated test from the specified level and the quantile of the test execution time is used. The gamma distribution model is used to describe a model of a student's random response time to a task. An algorithm is proposed for estimating the parameters of the gamma distribution for each task. It is assumed that task complexities are determined either by an expert or by using corresponding algorithms based on the Rush model. The results of a numerical experiment are presented. Keywords: distance learning system; statistical analysis; adaptive systems; quantile optimization ON COMPARATIVE EFFICIENCY OF CLASSIFICATION SCHEMES IN AN ENSEMBLE OF DATA SOURCES USING AVERAGE MUTUAL INFORMATION
Abstract: Given ensemble of data sources and different fusion schemes, an accuracy of multiclass classification of the collections of the source objects is investigated. Using the average mutual information between the datasets of the sources and a set of the classes, a new approach to comparing lower bounds to an error probability in two fusion schemes is developed. The authors consider the WMV (Weighted Majority Vote) scheme which uses a composition of the class decisions on the objects of the individual sources and the GDM (General Dissimilarity Measure) scheme based on a composition of metrics in datasets of the sources. For the above fusion schemes, the mean values of the average mutual information per one source are estimated. It is proved that the mean in the WMV scheme is less than the similar mean in the GDM scheme. As a corollary, the lower bound to the error probability in the WMV scheme exceeds the similar bound to the error probability in the GDM scheme. This theoretical result is confirmed by experimental error rates in face recognition of HSI color images that yield the ensemble of H, S, and I sources. Keywords: multiclass classification; ensemble of sources; fusion scheme; composition of decisions; composition of metrics; average mutual information; error probability DATA MODEL SELECTION IN MEDICAL DIAGNOSTIC TASKS
Abstract: Effective solution of medical diagnostics tasks requires the use of complex probabilistic models which allow one to adequately describe real data and permit the use of analytical methods of the supervised learning classification. Choosing a model of a mixture of normal distributions solves the posed problems but leads to the curse of dimensionality. The transition to the model of a mixture of probabilistic principal component analyzers allows one to formally set the task of choosing its structural parameters. The solution is proposed to search by combining the application of information criteria for the formation of initial approximations followed by refinement of the resulting estimates. Using the example of experiments to diagnose liver diseases and to predict the chemical composition of urinary stones, the capabilities of the described data analysis procedures are demonstrated. The proposed solutions give a source of improving the accuracy of classification, impetus to experts in the subject area to clarify the essence of the processes. Keywords: medical diagnostics; mixture of probabilistic principal component analyzers; model selection criterion; cross validation RESEARCH OF THE POSSIBILITY TO FORECAST CHANGES IN FINANCIAL STATE OF A CREDIT ORGANIZATION ON THE BASIS OF PUBLIC FINANCIAL STATEMENTS
Abstract: The mathematical model for forecasting of license revocation of a credit organization in the 6-month period based on public financial statements is considered. The model represents an ensemble of combinatorial and logical methods and decision trees of different types. Its effectiveness estimated by ROC AUC (area under receiver operating characteristic curve) is 0.74. The model allows distinguishing groups of credit organizations with higher and lower license revocation risks. Also, the ranking of different financial statement indicators has been performed which marked the importance of liquid and highly liquid assets. Keywords: forecasting; algorithm ensembles; financial state; credit organization THEORETICAL FOUNDATIONS OF CONTINUOUS VaR CRITERION OPTIMIZATION IN THE COLLECTION OF MARKETS
Abstract: The work continues studying the problems of using continuous VaR criterion (CC-VaR) in financial markets. The application of CC-VaR in a collection of theoretical markets of different dimensions that are mutually connected by their underliers is concerned. In a typical model of the collection of one two-dimensional market and two one-dimensional markets, the most general case of their conjoint functioning is considered.
Keywords: underliers; risk preferences function; continuous VaR criterion; cost and forecast densities; return relative function; Newman-Pearson procedure; combined portfolio; randomization; surrogate portfolio; idealistic portfolio THE OUTPUT STREAMS IN THE SINGLE SERVER QUEUEING SYSTEM WITH A HEAD OF THE LINE PRIORITY
Abstract: The paper studies a single server queuing system with two types of customers, head of the line priority, and an infinite number of positions in the queue. The arrival stream of customers of each type is a Poisson stream.
