Pub Date : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110727
R. I. Zaev, A. Romanov, R. Solovyev
Medical research has made tremendous progress in detecting various pathologies in the human body. There is still the problem of the speed of the process, and the lack of a sufficient number of trained professionals in this field. Detection of prostate cancer, in particular, without surgery is a very labor- intensive process. A neural network-based machine learning algorithm has been proposed to solve this problem, making it possible to see suspected areas of lesions in the organ. In this study, a comprehensive analysis of TRUS image processing approaches was carried out, and an algorithm architecture was developed to segment the affected areas. Based on this analysis, we have developed a system for automatic detection and segmentation of prostate cancer.
{"title":"Segmentation of Prostate Cancer on TRUS Images Using ML","authors":"R. I. Zaev, A. Romanov, R. Solovyev","doi":"10.1109/SmartIndustryCon57312.2023.10110727","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110727","url":null,"abstract":"Medical research has made tremendous progress in detecting various pathologies in the human body. There is still the problem of the speed of the process, and the lack of a sufficient number of trained professionals in this field. Detection of prostate cancer, in particular, without surgery is a very labor- intensive process. A neural network-based machine learning algorithm has been proposed to solve this problem, making it possible to see suspected areas of lesions in the organ. In this study, a comprehensive analysis of TRUS image processing approaches was carried out, and an algorithm architecture was developed to segment the affected areas. Based on this analysis, we have developed a system for automatic detection and segmentation of prostate cancer.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131106443","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 : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110747
A. Prasolov, S. Andreev, I. Nazarov
Hot blast stoves are intermittent heat exchangers designed to heat the air supplied into the blast furnace. To provide the continuity of the hot blast supply, separate stoves are combined into a unit. The definition of the time of heating and cooling periods for each stove is based on the calculation of providing the blast furnace with a hot blast with the maximum possible temperature for the entire period of operation of the hot blast stove's unit. Due to the different characteristics of the stove checkerworks, a reasonable heating time of each stove will be different; it changes over time, due to the checkerwork degradation. The studies conducted showed that to describe the state of the stove checkerworks in the unit, it is possible to use a linguistic variable with the terms ("low") and ("hard") characterizing the checkerwork capacity of heat accumulation. In this paper, a fuzzy controller is developed using the linguistic variable introduced. The controller allows controlling the heating time of each stove to provide the maximum heat absorption by the checkerwork taking into account its condition. Simulation modeling of the blast heating process showed that, providing a constant cycle time of the stove unit, the blowing temperature can be increased by 18 °C.
{"title":"The algorithm of Intelligent Control Adjustment of the Mode Map of Hot Blast Stove’s Unit Based on Fuzzy Logic","authors":"A. Prasolov, S. Andreev, I. Nazarov","doi":"10.1109/SmartIndustryCon57312.2023.10110747","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110747","url":null,"abstract":"Hot blast stoves are intermittent heat exchangers designed to heat the air supplied into the blast furnace. To provide the continuity of the hot blast supply, separate stoves are combined into a unit. The definition of the time of heating and cooling periods for each stove is based on the calculation of providing the blast furnace with a hot blast with the maximum possible temperature for the entire period of operation of the hot blast stove's unit. Due to the different characteristics of the stove checkerworks, a reasonable heating time of each stove will be different; it changes over time, due to the checkerwork degradation. The studies conducted showed that to describe the state of the stove checkerworks in the unit, it is possible to use a linguistic variable with the terms (\"low\") and (\"hard\") characterizing the checkerwork capacity of heat accumulation. In this paper, a fuzzy controller is developed using the linguistic variable introduced. The controller allows controlling the heating time of each stove to provide the maximum heat absorption by the checkerwork taking into account its condition. Simulation modeling of the blast heating process showed that, providing a constant cycle time of the stove unit, the blowing temperature can be increased by 18 °C.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127092367","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 : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110788
D. Perepelkin, M. Ivanchikova, Tin Nguyen
Technology software defined networks (SDNs) allows managing the network at the software level, expanding the functionality to control the process of data transfer in the network. SDNs provide a new level of network application abstraction to provide a level of quality of service. The purpose of the work is to develop a scientific approach to the study of multipath routing and load balancing in the SDNs based on the bird migration algorithm (BMA). The article studies and analyzes the bird migration algorithm to solve the problem of multipath routing in the SDNs. A visual software system SDNLoadBalancer has been developed and an experimental SDN topology has been designed, which makes it possible to study in detail the processes of multipath routing in the SDNs based on the proposed approach. The article also compares the proposed approach with the results of the genetic algorithm and the artificial bee colony algorithm. The obtained results show the effectiveness of the proposed approach in the SDNs, made it possible to obtain results close to optimal in an acceptable time, and also reduce the transmission delay jitter in the network.
