ANALYSIS OF MACHINE LEARNING METHODS IN THE TASK OF SEARCHING DUPLICATES IN THE SOFTWARE CODE

The object of the study is code in the Python programming language, analyzed by machine learning methods to identify clones. This work is devoted to the study of machine learning methods and implementation of the decision tree machine learning model in the problem of finding clones in the program code. The paper also analyzes existing machine learning approaches for detecting duplicates in program code. During the comparison, the advantages and disadvantages of each algorithm were determined, and the results were summarized in the corresponding comparison tables. As a result of the analysis, it was determined that the method based on the decision tree, which gives the best result in the task of finding clones in the program code, is the most optimal both from the point of view of accuracy and from the point of view of implementation. The result of the work is a created model that, with an accuracy of more than 99 %, classifies cloned and non-cloned codes on an automatically generated dataset in a minimal amount of time. This system has several open questions for future research, the list of which is presented in this work. The proposed model has the following ways of further development: universal datasets. The perspective of the work lies in training a decision tree model for accurate and fast detection of code clones, which can potentially be widely used for plagiarism detection in both educational institutions and IT companies. cloud services that do not provide such capabilities. Thanks to this, it is possible to obtain a predictive model of the behavior of bank customers with a fairly high degree of accuracy. It is worth noting that in order to make a prediction (forecast), it is necessary to study the context of the task, process the data, build various machine learning algorithms, evaluate the quality of the models and choose the best of them. results of their application, method of path metric computa-tion on the given topology by the amount of «hops» (transi-tions between network nodes), optimization of procedure for finding the path using SDN technology capabilities. These methods were united and integrated into the development of the overall routing algorithm, which is proposed in this article. The proposed multipath routing algorithm allows for the improvement of the process of traffic construction in the SDN network. This was achieved by decreasing the time complexity of the routing algorithm through the usage of previously known paths in the topology without the need to construct new ones. Involvement in the modification of the algorithm of forming distance vectors facilitated timely network reconfiguration in case its state changed. Using a centralized SDN controller made it possible to increase the stability of the network and save all configuration data in one place. The above factors make it possible to deploy an SDN network on an Edge architecture. Obtained results of the application of the multipath routing algorithm allow to consider it effective when compared with previously proposed algorithms, based on obtained results from a practical network model, where the proposed multipath routing algorithm is used. This is because the research task was formed correctly, and the solution for this task gave correct results. Results of using the described algorithm are demonstrated, and an analysis of the obtained results is conducted, which makes it possible to confirm the accuracy of scientific research. stability, mobility and throughput. A formalized description of the task of improving cyber security in a special purpose information and telecommunications network is provided. The components that affect the level of cyber security of the special purpose information and telecommunication network during the group’s operations have been established. The impact of the specified conditions and factors must be reflected: in the planning documents during the planning of the deployment and operation of the group’s information and telecommunications system; in the software, during operational management. tasks This especially applies to operational commands. Therefore, in order to suc-cessfully carry out changes in the system of topographic support regarding the necessary composition of forces and devices of this support, appropriate scientific researches were conducted. This, in turn, will provide an opportunity to improve existing methods and develop recommendations regarding the necessary composition of forces and devices of topographic support for the troops of the operational command. The men-tioned recommendations will make it possible to justify the composition of the forces and devices of topographical support of the troops of the operational command by the persons planning the operations of the groups of troops (forces). are the samples of weapons and military equipment that are the part of the groups of troops (forces). However, the results of the analysis show that the existing methods for substan-tiating the composition of samples of weapons and military equipment need improvement. First of all, this concerns the determination of the basic (support) version of the composition of samples of weapons and military equipment (WME). As in the existing methods, the basic (support) version of the composition is not determined, it is chosen by comparing the composition of one’s troops and the enemy’s troops according to their combat potential. This approach does not provide an opportunity to compare