Software & Systems

Articles of journal № 1 at 2020 year.

Order result by: Public date | Title | Authors

1. Design automation of adaptive user interfaces with elements of artificial intelligence [№1 за 2020 год]
Authors: Zubkova, T.M., Tagirova, L.F., V.K. Tagirov
Visitors: 5720
Design of the user interface is more, than distribution on the screen of controlling units of the program. The cognitive psychology explains how the person reads, thinks and remembers. It helps to design user interfaces in such a way that the software would work according to psychological opportunities and re-strictions of those people for which it is intended. In article describes the development technique of adaptive interfaces of application programs based on the professional quality assessment, psychophysiological features and an emotional status of users. Method using of artificial intelligence at software design allows considering features of the user au-dience. There are the characteristics based on which for users the adaptive interface will be selected. There are three groups of characteristics. The first group is the professional qualities during the work with the computer equipment, the second included psychophysiological features the third emotional statuses of the user. Each of characteristics has a three-point scale: the high, low and average level depending on the got points. The developed automated system, allows creating groups based on the user assessment, the adapted prototype of the application program interface is for each of which. For selection of an optimum proto-type of the interface in the application we use the indistinct expert system (ES) based on the produc-tional model of the knowledge representation. There is the example of the interface selection for the staff work of tax administration implementing data processing of individual and legal entities. There is a receiving two templates of the interface for experienced and not experienced users. As a result, users with different level of training and opportuni-ties can carry out more comfortably the professional activity, during the work with the application pro-gram.

2. Algorithm and software implementation for searching the parameter values deviations from industrial equipment norms [№1 за 2020 год]
Authors: A.E. Kolodenkova , S.S. Vereshchagina
Visitors: 4369
The paper proposes an algorithm for searching the parameters values deviations from industrial equip-ment norms in the context of statistical and fuzzy source data to further equipment diagnosing at the operational stage. The authors proposes generalized search pattern deviations of the industrial facility parameters with database using (technical data sheets, GOST standards and regulatory documentation), as well as the knowledge base containing the staff’s knowledge and experience of the equipment pa-rameters values, factors that may affect the equipment performance presented in the form of verbal de-scriptions and intervals. The authors proposed the algorithm, which searches the mistakes of parameters from the industrial facility norms with using of database. This algorithm in the decision support systems decision support systems for equipment diagnostics makes it possible not only to identify the time when the parameter value deviation occurred, but the duration and the average value of the parameter deviation as well. This paper describes an infological data model (a description of entities) with a set of attributes and re-lationships among them, which are in the process of researching both input and output data, as well as a logical data model (a display of logical relationships among data elements). As an example, there are the screen forms of the user interface of the developed software system in the C # programming lan-guage for searching deviations of parameter values. Using of the developed software will provide the staff on duty with the following: 1) lack of neces-sity to constantly access the technical data sheets, GOST standards and regulatory documentation pre-sented in paper form; 2) rejection of an option to choose possible causes of deviations; 3) elimination of unreasonable equipment repairs.

3. Algorithm of route integration for the development of an electronic model of a territorial waste management scheme [№1 за 2020 год]
Authors: O.I. Khristodulo, A.Kh. Abdullin , G.F. Bagamanshina
Visitors: 3878
The reform of the waste products distribution touches many aspects. The questions of the rational or-ganization of the waste transportation are actual for the all-Russian regions. The one of the directions of the development territorial arrangement is a reduction in expenses for fuel waste delivery to the waste polygons or the places of their utilization, a tariff cut for the citizens. For developed the under-lying territorial scheme of a waste management we need the carrying-out of an analysis and the waste direction control with consideration to the waste formation dynamics, the spatial features of the munic-ipal territories, accumulation rate determination of the waste on the polygons. Specialists solve the problem of the road complexation of the refuse transport for the electronic model development of the waste management scheme. The paper has a look at the number of the most identified models and ways to the material allocation in the spatially-regular system. There is an estimate of its using for modifying of the territorial scheme of a waste management. The specialists recognized that one of such approach-es must not directly use to the waste management. The using the well-known algorithms of the transport logistic disabled to take stock of the combination of the factors, representative for a mature model in a transportation of waste region. There needs their combination and refinement. There is a collection or-ganization description, utilization and waste disposal, and a direction control of a municipal solid waste from the local centers to the polygons. The authors showed the algorithm of the complexation cargo transportation route. There is a script package based on the offered algorithm, with help of this the specialists made the analysis of the mod-eling waste transport flows. The complexation results of the march allow developing the whole and consistent electronic model of the territorial waste management scheme.