{"title":"Research of Multipath Routing and Load Balancing Processes in Software Defined Networks Based on Bird Migration Algorithm","authors":"D. Perepelkin, M. Ivanchikova, Tin Nguyen","doi":"10.1109/SmartIndustryCon57312.2023.10110788","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110788","url":null,"abstract":"Technology software defined networks (SDNs) allows managing the network at the software level, expanding the functionality to control the process of data transfer in the network. SDNs provide a new level of network application abstraction to provide a level of quality of service. The purpose of the work is to develop a scientific approach to the study of multipath routing and load balancing in the SDNs based on the bird migration algorithm (BMA). The article studies and analyzes the bird migration algorithm to solve the problem of multipath routing in the SDNs. A visual software system SDNLoadBalancer has been developed and an experimental SDN topology has been designed, which makes it possible to study in detail the processes of multipath routing in the SDNs based on the proposed approach. The article also compares the proposed approach with the results of the genetic algorithm and the artificial bee colony algorithm. The obtained results show the effectiveness of the proposed approach in the SDNs, made it possible to obtain results close to optimal in an acceptable time, and also reduce the transmission delay jitter in the network.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129454071","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 : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110839
A. Zakharov, Irina G. Zakharova, Dmitriy Korenev
The article proposes a new approach to the transfer and storage of data, including from IoT devices, for medical systems to improve the information security of these systems. The proposed approach uses the blockchain methodology for storing batch and streaming information from various wearable devices designed to take medical indicators (blood pressure, pulse, temperature, etc.) in real time. For this, a blockchain from blockchains is used, which allows considering the peculiarity of biomedical data when they are presented in a distributed information storage. The article presents the storage architecture, describes data storage structures, and proposes pseudocode for creating a blockchain and adding information to it. A prototype of the system is presented, including services for receiving and processing information, as well as information retrieval. Computational experiments were performed to evaluate the performance of the system prototype on a test data set. Their results showed that the creation of new blockchains can be more time-consuming compared to traditional data storage. However, the proposed architecture, along with ensuring data protection, allows you to not lose the search speed.
{"title":"Blockchain Architecture for Secure Storage of IoT Data","authors":"A. Zakharov, Irina G. Zakharova, Dmitriy Korenev","doi":"10.1109/SmartIndustryCon57312.2023.10110839","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110839","url":null,"abstract":"The article proposes a new approach to the transfer and storage of data, including from IoT devices, for medical systems to improve the information security of these systems. The proposed approach uses the blockchain methodology for storing batch and streaming information from various wearable devices designed to take medical indicators (blood pressure, pulse, temperature, etc.) in real time. For this, a blockchain from blockchains is used, which allows considering the peculiarity of biomedical data when they are presented in a distributed information storage. The article presents the storage architecture, describes data storage structures, and proposes pseudocode for creating a blockchain and adding information to it. A prototype of the system is presented, including services for receiving and processing information, as well as information retrieval. Computational experiments were performed to evaluate the performance of the system prototype on a test data set. Their results showed that the creation of new blockchains can be more time-consuming compared to traditional data storage. However, the proposed architecture, along with ensuring data protection, allows you to not lose the search speed.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"34 11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134361840","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 : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110807
S. Staroletov
The current pace of development of cyber-physical systems requires the elaboration of fast methods for analyzing data circulating in them. Anomalies are patterns of data that do not conform to the concept of normal (expected) behavior. The data available on vehicular CAN bus reflects the current state of a vehicle, and it is the product of the engine control system and various sensors. In the present paper, we present software to process data from the CAN bus with the goal to detect anomalies in it. Often the data circulated in a vehicle is vendor-specific, in addition, we consider various methods for finding anomalies, therefore, it is advisable to design extensible software in the form of a software framework. The work is intended for the Jetson Nano platform, but can be run on another embedded Linux platform with restrictions on detection methods. We discuss hardware and software methods to obtain information on current state of the vehicle in real time, and then we briefly study how to implement anomaly analysis methods on the received data. Evaluation of detection methods is not included in the goals of the work; we mainly provide infrastructural methods for receiving data from the bus, decoding it and passing it to an anomaly predictor. Software was implemented in C++ with the ability to run Python code for the prediction, tests were carried out on a Mazda 6 first generation car and its ECU.
