Pub Date : 1900-01-01DOI: 10.18127/j19997493-202102-05
I.S. Markov, N.V. Pivovarova
Formulation of the problem. The problem of using large neural networks with complex architectures on modern devices is considered. They work well, but sometimes their speed is unacceptably low and the amount of memory required to place them on the device is not always available. Briefly describes how to solve these problems by using pruning and quantization. It is proposed to consider an unconventional type of neural networks that can meet the requirements for the occupied memory space, speed and quality of work and describe approaches to training this type of networks. The aim of the work is to describe modern approaches to reducing the size of neural networks with minimal loss of the quality of their work and to propose an alternative type of networks of small size and high accuracy. Results. The proposed type of neural network has a large number of advantages in terms of the size and flexibility of layer settings. By varying the parameters of the layers, you can control the size, speed and quality of the network. However, the greater accuracy, the greater the memory volume. To train such a small network, it is proposed to use specific techniques that allow learning complex dependencies based on a more complex and voluminous network. As a result of this learning procedure, it is assumed that only a small network is used, which can then be placed on low-power devices with a small amount of memory. Practical significance. The described methods allow the use techniques to reduce the size of networks with minimal loss of quality of their work. The proposed architecture makes it possible to train simpler networks without using their size reduction techniques. These networks can work with various data, be it pictures, text, or other information encoded in a numerical vector.
{"title":"Methods for constructing effective neural networks for low-power systems","authors":"I.S. Markov, N.V. Pivovarova","doi":"10.18127/j19997493-202102-05","DOIUrl":"https://doi.org/10.18127/j19997493-202102-05","url":null,"abstract":"Formulation of the problem. The problem of using large neural networks with complex architectures on modern devices is considered. They work well, but sometimes their speed is unacceptably low and the amount of memory required to place them on the device is not always available. Briefly describes how to solve these problems by using pruning and quantization. It is proposed to consider an unconventional type of neural networks that can meet the requirements for the occupied memory space, speed and quality of work and describe approaches to training this type of networks. The aim of the work is to describe modern approaches to reducing the size of neural networks with minimal loss of the quality of their work and to propose an alternative type of networks of small size and high accuracy. Results. The proposed type of neural network has a large number of advantages in terms of the size and flexibility of layer settings. By varying the parameters of the layers, you can control the size, speed and quality of the network. However, the greater accuracy, the greater the memory volume. To train such a small network, it is proposed to use specific techniques that allow learning complex dependencies based on a more complex and voluminous network. As a result of this learning procedure, it is assumed that only a small network is used, which can then be placed on low-power devices with a small amount of memory. Practical significance. The described methods allow the use techniques to reduce the size of networks with minimal loss of quality of their work. The proposed architecture makes it possible to train simpler networks without using their size reduction techniques. These networks can work with various data, be it pictures, text, or other information encoded in a numerical vector.","PeriodicalId":370962,"journal":{"name":"Dynamics of Complex Systems - XXI century","volume":"100 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129623810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.18127/j19997493-202102-01
D. Berezkin, Shi Runfang, Liu Tengjiao
This experiment compared the performance of four machine learning algorithms in detecting bank card fraud. At the same time, the strong imbalance of the classes in the training sample was taken into account, as well as the difference in transaction amounts, and the ability of different machine learning methods to recognize fraudulent behavior was assessed taking into account these features. It has been found that a method that works well with indicators for assessing a classification is not necessarily the best in terms of assessing the magnitude of economic losses. Logistic regression is a good proof of this. The results of this work show that the problem of detecting fraud with bank cards cannot be regarded as a simple classification problem. AUC data is not the most appropriate metric for fraud detection tasks. The final choice of the model depends on the needs of the bank, that is, it is necessary to take into account which of the two types of errors (FN, FP) will lead to large economic losses for the bank. If the bank believes that the loss caused by identifying fraudulent transactions as regular transactions is the main one, it should choose the algorithm with the lowest FN value, which in this experiment is Adaboost. If the bank believes that the negative impact of identifying regular transactions as fraudulent is also very important, it should choose an algorithm with relatively small FN and FP data. In this experiment, the overall performance of the random forest is better. Further, by evaluating the economic losses caused by false positives (identifying an ordinary transaction as fraudulent), a quantitative analysis of the economic losses caused by each algorithm can be used to select the optimal algorithm model.