Keywords: output stream; head of the line priority; embedded Markov chain; single server THE MEAN SQUARE RISK OF NONLINEAR REGULARIZATION IN THE PROBLEM OF INVERSION OF LINEAR HOMOGENEOUS OPERATORS WITH A RANDOM SAMPLE SIZE
Abstract: The problems of constructing estimates from observations, which represent a linear transformation of the initial data, arise in many application areas, such as computed tomography, optics, plasma physics, and gas dynamics. In the presence of noise in the observations, as a rule, it is necessary to apply regularization methods. Recently, the methods of threshold processing of wavelet expansion coefficients have become popular. This is explained by the fact that such methods are simple, computationally efficient, and have the ability to adapt to functions which have different degrees of regularity at different areas. The analysis of errors of these methods is an important practical task, since it allows assessing the quality of both the methods themselves and the equipment used. When using threshold processing methods, it is usually assumed that the number of expansion coefficients is fixed and the noise distribution is Gaussian. This model is well studied in literature and optimal threshold values are calculated for different classes of signal functions. However, in some situations, the sample size is not known in advance and has to be modeled by a random variable. In this paper, the author considers a model with a random number of observations containing Gaussian noise and estimates the order of the mean-square risk with an increasing sample size. Keywords: wavelets; threshold processing; linear homogeneous operator; random sample size; mean square risk MIXED POLICIES FOR ONLINE JOB ALLOCATION IN ONE CLASS OF SYSTEMS WITH PARALLEL SERVICE
Abstract: Consideration is given to the problem of efficient job allocation in the class of systems with parallel service on independently working single-server stations each equipped with the infinite capacity queue. There is one dispatcher which routes jobs, arriving one by one, to servers. The dispatcher does not have a queue to store the jobs and, thus, the routing decision must be made on the fly. No jockeying between servers is allowed and jobs cannot be rejected. For a job, there is the soft deadline (maximum waiting time in the queue). If the deadline is violated, a fixed cost is incurred and the job remains in the system and must be served. The goal is to find the job allocation policy which minimizes both the job's stationary response time and probability of job's deadline violation. Based on simulation results, it is demonstrated that the goal may be achieved (to some extent) by adopting a mixed policy, i. e. a proper dispatching rule and the service discipline in the server. Keywords: parallel service; dispatching policy; service discipline; sojourn time; deadline violation DISCRETE-TIME Geo/G/l/infinity LIFO QUEUE WITH RESAMPLING POLICY
Abstract: Consideration is given to the problem of estimation of the true stationary mean response time in the discrete-time single-server queue of infinite capacity, with Bernoulli input, round-robin scheduling, and inaccurate information about the service time distribution which is considered to be general arithmetic. It is shown that the upper bound for the true value may be provided by the mean response time in the discrete-time single-server queue with LIFO (last in, first out) service discipline and resampling policy. The latter implies that a customer arriving to the nonidle system assigns new remaining service time for the customer in the server. For the case when the true service time distribution is geometric and the error in the service times is of multiplicative type, conditions are provided which, when satisfied, guarantee that the proposed method yields the upper bound across all possible values of the system's load. Keywords: discrete time; inverse service order; inaccurate service time; round robin scheduling; resampling policy NUMERICAL SCHEMES OF MARKOV JUMP PROCESS FILTERING GIVEN DISCRETIZED OBSERVATIONS I: ACCURACY CHARACTERISTICS
Abstract: The note is the initial in the series of the papers devoted to the numerical realization of the optimal state filtering of Markov jump processes given the indirect observations corrupted by the additive and/or multiplicative Wiener noises. This problem is solved by the time discretization of the observations with their subsequent processing.