the groups, taking into account the specifics of the use of their striking equipment and to create the necessary balance of forces at all stages of the operation. Taking into account the above, we conducted the researches that made it possible to determine that with certain proportions, characteristic of the organizational structure of the absolute majority of military formations, there is a close to linear relationship between the number of their personnel and WME and combat potential. Based on the research results, an improved method of determining the basic (support) composition of samples of weapons and military equipment in operation is proposed. were carried out according to DBN B.1.1-31:2013. Data obtained theoretically, during modeling, and experimentally are compared with each other. The expediency of taking into account the results obtained with the help of the INSUL software when writing a new methodology for the state standard or for improving the lat-ter is considered. Comparing real and simulated amplitude-frequency characteristics, it is possible to conclude that the INSUL software complex is not accurate at high frequencies. This phenomenon may be related to the inherent resonances of the existing partition, or to the quality of the installation, which can be equated to a setup error, due to the lack of pos-sibility of influencing the quality of the construction process of the measurement object. A possible error occurs when the soundproof partition is violently connected to the existing structures of the building, or due to the presence of resonat-ing technical elements in the cinema hall (air ducts, metal structures of the stage and seats, etc.). This error is not in-fluential due to the not significantly small wavelength in the frequency of 3000–8000 Hz. Therefore, as a conclusion, the INSUL software can be used in the development of the methodology of state standards for sound insulation in buildings. This paper is the first result of an experimental study of manufacturing deformation monitoring equipment using a position sensor, an efficient solution for contractors in bridge monitoring during either process of construction or operation in the context of the increasing number of large bridges. The object of the research is bridgework. And the work is aimed at achieving two tasks, which include meeting the need of the surveyors in real-time monitoring and warn-ing during the construction process with a suitable price and reducing the reliance on manufacturers in supplying and operating monitoring systems. The study used a KTR position sensor, a Linear Vari-able Differential Transformer, and an Arduino board for embedded coding. Data from the KTR sensor are acquired, processed, and controlled by a program written in Python. The results of the study are compared to the true observa-tions from a Mitutoyo Palmer which has 0.01 mm accuracy in a laboratory.The initial results from the laboratory show prospects for applications in reality, and accuracy is suitable for technical measurements. This is also a good choice for surveyors in acquiring continuously monitoring data at high accuracy for bridge monitoring in general and structure in particular. The study is fundamental to expanding the number of channels from 1 to 8 or 16 for monitoring a full cross-section. At the same time, cable connection mode will be developed to Wi-Fi or Bluetooth mode for online observation. The results of the study confirm the scientificity, and feasibility of the solution. This solution can be applied for either bridge monitoring or other monitoring fields and can be produced by contractors. The cost of monitoring projects will be significantly reduced and there will be no more disruption of monitoring projects after suppliers leave as a new system will be quickly added or replaced.

The object of the research is modern online services and machine learning libraries for predicting the probability of the bank client's consent to the provision of the proposed services. One of the most problematic areas is the high unpredictability of the result in the field of banking marketing using the most common technique of introducing new services for clients -the so-called cold calling. Therefore, the question of assessing the probability and predicting the behavior of a potential client when promoting new banking services and services using cold calling is particularly relevant.
In the course of the study, libraries of machine learning methods and data analysis of the Python programming language were used. A program was developed to build a model for predicting the behavior of bank customers using data processing methods using gradient boosting, regularization of gradient boosting, random forest algorithm and recurrent neural networks. Analogous models were built using cloud machine learning services Azure ML, BigML and the Autosklearn library.
Data analysis and prediction models built using Python language libraries have a fairly high quality -an average of 94.5 %. Using the Azure ML cloud service, a predictive model with an accuracy of 88.6 % was built. The BigML machine learning service made it possible to build a model with an accuracy of 88.8 %. Machine learning methods from the Auto-sklearn library made it possible to obtain a model with a higher quality -94.9 %. This is due to the fact that the proposed libraries of the Python programming language allow better customization of data processing methods and machine learning to obtain more accurate models than free cloud services that do not provide such capabilities.