{"title":"Applying and comparing multiple machine learning techniques to detect fraudulent credit card transactions","authors":"D. Berezkin, Shi Runfang, Liu Tengjiao","doi":"10.18127/j19997493-202102-01","DOIUrl":"https://doi.org/10.18127/j19997493-202102-01","url":null,"abstract":"This experiment compared the performance of four machine learning algorithms in detecting bank card fraud. At the same time, the strong imbalance of the classes in the training sample was taken into account, as well as the difference in transaction amounts, and the ability of different machine learning methods to recognize fraudulent behavior was assessed taking into account these features. It has been found that a method that works well with indicators for assessing a classification is not necessarily the best in terms of assessing the magnitude of economic losses. Logistic regression is a good proof of this. The results of this work show that the problem of detecting fraud with bank cards cannot be regarded as a simple classification problem. AUC data is not the most appropriate metric for fraud detection tasks. The final choice of the model depends on the needs of the bank, that is, it is necessary to take into account which of the two types of errors (FN, FP) will lead to large economic losses for the bank. If the bank believes that the loss caused by identifying fraudulent transactions as regular transactions is the main one, it should choose the algorithm with the lowest FN value, which in this experiment is Adaboost. If the bank believes that the negative impact of identifying regular transactions as fraudulent is also very important, it should choose an algorithm with relatively small FN and FP data. In this experiment, the overall performance of the random forest is better. Further, by evaluating the economic losses caused by false positives (identifying an ordinary transaction as fraudulent), a quantitative analysis of the economic losses caused by each algorithm can be used to select the optimal algorithm model.","PeriodicalId":370962,"journal":{"name":"Dynamics of Complex Systems - XXI century","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129388861","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.18127/j19997493-202101-06
M. Basarab, B. Lunin, E. Chumankin
Wave solid-state gyroscopes (WSG) are among the most modern navigation devices. Based on the phenomenon of precession of elastic waves in thin-walled axisymmetric bodies, WSGs have a simple design, including 2-3 fixed parts, and have a number of advantages over other types of gyroscopes: great resource of work; small random error; resistance to severe operating conditions (overload, vibration, gamma radiation); relatively small overall dimensions, weight and power consumption; preservation of inertial information during short-term power outages. From the point of view of practical application and technologies used, three main groups of WSG can be distinguished. Wave solid-state gyroscopes of high precision. In such devices, high-quality (with a Q-factor of over 1·107) quartz resonators, contactless sensors and actuators, as well as complex electronic control systems are used. The field of application today, for various reasons, is limited to space technology, which requires, along with high precision, a long working life. Micromechanical devices of low accuracy for mass use (laptop computers, toys, industrial equipment, etc.) Integration of micromechanical WSGs with satellite systems makes it possible to create small-sized inexpensive navigation systems for widespread use. This market segment is developing very quickly, but production of such devices requires a very high the level of development of the microelectronic industry. An intermediate group consists of sensors of general use with metal resonators. Although these devices are larger than micromechanical devices, their production technology is much simpler. Metal resonators with a quality factor of (3 ... 5)∙104 can be manufactured using universal metal-cutting equipment; such devices have a simple design, do not require the creation of a high vacuum in their housing, and widespread radioelements can be used in control units. As a result, devices of this group, possessing insignificant power consumption and long working life, have a low cost price. On the other hand, the comparatively large dimensions of the resonator allow their precise tuning, which makes it possible to sharply increase the accuracy of the gyro instruments. From these points of view, a general-purpose WSG with a metal resonator is the most promising device that should replace the rotary-type electromechanical gyroscopes used today, and the production of which can be quickly mastered by the domestic industry. The development of such sensors requires solving a number of scientific and technical problems. Since all the main characteristics of such a device are determined by the properties of the resonator, special attention should be paid to its design and production technology. One of the most difficult and expensive operations in the WSG technology is the balancing of the resonator, carried out to eliminate the mass imbalance that arises during its manufacture due to inevitable deviations from the ideal axisymmetric shape