{"title":"A Software Framework for Jetson Nano to Detect Anomalies in CAN Data","authors":"S. Staroletov","doi":"10.1109/SmartIndustryCon57312.2023.10110807","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110807","url":null,"abstract":"The current pace of development of cyber-physical systems requires the elaboration of fast methods for analyzing data circulating in them. Anomalies are patterns of data that do not conform to the concept of normal (expected) behavior. The data available on vehicular CAN bus reflects the current state of a vehicle, and it is the product of the engine control system and various sensors. In the present paper, we present software to process data from the CAN bus with the goal to detect anomalies in it. Often the data circulated in a vehicle is vendor-specific, in addition, we consider various methods for finding anomalies, therefore, it is advisable to design extensible software in the form of a software framework. The work is intended for the Jetson Nano platform, but can be run on another embedded Linux platform with restrictions on detection methods. We discuss hardware and software methods to obtain information on current state of the vehicle in real time, and then we briefly study how to implement anomaly analysis methods on the received data. Evaluation of detection methods is not included in the goals of the work; we mainly provide infrastructural methods for receiving data from the bus, decoding it and passing it to an anomaly predictor. Software was implemented in C++ with the ability to run Python code for the prediction, tests were carried out on a Mazda 6 first generation car and its ECU.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130768478","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 : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110825
I. Kalmykov, D. V. Dukhovnyj, Alexander A. Olenev
Currently, the problem of ensuring universal access to the Internet resources in the Far North is very acute. It is possible to solve this problem by using low-orbit satellite Internet systems (LOSIS). As the number of LOSIS increases, the probability of unauthorized content being imposed by the offending satellite increases. This situation can be prevented by using the Spacecraft (SC) authentication protocol. The use of the modular code of residue number system (MC RNS) in the implementation of the protocol will reduce the authentication time of the spacecraft due to the transition from calculations with large numbers (140 or more digits) to parallel calculations with low-bit residues. The goal is to reduce the time spent on satellite authentication through the use of MC RNS, which will reduce the time for selecting the correct response signal by the intruder satellite and increase the information security of the LOSIS.