4. A hybrid method for automatic summarization of scientific and technical texts based on rhetorical analysis [№1 за 2020 год]
Authors: Batura T.V., A.M. Bakieva
Visitors: 3479
Continuously increasing amounts of textual information makes the problem of semantic text pressure more and more important. Relations between discursive markers in a text set the hierarchy, which al-lows solving the various text-processing problems in natural language, including the problem of the au-tomatic gist abstracting from high-cube text. This article describes the original method of auto reviewing of scientific and technical texts based on rhetorical analysis and using topic modeling. The suggested method combines the usage of a linguis-tic knowledge base, graph representation of texts and machine learning. There is a topic modeling for the detection of essential terms. First, there is a producing of unigram topic models. They contain only one-word terms. Further, these is an expansion of these models by adding multiword terms. The determination of the most important fragments of the source document are in the process of rhetorical analysis with the help of discursive markers. Representation of texts in the form of graphs helps to demonstrate the transformation with the text necessary to extract important fragments. In assessing the importance of text fragments, keywords, multiword terms, and scientific lexicon characterizing scientific and technical texts are also conceded. The creating of the linguistic knowledge base was for the data storage markers. The final step of developing abstract is smoothing – a procedure that makes the text of the summary more coherent and consistent. As the experiments showed, the proposed method is effective and can be adapted for processing texts of different subjects in other languages.

5. Integrated development environment with structured editing support for Go programming language [№1 за 2020 год]
Authors: N.V. Vanyasin, I.G. Sidorkina
Visitors: 5229
The paper describes software implementation of an integrated development environment with struc-tured editing support that allows creating and editing software projects using Go programming lan-guage. Structure editors allow editing a source code as a hierarchy of embedded blocks in opposite to text editors, which allow editing a source as a list of characters and lines. The proposed software implementation is different from the known systems by the storage repre-sentation method that allows speeding up the performing operations with a source code. The proposed editor also allows increasing the overall efficiency of a software development process. This implemen-tation solves the problem of circular translation of a source code into an intermediate representation and vice versa, which is common among modern structured editors of a source code. The solution for this problem improves the development environment performance and program-mer’s productivity in several ways. The study lists the basic subsystems and components of the pro-posed information system, their purpose, functions and work principle. The paper describes the user interface and its basic features. One of the main features of the pro-posed implementation is a version control system that allows tracking the differences between source code hierar- chical structures, not the differences between text files. It allows a software developer to get more de-tailed and unambiguous information when browsing a version history as well when performing merge or conflict resolution. The paper presents test results of the integrated development environment performance and the ef-fectiveness of software development using this environment. It also describes the implementation re-sults of the proposed system in the development process of industrial companies and educational organization. Finally, there are some insights on further development of the study.

6. Intelligent analysis video data to recognize car theft situations in a parking lot [№1 за 2020 год]
Authors: A.Yu. Kruchinin, R.R. Galimov
Visitors: 2608
Modern video surveillance systems are very popular to prevent and investigate various illegal inci-dents. The main disadvantage of traditional decisions in this area is that the operator directly assesses the current situation and makes a decision. With a large number of monitored cameras and due to the human factor, there is a high probability of increasing the time delay in determining the dangerous situ-ation. This leads to significant damage, especially in cases where it is necessary to respond promptly to an incident. Such cases include, for example, the situation of carjacking. These circumstances necessi-tate the introduction of systems for the intellectual analysis of video data. This paper submits a method for recognizing situations of car theft from parking based on stochastic grammars and deep neural networks. Recognition of an incident is at two main levels: at the lower lev-el, event recognition occurs, at the upper level, the situation as the most probable chain of events cor-responding to the grammar of car theft signatures. The analysis of the identified objects, their relative position, the dynamic characteristics of the trajectory of movement and the characteristics of the peo-ple’s postures, fills up the detection of possible hijacking events. The deep neural networks recognize the objects and people poses. These networks have a high degree of reliability. The article developed a simulation model of the system of recognition of the situations of theft of road transport, which based on the object detection module using a deep neural network. The history of certain objects in previous frames and, if necessary, data on the posture of a person improves the as-sessment of the reliability of event recognition. To describe the possible scenarios of car theft, the authors developed a stochastic grammar and cre-ated the test utility based on it. The test results of the developed method on the Mini-deone video da-taset showed its efficiency.

7. Using self-learning optimization algorithms to manage dynamically changing systems [№1 за 2020 год]
Author: V.A. Kostenko
Visitors: 4334
The paper proposes an approach to the management of dynamically changing systems. Such systems has a response that during the system operation their state may: the list of the equipment, the load of the system and the functions performed by the system. It is required to choose the values of control pa-rameters depending on the system state in such a way as to provide the required values of the charac-teristics of the system. To solve this problem, the authors plan to use optimization algorithms with self-learning: direction-al random search algorithms with self-learning and ant algorithms. These algorithms, operating by trial and error method, allow you to tune in to the current state of the system. The introducing into the algo-rithms the memory of successes and failures on the previous steps helps to achieve it. The main idea of using self-learning algorithms to control the system is that we will consider the control problem as an unconditional optimization problem. The elements of the vector of optimized variables are the control parameters of the system. The authors consider the algorithm steps as possible actions of system control (operations of changing control parameter values). The objective function can be set in one of two ways: by the calculation of the total mean probable characteristics error of the system from the required ones. The weights help take into account the characteristics importance by summing deviations; the researching of the maximum characteristics deviation of the system operation from the required ones. The proposed management approach allows the unstable environmental behavior, limited managed system information and allows you to take into account the presence of many system characteristics, their values must be maintained within specified limits. If the system elements failure does not lead to the whole system failure, but only worsens the system characteristics values, this approach will mini-mize the characteristics values deviations from the required. A limitation of this approach is the transition process allow ability by changing the system state. In this way, the algorithm will require a number of steps for retraining. Some steps cause disturbance the required system performance.