{"title":"Balancing of metal resonators of wave solid-state gyroscopes of general use","authors":"M. Basarab, B. Lunin, E. Chumankin","doi":"10.18127/j19997493-202101-06","DOIUrl":"https://doi.org/10.18127/j19997493-202101-06","url":null,"abstract":"Wave solid-state gyroscopes (WSG) are among the most modern navigation devices. Based on the phenomenon of precession of elastic waves in thin-walled axisymmetric bodies, WSGs have a simple design, including 2-3 fixed parts, and have a number of advantages over other types of gyroscopes: great resource of work; small random error; resistance to severe operating conditions (overload, vibration, gamma radiation); relatively small overall dimensions, weight and power consumption; preservation of inertial information during short-term power outages. From the point of view of practical application and technologies used, three main groups of WSG can be distinguished. Wave solid-state gyroscopes of high precision. In such devices, high-quality (with a Q-factor of over 1·107) quartz resonators, contactless sensors and actuators, as well as complex electronic control systems are used. The field of application today, for various reasons, is limited to space technology, which requires, along with high precision, a long working life. Micromechanical devices of low accuracy for mass use (laptop computers, toys, industrial equipment, etc.) Integration of micromechanical WSGs with satellite systems makes it possible to create small-sized inexpensive navigation systems for widespread use. This market segment is developing very quickly, but production of such devices requires a very high the level of development of the microelectronic industry. An intermediate group consists of sensors of general use with metal resonators. Although these devices are larger than micromechanical devices, their production technology is much simpler. Metal resonators with a quality factor of (3 ... 5)∙104 can be manufactured using universal metal-cutting equipment; such devices have a simple design, do not require the creation of a high vacuum in their housing, and widespread radioelements can be used in control units. As a result, devices of this group, possessing insignificant power consumption and long working life, have a low cost price. On the other hand, the comparatively large dimensions of the resonator allow their precise tuning, which makes it possible to sharply increase the accuracy of the gyro instruments. From these points of view, a general-purpose WSG with a metal resonator is the most promising device that should replace the rotary-type electromechanical gyroscopes used today, and the production of which can be quickly mastered by the domestic industry. The development of such sensors requires solving a number of scientific and technical problems. Since all the main characteristics of such a device are determined by the properties of the resonator, special attention should be paid to its design and production technology. One of the most difficult and expensive operations in the WSG technology is the balancing of the resonator, carried out to eliminate the mass imbalance that arises during its manufacture due to inevitable deviations from the ideal axisymmetric shape ","PeriodicalId":370962,"journal":{"name":"Dynamics of Complex Systems - XXI century","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115252171","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.18127/j19997493-202102-04
E.Yu. Silantieva, V.A. Zabelina, G.A. Savchenko, I.M. Chernenky
This study presents an analysis of autoencoder models for the problems of detecting anomalies in network traffic. Results of the training were assessed using open source software on the UNB ICS IDS 2017 dataset. As deep learning models, we considered standard and variational autoencoder, Deep SSAD approaches for a normal autoencoder (AE-SAD) and a variational autoencoder (VAE-SAD). The constructed deep learning models demonstrated different indicators of anomaly detection accuracy; the best result in terms of the AUC metric of 98% was achieved with VAE-SAD model. In the future, it is planned to continue the analysis of the characteristics of neural network models in cybersecurity problems. One of directions is to study the influence of structure of network traffic on the performance indicators of using deep learning models. Based on the results, it is planned to develop an approach of robust identification of security events based on deep learning methods.