Keywords: Markov jump process; optimal filtering; additive and multiplicative observation noises; stochastic differential equation; analytical and numerical approximation ON THE REPRESENTATION OF GAMMA-EXPONENTIAL AND GENERALIZED NEGATIVE BINOMIAL DISTRIBUTIONS
Abstract: For more than a century and a half, gamma-type distributions have shown their adequacy in modeling real processes and phenomena. Over time, designs using distributions from the gamma family are becoming more complex in order to improve the applicability of mathematical models to relevant aspects of life. The paper presents a number of results both generalizing and simplifying some classical forms used in the analysis of large-scale and structural mixtures of generalized gamma laws. The gamma-exponential distribution is introduced and its characteristics are described. An explicit form for integral representations of partial probabilities of the generalized negative binomial distribution is given. The results are formulated in terms of the gamma exponential function. The obtained results can be widely used in models that use scale and structural mixtures of distributions with positive unrestricted support to describe processes and phenomena. Keywords: gamma exponential function; generalized gamma distribution; generalized negative binomial distribution; gamma-exponential distribution; mixed distributions CONCEPTS FORMING ON THE BASIS OF SMALL SAMPLES
Abstract: Monitoring systems of information security of information systems obtain information in the form of chains of short messages which can be considered as chains of small samples. Often, owing to an inertance of information systems, these chains reflect close statuses of the computing system or network. In the paper, it is supposed that work of the system can be presented in the form of a finite set of modes which are called concepts. Violations of security are detected by means of anomalies that are associated with emergence of new concepts. The known technologies of identification of anomalies are based on creation of a model of a normal system's behavior. Concepts correspond to normal types of a system's behavior. In the paper, the problem of creation of concepts on the basis of machine learning based on chains of small samples is considered. The algorithm of concepts forming is constructed and its efficiency is proved. Keywords: information security monitoring; small samples; small sample learning; concepts forming USING METADATA TO IMPLEMENT MULTILEVEL SECURITY POLICY REQUIREMENTS
Abstract: A distributed information computing system which objects contain both valuable information (or are themselves valuable) and open (non-valuable) information is considered. To protect valuable information, multilevel security (MLS) policy is used that prohibits information flows from objects with valuable information to objects with open information. Objects with valuable information form a class of high-level objects, and objects with open information form a class of low-level objects. Metadata is created to manage network connections. Metadata is a simplification of mathematical models of business processes and is the basis of a permission system for host connections in a distributed information computing system. The paper constructs MLS security policy rules, and based on metadata-related infrastructure, shows the ability to implement this security policy in the distributed information computing system. The only trusted process required to implement the MLS security policy is at the connection management level. This layer is unrelated to the data plane and can be isolated to ensure its information security. Keywords: MLS security policy; information flows; metadata TEMPORAL DATA IN LEXICOGRAPHIC DATABASES
Abstract: The paper describes an approach to design of the Lexicographic Knowledge Base (LKB), which aims to fulfill two interrelated tasks: (i) goal-oriented development of linguistic typologies; and (ii) creation and updating of electronic bilingual dictionaries based on the developed typologies. In the LKB, some of the fields assigned by lexicographers to represent new knowledge on words' meanings (concepts) and translations are temporal. The content of these fields is time-dependent because lexicographers can change description of concepts with time.
Keywords: lexicographic knowledge base; temporal structure of a dictionary entry; bilingual dictionaries; linguistic typologies; parallel texts; evolution of concepts DIGITAL ENCODING OF CONCEPTS
Abstract: The tasks of encoding concepts of human knowledge in the digital medium of computers and networks are of particular relevance in connection with the widespread use of artificial intelligence systems in the world. In the process of expanding the scope of their applications, the range of categories of encoded concepts is increasing.
Keywords: knowledge encoding; polyadic computing; digital medium; artificial intelligence; categories of concepts; socialization of knowledge concepts UNDERSTANDING OF COMPLEX SYSTEMS USING THE LAWS OF SYNERGETICS AND INFORMATICS
Abstract: The author shows how the laws of informatics and synergetics can be used to explain the genesis and evolution of such a complex natural system as petroleum. When one creates the matrix of hydrocarbon molecules using the laws of informatics, the latter imply the ambiguity in the quantum behavior of the electrons. This dynamic and static uncertainty comes into play during the oil field location process. Consideration is given to the laws of synergetics, which demonstrate the self-organization ability of the molecules. A new type of molecules is formed in the hydrocarbon fluid near the bifurcation points, associated with the variation of the thermodynamics, structure, and mixture of the geological environment. In the analysis of the petroleum formation process, consideration is also given to the notion of attractor. It serves as the basin of attraction for all hydrocarbon molecules, in which the exact petroleum molecular composition is formed. Keywords: synergetics; petroleum formation; informatics and oil location; bifurcation and composition of hydrocarbon molecules; attractor and petroleum molecular composition
|
 | |
 Phone of the Center: +7 (499) 135-62-60E-mail of the Center: ipiran@ipiran.ru | 
RUS
|