{"title":"Development of an Imitation-Resistant Satellite Authentication Protocol Using Modular Codes of Residue Number System","authors":"I. Kalmykov, D. V. Dukhovnyj, Alexander A. Olenev","doi":"10.1109/SmartIndustryCon57312.2023.10110825","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110825","url":null,"abstract":"Currently, the problem of ensuring universal access to the Internet resources in the Far North is very acute. It is possible to solve this problem by using low-orbit satellite Internet systems (LOSIS). As the number of LOSIS increases, the probability of unauthorized content being imposed by the offending satellite increases. This situation can be prevented by using the Spacecraft (SC) authentication protocol. The use of the modular code of residue number system (MC RNS) in the implementation of the protocol will reduce the authentication time of the spacecraft due to the transition from calculations with large numbers (140 or more digits) to parallel calculations with low-bit residues. The goal is to reduce the time spent on satellite authentication through the use of MC RNS, which will reduce the time for selecting the correct response signal by the intruder satellite and increase the information security of the LOSIS.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122215704","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 : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110755
I. F. Yasinskiy, Tatyana V. Gvozdeva, V. Tyutikov
An important issue that arises before every person is the choice of a profession. It is obvious that the potential of success in different areas can be influenced both by the knowledge and skills acquired in the process of study, as well as the character traits of a person. The article proposes an intelligent predictive system that allows to assess the student's capabilities in the area of analytics. The topic relevance is explained by the need to increase the importance of connection between the employer and the university on the formation of knowledge, skills and abilities of the student that are in demand on the labor market. When designing the prognostic structure of the system, a hybrid intellectual approach is used that combines the advantages of known methods. It includes a neural network model and a method of accounting of arguments groups. The most demanded professions in the labor market have been identified. Professional skill maps are compiled, based on the description of the requirements. The training samples of are supplemented with images generated by the Monte Carlo method. Using data on the student's progress in selected key disciplines, as well as other available information, the system offers a numerical equivalent of the potential for the declared professions. Such recommendation allows the student to timely and consciously adjust the orientation in the educational process, which positively affects the competitiveness of the labor resources produced by the higher education institution.
{"title":"Combined Method of Cognitive Assessment of the Specialist Professional Potential","authors":"I. F. Yasinskiy, Tatyana V. Gvozdeva, V. Tyutikov","doi":"10.1109/SmartIndustryCon57312.2023.10110755","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110755","url":null,"abstract":"An important issue that arises before every person is the choice of a profession. It is obvious that the potential of success in different areas can be influenced both by the knowledge and skills acquired in the process of study, as well as the character traits of a person. The article proposes an intelligent predictive system that allows to assess the student's capabilities in the area of analytics. The topic relevance is explained by the need to increase the importance of connection between the employer and the university on the formation of knowledge, skills and abilities of the student that are in demand on the labor market. When designing the prognostic structure of the system, a hybrid intellectual approach is used that combines the advantages of known methods. It includes a neural network model and a method of accounting of arguments groups. The most demanded professions in the labor market have been identified. Professional skill maps are compiled, based on the description of the requirements. The training samples of are supplemented with images generated by the Monte Carlo method. Using data on the student's progress in selected key disciplines, as well as other available information, the system offers a numerical equivalent of the potential for the declared professions. Such recommendation allows the student to timely and consciously adjust the orientation in the educational process, which positively affects the competitiveness of the labor resources produced by the higher education institution.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127200769","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 : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110787
R. Galin, Saniya B. Galina
The presented article considers the formation and engagement of a mixed heterogeneous team within a robotic system designed for completing collaborative tasks. A qualitative comparative analysis of approaches, methods and algorithms related to efficient task allocation in a collaborative robotic system by reducing the time and/or costs is carried out. A method and its corresponding algorithm for task allocation in a mixed heterogeneous team of a collaborative robotic system are presented. The proposed algorithm for task allocation with cost minimization is applicable for a mixed heterogeneous team consisting of humans and collaborative robots (i.e., heterogeneous groups of participants of various types). Thus, the within the problem statement the following is considered: different cost functions for different types of team members; limited robot activity; the dependence of the cost of robots for performing a certain type of work on the number of people. The developed method and algorithm of task allocation and work distribution within a CRS with minimizing the costs is applicable to a mixed heterogeneous team consisting of humans and cobots (groups of participants of different types), taking into account the specific heterogeneity of team members, as well as the mandatory requirements for safe interaction of participants within a CRS. The simulation of five scenarios of task allocation between the participants using the proposed solution is carried out. The results of the experiment showed the increased efficiency of a mixed heterogeneous team functioning.