{"title":"Detecting internet attacks using a neural network","authors":"E.Yu. Silantieva, V.A. Zabelina, G.A. Savchenko, I.M. Chernenky","doi":"10.18127/j19997493-202102-04","DOIUrl":"https://doi.org/10.18127/j19997493-202102-04","url":null,"abstract":"This study presents an analysis of autoencoder models for the problems of detecting anomalies in network traffic. Results of the training were assessed using open source software on the UNB ICS IDS 2017 dataset. As deep learning models, we considered standard and variational autoencoder, Deep SSAD approaches for a normal autoencoder (AE-SAD) and a variational autoencoder (VAE-SAD). The constructed deep learning models demonstrated different indicators of anomaly detection accuracy; the best result in terms of the AUC metric of 98% was achieved with VAE-SAD model. In the future, it is planned to continue the analysis of the characteristics of neural network models in cybersecurity problems. One of directions is to study the influence of structure of network traffic on the performance indicators of using deep learning models. Based on the results, it is planned to develop an approach of robust identification of security events based on deep learning methods.","PeriodicalId":370962,"journal":{"name":"Dynamics of Complex Systems - XXI century","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131262801","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.18127/j19997493-202101-02
O. Rogova, V. Stroganov, D. Stroganov
The article deals with the analysis of the behavior of controlled simulation models for solving the choice of extreme values of the functional, which it determines on the basis of the average integral estimate. It is assumed that the search engine optimization algorithm is directly included in the model. Of interest is the problem of estimating the duration of the control interval, i.e. system simulation time with different parameters to select the search direction. The smaller the control interval, the lower the accuracy of the estimates of the functional and, accordingly, the lower the probability of choosing the correct search direction. However, with a general limitation on the simulation time, the search algorithm performs a larger number of steps, which increases the rate of convergence to the extreme value. Thus, the choice of the duration of the control interval raises a question. The aim of the work is to build a model of a controlled process, i.e. the process of changing the controlled parameters, to estimate the rate of convergence of the optimization algorithm depending on the duration of the control interval. The analysis of the convergence of the optimization process directly on the simulation model is practically impossible due to the nonstationary nature of all ongoing processes. In this regard, the article introduces a class of conditionally non-stationary Gaussian processes, on which the efficiency of a controlled simulation model is evaluated. It is assumed that a symmetric design is used to choose the direction, and all realizations of the nonstationary process at the current point have the same initial state. As a result of the analysis of such a model, analytical expressions were obtained for estimating the accuracy of the position of the extremum depending on the duration of the control interval. The results obtained make it possible, with a general limitation of the time for conducting experiments with a simulation model, to construct a sequential analysis plan, which improves the accuracy of solving the optimization problem.
{"title":"Analysis of the behavior of search engine optimization algorithms on transient modes of non-stationary simulation processes","authors":"O. Rogova, V. Stroganov, D. Stroganov","doi":"10.18127/j19997493-202101-02","DOIUrl":"https://doi.org/10.18127/j19997493-202101-02","url":null,"abstract":"The article deals with the analysis of the behavior of controlled simulation models for solving the choice of extreme values of the functional, which it determines on the basis of the average integral estimate. It is assumed that the search engine optimization algorithm is directly included in the model. Of interest is the problem of estimating the duration of the control interval, i.e. system simulation time with different parameters to select the search direction. The smaller the control interval, the lower the accuracy of the estimates of the functional and, accordingly, the lower the probability of choosing the correct search direction. However, with a general limitation on the simulation time, the search algorithm performs a larger number of steps, which increases the rate of convergence to the extreme value. Thus, the choice of the duration of the control interval raises a question. The aim of the work is to build a model of a controlled process, i.e. the process of changing the controlled parameters, to estimate the rate of convergence of the optimization algorithm depending on the duration of the control interval. The analysis of the convergence of the optimization process directly on the simulation model is practically impossible due to the nonstationary nature of all ongoing processes. In this regard, the article introduces a class of conditionally non-stationary Gaussian processes, on which the efficiency of a controlled simulation model is evaluated. It is assumed that a symmetric design is used to choose the direction, and all realizations of the nonstationary process at the current point have the same initial state. As a result of the analysis of such a model, analytical expressions were obtained for estimating the accuracy of the position of the extremum depending on the duration of the control interval. The results obtained make it possible, with a general limitation of the time for conducting experiments with a simulation model, to construct a sequential analysis plan, which improves the accuracy of solving the optimization problem.","PeriodicalId":370962,"journal":{"name":"Dynamics of Complex Systems - XXI century","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125896805","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.18127/j19997493-202103-02
D.N. Bogomolov, B. Goryachkin
The article is devoted to the problem of the efficiency of the support service in the structure of a modern office and ways to improve it by connecting intelligent modules. The paper describes the Service Desk system and details the algorithm of its operation. The bottlenecks of the basic algorithm are noted and ways of its improvement are suggested. Attention is drawn to how machine learning methods can improve the efficiency of the system as a whole.
