РУС ENG

Problems of Artificial Intelligence № 3(14)'2019

About

News
Aims and Scope
Founder and Publisher
Editorial board
Licensing terms
Privacy Statement
Plagiarism policy
Publication ethics
Archiving Policy
Subscription


For Authors

Instructions for authors
Reviewing proccess
Copyright Notice
License agreement
Article processing charges


Archive

All issues
Search


Contacts

Contacts


UDC 538.91

Выполнены первопринципные расчеты электронной структуры нанопроводов железа с поперечными размерами до 12.6Å. В поперечном сечении структура нанопроводов с ростом диаметра приближается к структуре объемно-центрированного кристалла. Близость электронной структуры нанопровода и объемного кристалла определяется, характеризующим нанопровод, средним количеством ближайших соседей.
Ключевые слова: теория функционала плотности, электронная структура, нанопровод, уровень Ферми.

V. G. Boutko, A. A. Gusev MODELING GROWTH OF THE DIAMETER IN IRON NANOWIRE. STRUCTURAL AND ELECTRONIC PROPERTIES
The first-principle calculations of the electronic structure of iron nanowires with transverse dimensions up to 12.6Å were performed. In cross section, the structure of nanowires with increasing diameter approaches the structure of a body-centered crystal. The proximity of the electronic structure of the nanowire and the bulk crystal is determined, characterizing the nanowire, by the average number of nearest neighbors.
Key words: density functional theory, electron structure, nanowire, Fermi level.


UDC 51–77

V. I. Levin QUANTITATIVE METHODS FOR DETECTING AND MEASURING CORRUPTION
The problem of mathematical modeling, measurement and detection of corruption is formulated. The
model of corrupted system is constructed. Methods of measurement and detection of corruption are
offered and real practical examples of problem solving are indicated.
Keywords: corruption, mathematical modeling, detection, localization, measuring, organizational control, expertise.


UDC 681.518.9; 621.384.3

S. S. Antsyferov, K. N. Fazilova EVALUATION OF COGNITIVE SYSTEMS STRUCTURAL ELEMENTS EFFECTIVENESS
Cognitive systems have a number of properties, the most important of which is the property of nonequilibrium stability. A method for determining quality indicators is proposed, which includes the development of a model of nonequilibrium stability, a real-time structural element evaluation model, and numerical modeling of the proposed models.
Keywords: cognitive system, structural element, nonequilibrium stability, efficiency indicators of structural element, entropy.


UDC 004.89
T. V. Yermolenko CLASSIFICATION OF ERRORS IN THE TEXT BASED ON DEEP LEARNING
The article considers the task of classifying errors made in Russian-language texts using deep neural networks. To solve this problem, the QRNN architecture was selected using the attention layer; instead of vector representations of words, the penultimate layer from the trained language models was used. Language models were trained using a recursive network with LSTM architecture. Numerical studies have shown the effectiveness of the proposed architecture, the accuracy of the classification of errors in the test sample exceeded 95%.
Key words: natural language processing, recurrent neural networks, quasi-recurrent neural networks.


UDC 608.2

E. P. Miroshnikova, D. K. Levonevskiy, A. I. Motienko MODULES FOR IMPORT, EXPORT AND DATA ANALYTICS IN THE ELECTRONIC JOURNAL MANAGEMENT SYSTEM OF THE “SPIIRAS PROCEEDINGS” JOURNAL FOR AUTOMATED INTERACTION WITH GLOBAL INDICES AND AGGREGATORS
This paper considers current problems that may be encountered by a scientific journal while introducing an electronic journal management system and demonstrates the way of their solution by the “SPIIRAS Proceedings” journal. Also the analysis of existing journal management systems, services and their relevance to the demands of scientific journals in Russia is provided. The article describes the advantages of using a journal management system as well as original modules and functions included in the standard Open Journal Systems package.
Keywords: electronic journal management systems, Open Journal Systems, scientific journals, metadata, global citation indices.


UDC 004.932.2

N. Y. Tarasov, I. A. Tarasova SYSTEM OF EVALUATION OF RESULTS OF SHOOTING TRAINING
The paper describes all the stages of developing a system for evaluating the results of shooting training. A study of image processing and analysis methods has been carried out, on the basis of which the border detection method has been selected for the implementation of the system. Models and an algorithm for the functioning of the system are developed. The software implementation of the system for evaluating the results of shooting training was implemented, the introduction of which will improve the quality of shooting training.
Key words: system, shooting training, analysis, image processing, model.