{"title":"Approach to Efficient Task Allocation in a Collaborative Robotic System Using Modified Cost Functions","authors":"R. Galin, Saniya B. Galina","doi":"10.1109/SmartIndustryCon57312.2023.10110787","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110787","url":null,"abstract":"The presented article considers the formation and engagement of a mixed heterogeneous team within a robotic system designed for completing collaborative tasks. A qualitative comparative analysis of approaches, methods and algorithms related to efficient task allocation in a collaborative robotic system by reducing the time and/or costs is carried out. A method and its corresponding algorithm for task allocation in a mixed heterogeneous team of a collaborative robotic system are presented. The proposed algorithm for task allocation with cost minimization is applicable for a mixed heterogeneous team consisting of humans and collaborative robots (i.e., heterogeneous groups of participants of various types). Thus, the within the problem statement the following is considered: different cost functions for different types of team members; limited robot activity; the dependence of the cost of robots for performing a certain type of work on the number of people. The developed method and algorithm of task allocation and work distribution within a CRS with minimizing the costs is applicable to a mixed heterogeneous team consisting of humans and cobots (groups of participants of different types), taking into account the specific heterogeneity of team members, as well as the mandatory requirements for safe interaction of participants within a CRS. The simulation of five scenarios of task allocation between the participants using the proposed solution is carried out. The results of the experiment showed the increased efficiency of a mixed heterogeneous team functioning.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127886779","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 : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110798
V. Gusev
The problem of object management in conditions of uncertainty, when there is no formalized description, and measurements are made at certain intervals, is considered. It is assumed that the dependence of the objective function on the control parameter is quite smooth and does not have sharp jumps. The construction of control mechanisms in such a situation is associated with the problem of an extreme regulator that automatically searches for and maintains the extreme value of the regulated value. The problem boils down to the problem of finding the extremum for a nonstationary objective function, when its values, depending on the components of the control vector, are set only on a discrete set of moments. To find a solution, a discrete method of unconditional optimization of the gradient type is proposed. The conditions of its application are considered. The application of the method is demonstrated on the numerical model of an extreme regulator designed to manage a non-stationary object with a nonlinear objective function of tax revenues from the tax rate.
{"title":"Application of an Extreme Regulator to Control an Inertial Object","authors":"V. Gusev","doi":"10.1109/SmartIndustryCon57312.2023.10110798","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110798","url":null,"abstract":"The problem of object management in conditions of uncertainty, when there is no formalized description, and measurements are made at certain intervals, is considered. It is assumed that the dependence of the objective function on the control parameter is quite smooth and does not have sharp jumps. The construction of control mechanisms in such a situation is associated with the problem of an extreme regulator that automatically searches for and maintains the extreme value of the regulated value. The problem boils down to the problem of finding the extremum for a nonstationary objective function, when its values, depending on the components of the control vector, are set only on a discrete set of moments. To find a solution, a discrete method of unconditional optimization of the gradient type is proposed. The conditions of its application are considered. The application of the method is demonstrated on the numerical model of an extreme regulator designed to manage a non-stationary object with a nonlinear objective function of tax revenues from the tax rate.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115336793","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 : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110769
N. Smirnov, A. S. Trifonov
This paper deals with the task of text messages classification. The authors analyzed and reviewed the results of other researchers in this task and provided a brief overview of the machine learning and deep learning methods used in the study. The dataset of 1200 incoming messages of university admission campaign was used in the study. The authors pre-processed message texts, classified messages in three ways and applied three types of text vectorization. Based on machine learning and deep learning methods, the authors developed and applied multiclass and binary message classifiers. The paper presents classification metrics and confusion matrices for tasks of multiclass and multilabel classification. The models that provide the highest f1 score were selected as the best models.
{"title":"Classification of Incoming Messages of the University Admission Campaign","authors":"N. Smirnov, A. S. Trifonov","doi":"10.1109/SmartIndustryCon57312.2023.10110769","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110769","url":null,"abstract":"This paper deals with the task of text messages classification. The authors analyzed and reviewed the results of other researchers in this task and provided a brief overview of the machine learning and deep learning methods used in the study. The dataset of 1200 incoming messages of university admission campaign was used in the study. The authors pre-processed message texts, classified messages in three ways and applied three types of text vectorization. Based on machine learning and deep learning methods, the authors developed and applied multiclass and binary message classifiers. The paper presents classification metrics and confusion matrices for tasks of multiclass and multilabel classification. The models that provide the highest f1 score were selected as the best models.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130320974","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：