{"title":"Improving the efficiency of a modern office through an intelligent Service Desk user support system","authors":"D.N. Bogomolov, B. Goryachkin","doi":"10.18127/j19997493-202103-02","DOIUrl":"https://doi.org/10.18127/j19997493-202103-02","url":null,"abstract":"The article is devoted to the problem of the efficiency of the support service in the structure of a modern office and ways to improve it by connecting intelligent modules. The paper describes the Service Desk system and details the algorithm of its operation. The bottlenecks of the basic algorithm are noted and ways of its improvement are suggested. Attention is drawn to how machine learning methods can improve the efficiency of the system as a whole.","PeriodicalId":370962,"journal":{"name":"Dynamics of Complex Systems - XXI century","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127155689","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.18127/j19997493-202101-01
D. Stroganov, V. Chernenky
Thus, the formal formulation of the problem of evaluating the effectiveness of search optimization procedures on simulation models of regenerating processes under strict time constraints has been completed. A procedure for parametric tuning of the optimization algorithm has been developed, which sequentially refines the values of the functional specified by the model and redistributes the remaining model regeneration cycles between the investigated values of the controlled parameter. The problem of maximizing the probability of the correct choice is posed and solved, i.e. selection, according to the results of the simulation experiment on the model of the regenerating process, of the value of the controlled parameter that delivers the true maximum to the investigated functional. Based on the transition to the Lagrangian, the solution to the constrained optimization problem is reduced to an unconstrained optimization problem. Analytical expressions are obtained to assess the optimal distribution of regeneration cycles. It is shown that the simulation model with the included search engine optimization algorithm provides solutions that are quite effective in terms of computational costs. As a result, a method is proposed for a simple extension of the developed simulation models by including a search optimization algorithm, which makes it possible to move from modeling the system to optimizing its objective function on a given area of controlled parameters.
{"title":"Evaluation of the efficiency of the problem of choosing the extreme values of the parameters of the parametric family of regenerating processes","authors":"D. Stroganov, V. Chernenky","doi":"10.18127/j19997493-202101-01","DOIUrl":"https://doi.org/10.18127/j19997493-202101-01","url":null,"abstract":"Thus, the formal formulation of the problem of evaluating the effectiveness of search optimization procedures on simulation models of regenerating processes under strict time constraints has been completed. A procedure for parametric tuning of the optimization algorithm has been developed, which sequentially refines the values of the functional specified by the model and redistributes the remaining model regeneration cycles between the investigated values of the controlled parameter. The problem of maximizing the probability of the correct choice is posed and solved, i.e. selection, according to the results of the simulation experiment on the model of the regenerating process, of the value of the controlled parameter that delivers the true maximum to the investigated functional. Based on the transition to the Lagrangian, the solution to the constrained optimization problem is reduced to an unconstrained optimization problem. Analytical expressions are obtained to assess the optimal distribution of regeneration cycles. It is shown that the simulation model with the included search engine optimization algorithm provides solutions that are quite effective in terms of computational costs. As a result, a method is proposed for a simple extension of the developed simulation models by including a search optimization algorithm, which makes it possible to move from modeling the system to optimizing its objective function on a given area of controlled parameters.","PeriodicalId":370962,"journal":{"name":"Dynamics of Complex Systems - XXI century","volume":"104 5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126222359","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.18127/j19997493-202101-05
B. Goryachkin, M.D. Kaurov
Currently, scientists identify such a basic human need as information hunger. This means that a person strives to obtain a certain amount of information every day. However, it should not just be the volume of any information, it also has various requirements, such as: volume, reliability, saturation, openness, value and timeliness. In order to ensure the need for rapid access to information, the trend towards the development of digital technologies is becoming increasingly evident in the modern world. The development of information access technologies has led to the creation of many ways to provide a person with the information they requested quickly and efficiently, but the most modern means include virtual reality (VR) and augmented reality (AR) technologies. While virtual reality technologies provide a potentially larger amount of information, augmented reality technologies are more in demand in everyday life, because, unlike VR technologies, they allow the user to observe the real world with the addition of the information he needs. When developing this technology, the goal has always been to most effectively recognize objects in the information field and bring them into the field of view of the operator. The hardware part of the implementation of this technology must have a camera and a monitor. The most common and affordable digital device is a smartphone that fully meets the requirements of this technology. The article discusses the current state of AR technology, primarily the means of developing AR applications for Android. To write any applications, in particular, AR applications for Android, you must use a set of development tools (Softaware Development Kit), otherwise the SDK.