8. An anti-spoofing methodology for a limited number of photos [№1 за 2020 год]
Authors: K.D. Rusakov , A.A. Genov, S.Sh. Hill
Visitors: 6144
Nowadays, the problem of providing a high security level of mobile devices, such as smartphones and tablets, through biometric approaches is particularly relevant. The paper proposes an anti-spoofing method for a limited number of images. The authors investigate the detection of spoof attacks using printed photos and screens of mobile devices and monitors. They show the relevance of the research and current unresolved problems. The paper considers the structure of a typical Liveness system consisting of a source (camera), pre-processing of received images, face detection, feature module and a classifier. The study shows that pre-processing of the obtained images is one of the most important parts of the system, since the signs of spoof attacks are difficult to detect due to the extensiveness of augmentations. There is a small review of modern architectures of convolutional neural networks (called encoders in the current architecture terminology). It is also shown that the linear output of convolutional neural networks can be used as an input for LSTM-type recurrent neural networks. It is noted that the best face detection algorithm in the current architecture is MMOD method. The resulting Liveness system architecture is presented as a combined approach consisting of two components: a convolutional neural network receiving embeddings from each frame and a LSTM re-current neural network that uses these insertions on the input and learns to remember sequences of cer-tain frames and their characteristics. The authors present experimental results showing the accuracy indicators of the current develop-ments in this field when computing power requirements are low. This methodology allows determining fake photos by several frames.

9. Modeling the simultaneous liquid and gas flow: a practical approach [№1 за 2020 год]
Authors: A.P. Koryakov , I.B. Kostyukov , M.N. Rybakov
Visitors: 2450
The paper deals with the problem of modeling the liquid and gas flow in systems consisting of contain-ers of various volumes interconnected by pipes of various lengths and widths. The proposed model is based on a system of differential equations describing the change in sub-stance mass in containers as well as the change in the substance flow magnitude in pipes. In addition to container and pipe parameters, the model takes into account various parameters of liquids and gases, such as density, temperature, and molar mass. The description is accompanied by formulas with neces-sary explanations. The paper also discusses the problems caused by high computational costs to run such models, time costs. It proposes a number of heuristics to reduce those costs significantly. The authors discuss each idea, interconnections between applications of the ideas, as well as their limitations. The authors describe the key technical aspects of model implementation, including the proposed heuristics and the ability to configure various parameters that affect the accuracy of the calculation and its running time. The paper shows that the proposed model can be extended by adding the elements that model the operation of various devices, such as valves, gearboxes, and pumps. Each device has an implementa-tion model. Finally, the scope of the model and a computer implementation are discussed. The authors also mention some issues beyond the paper scope, such as implementation of parallel computing.

10. Module for morphological disambiguation: architecture and database organization [№1 за 2020 год]
Authors: D.R. Mukhamedshin, D.Sh. Sulejmanov
Visitors: 6594
Corpus data management systems today help to solve a number of urgent problems connected with the computer linguistics. Particularly, this is the source documents processing, automatic morphological markup of them, storage and modification of corpus data with morphological markup, the search in-quiry execution (selecting data), and integration with other applications. In the automatic morphological text markup of the electronic corpus, there is a problem when two or more sets of morphological properties marked out the same word form. It is a morphological ambiguity. The authors developed a module for manual resolution of morphological ambiguity, which can help in solving this problem. This article discusses the architecture of the module for resolving the morpholog-ical ambiguity of the corpus data management system developed for managing the Tatar corpus, and describes the organization of the database based on the transactional approach to data storage. The article proposes in detail the architecture of the rupture modulus of the software morphological ambiguousness of corpus data. It is for the tartaric corpus management. In addition, the author de-scribed the database organization. It relies on the transactional way to the strategy for the data storage. The paper in detail describes the architecture of the module software, which is part of the Tugan Tel corpus data management system. Due to the architecture of the corpus data management system, such modules can quickly integrate into the system and have no effect on the other module operation. The module database organization for the morphological disambiguation described in the paper, al-low going over for solving the other computer linguistic problems, such as automatic morphological disambiguation and improving automatic morphological markup of texts in the electronic corpus.

| 1 | 2 | 3 | Next → ►