{"title":"Affordable augmented reality for smartphones","authors":"B. Goryachkin, M.D. Kaurov","doi":"10.18127/j19997493-202101-05","DOIUrl":"https://doi.org/10.18127/j19997493-202101-05","url":null,"abstract":"Currently, scientists identify such a basic human need as information hunger. This means that a person strives to obtain a certain amount of information every day. However, it should not just be the volume of any information, it also has various requirements, such as: volume, reliability, saturation, openness, value and timeliness. In order to ensure the need for rapid access to information, the trend towards the development of digital technologies is becoming increasingly evident in the modern world. The development of information access technologies has led to the creation of many ways to provide a person with the information they requested quickly and efficiently, but the most modern means include virtual reality (VR) and augmented reality (AR) technologies. While virtual reality technologies provide a potentially larger amount of information, augmented reality technologies are more in demand in everyday life, because, unlike VR technologies, they allow the user to observe the real world with the addition of the information he needs. When developing this technology, the goal has always been to most effectively recognize objects in the information field and bring them into the field of view of the operator. The hardware part of the implementation of this technology must have a camera and a monitor. The most common and affordable digital device is a smartphone that fully meets the requirements of this technology. The article discusses the current state of AR technology, primarily the means of developing AR applications for Android. To write any applications, in particular, AR applications for Android, you must use a set of development tools (Softaware Development Kit), otherwise the SDK.","PeriodicalId":370962,"journal":{"name":"Dynamics of Complex Systems - XXI century","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129581839","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.18127/j19997493-202101-03
M. Kiwan, D. Berezkin, M. Raad, B. Rasheed
Statement of a problem. One of the main tasks today is to prevent accidents in complex systems, which requires determining their cause. In this regard, several theories and models of the causality of accidents are being developed. Traditional approaches to accident modeling are not sufficient for the analysis of accidents occurring in complex environments such as socio-technical systems, since an accident is not the result of individual component failure or human error. Therefore, we need more systematic methods for the investigation and modeling of accidents. Purpose. Conduct a comparative analysis of accident models in complex systems, identify the strengths and weaknesses of each of these models, and study the feasibility of their use in risk management in socio-technical systems. The paper analyzes the main approaches of accident modeling and their limitations in determining the cause-and-effect relationships and dynamics of modern complex systems. the methodologies to safety and accident models in sociotechnical systems based on systems theory are discussed. The complexity of sociotechnical systems requires new methodologies for modeling the development of emergency management. At the same time, it is necessary to take into account the socio-technical system as a whole and to focus on the simultaneous consideration of the social and technical aspects of the systems. When modeling accidents, it is necessary to take into account the social structures and processes of social interaction, the cultural environment, individual characteristics of a person, such as their abilities and motivation, as well as the engineering design and technical aspects of systems. Practical importance. Based on analyzing various techniques for modeling accidents, as well as studying the examples used in modeling several previous accidents and review the results of this modeling, it is concluded that it is necessary to improve the modeling techniques. The result was the appearance of hybrid models of risk management in socio-technical systems, which we will consider in detail in our next work.
{"title":"Analyzing common approaches of accident models for risk management in socio-technical systems","authors":"M. Kiwan, D. Berezkin, M. Raad, B. Rasheed","doi":"10.18127/j19997493-202101-03","DOIUrl":"https://doi.org/10.18127/j19997493-202101-03","url":null,"abstract":"Statement of a problem. One of the main tasks today is to prevent accidents in complex systems, which requires determining their cause. In this regard, several theories and models of the causality of accidents are being developed. Traditional approaches to accident modeling are not sufficient for the analysis of accidents occurring in complex environments such as socio-technical systems, since an accident is not the result of individual component failure or human error. Therefore, we need more systematic methods for the investigation and modeling of accidents. Purpose. Conduct a comparative analysis of accident models in complex systems, identify the strengths and weaknesses of each of these models, and study the feasibility of their use in risk management in socio-technical systems. The paper analyzes the main approaches of accident modeling and their limitations in determining the cause-and-effect relationships and dynamics of modern complex systems. the methodologies to safety and accident models in sociotechnical systems based on systems theory are discussed. The complexity of sociotechnical systems requires new methodologies for modeling the development of emergency management. At the same time, it is necessary to take into account the socio-technical system as a whole and to focus on the simultaneous consideration of the social and technical aspects of the systems. When modeling accidents, it is necessary to take into account the social structures and processes of social interaction, the cultural environment, individual characteristics of a person, such as their abilities and motivation, as well as the engineering design and technical aspects of systems. Practical importance. Based on analyzing various techniques for modeling accidents, as well as studying the examples used in modeling several previous accidents and review the results of this modeling, it is concluded that it is necessary to improve the modeling techniques. The result was the appearance of hybrid models of risk management in socio-technical systems, which we will consider in detail in our next work.","PeriodicalId":370962,"journal":{"name":"Dynamics of Complex Systems - XXI century","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123885604","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.18127/j19997493-202102-03
M. V. Vinogradova, A. S. Larionov, V. Chernenky
Currently, most modern manufacturing companies use automated information systems to account for resources and plan their activities. The functionality of these systems often becomes very extensive and actively changes over time. It is required to maintain the competence of employees when working with them, for which enterprise knowledge bases are created, which, however, themselves require significant labor costs to build. This paper describes methods and tools for collecting and analyzing information about the actions of users in an automated enterprise management system, which allows you to identify problems in its business processes and the most frequent mistakes of employees. Under the basic information system of the enterprise we mean the software complexes for resource management and control of production activities. One of the most common samples are products based on 1C: Enterprise platform. These systems include advanced functionality of the accounting of user actions, where the type and time of events is recorded, as well as instances of related objects. Common intelligent systems for automating the filling of knowledge bases concentrate on the generation of materials directly from the data structure of the main information system, or analysis of already existing documents in natural language. The approach we propose is not aimed at completely replacing a human expert, but at providing him with up-to-date information about the needs of users in the order of compiling knowledge base materials. The information system of an enterprise is represented by a discrete dynamic system, whose elements change state as a result of user actions. Users' interactions with the information system are recorded as time-ordered sequences of elementary events, which are stored in a special logbook. The subject area structure and business processes of the core information system are described in a knowledge base in the form of a graph structure. Its elements include human-readable materials and a set of metadata, allowing their analysis by algorithmic methods. Based on the results of the work, an approach to identify the most relevant materials for the composition of the knowledge base materials is developed. It is based on algorithmic analysis of the actions of users of the basic information system and the calculation of the amount of time that users spend to eliminate emerging problems. To assess the effectiveness of the proposed approach developed a simulation model and conducted experiments to estimate the downtime of users, depending on the method of filling the knowledge base.
{"title":"Intelligent system for collecting and analyzing information about the actions of users of an automated enterprise management system","authors":"M. V. Vinogradova, A. S. Larionov, V. Chernenky","doi":"10.18127/j19997493-202102-03","DOIUrl":"https://doi.org/10.18127/j19997493-202102-03","url":null,"abstract":"Currently, most modern manufacturing companies use automated information systems to account for resources and plan their activities. The functionality of these systems often becomes very extensive and actively changes over time. It is required to maintain the competence of employees when working with them, for which enterprise knowledge bases are created, which, however, themselves require significant labor costs to build. This paper describes methods and tools for collecting and analyzing information about the actions of users in an automated enterprise management system, which allows you to identify problems in its business processes and the most frequent mistakes of employees. Under the basic information system of the enterprise we mean the software complexes for resource management and control of production activities. One of the most common samples are products based on 1C: Enterprise platform. These systems include advanced functionality of the accounting of user actions, where the type and time of events is recorded, as well as instances of related objects. Common intelligent systems for automating the filling of knowledge bases concentrate on the generation of materials directly from the data structure of the main information system, or analysis of already existing documents in natural language. The approach we propose is not aimed at completely replacing a human expert, but at providing him with up-to-date information about the needs of users in the order of compiling knowledge base materials. The information system of an enterprise is represented by a discrete dynamic system, whose elements change state as a result of user actions. Users' interactions with the information system are recorded as time-ordered sequences of elementary events, which are stored in a special logbook. The subject area structure and business processes of the core information system are described in a knowledge base in the form of a graph structure. Its elements include human-readable materials and a set of metadata, allowing their analysis by algorithmic methods. Based on the results of the work, an approach to identify the most relevant materials for the composition of the knowledge base materials is developed. It is based on algorithmic analysis of the actions of users of the basic information system and the calculation of the amount of time that users spend to eliminate emerging problems. To assess the effectiveness of the proposed approach developed a simulation model and conducted experiments to estimate the downtime of users, depending on the method of filling the knowledge base.","PeriodicalId":370962,"journal":{"name":"Dynamics of Complex Systems - XXI century","volume":"100 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122603143","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：