Thanks to this, it is possible to obtain a predictive model of the behavior of bank customers with a fairly high degree of accuracy. It is worth noting that in order to make a prediction (forecast), it is necessary to study the context of the task, process the data, build various machine learning algorithms, evaluate the quality of the models and choose the best of them.
Keywords: artificial intelligence, machine learning methods, banking services, credit scoring, credit risk.  The object of the research is the creation of a multipath routing algorithm for software-defined networks (SDN) including known paths, subject of the research is network parameters, designed according to a certain topology and using the developed algorithm. One of the most problematic aspects of contemporary computer networks (including applied SDN networks) is overloading. This results in harder control and limiting traffic and amount of users. Most routing algorithms that are used today have a rather large time complexity.
In the course of the study, the following methods were used: study of known routing solutions for SDN networks and ISSN 2664-9969 results of their application, method of path metric computation on the given topology by the amount of «hops» (transitions between network nodes), optimization of procedure for finding the path using SDN technology capabilities. These methods were united and integrated into the development of the overall routing algorithm, which is proposed in this article.
The proposed multipath routing algorithm allows for the improvement of the process of traffic construction in the SDN network. This was achieved by decreasing the time complexity of the routing algorithm through the usage of previously known paths in the topology without the need to construct new ones. Involvement in the modification of the algorithm of forming distance vectors facilitated timely network reconfiguration in case its state changed. Using a centralized SDN controller made it possible to increase the stability of the network and save all configuration data in one place. The above factors make it possible to deploy an SDN network on an Edge architecture.
Obtained results of the application of the multipath routing algorithm allow to consider it effective when compared with previously proposed algorithms, based on obtained results from a practical network model, where the proposed multipath routing algorithm is used. This is because the research task was formed correctly, and the solution for this task gave correct results. Results of using the described algorithm are demonstrated, and an analysis of the obtained results is conducted, which makes it possible to confirm the accuracy of scientific research.
Keywords: software-defined networking, multipath routing, known routes, distance vector.  The increase in the number of cases of failure of information and telecommunication networks due to cyber attacks determines the need to protect them from this type of attacks. The issue of increasing cyber security in the conditions of conducting operations by groups of troops (forces) is very important. Due to the armed conflict in the East of Ukraine, the military-political instability in the Middle East, the struggle for influence on world financial and energy flows, the global military-political instability is intensifying. This is due to an increase in the number of communication devices in information and telecommunication networks, as well as an increase in the number of possible attacks that can be used to disrupt the operation of an information and telecommunication network. Considering the above, the object of research is a special purpose information and telecommunication system. The subject of research is cyber security of a special purpose information and telecommunication system. Classical methods of scientific knowledge, namely analysis and synthesis, were used during the research. The research identifies factors that affect cyber security in a special purpose information and telecommunications system. All this must be taken into account while planning and deploying a special purpose information and telecommunication system. The analysis of the consequences of the impact on the information and telecommunications system of special purpose of modern devices of defeat and the impact of devices of radio-electronic suppression and other factors was carried out. A typical special purpose information and telecommunication system does not fully meet the requirements for constant readiness to ensure the management of troops (forces), ISSN 2664-9969 stability, mobility and throughput. A formalized description of the task of improving cyber security in a special purpose information and telecommunications network is provided. The components that affect the level of cyber security of the special purpose information and telecommunication network during the group's operations have been established. The impact of the specified conditions and factors must be reflected: in the planning documents during the planning of the deployment and operation of the group's information and telecommunications system; in the software, during operational management.
Keywords: cyber security, radio-electronic environment, special purpose information and telecommunication system, operational management, cyber security.  The object of the research is the forces and devices of topographic support of the operational command troops. The work highlights the recommendations regarding the composition of the forces and devices of topographical support of the operational command troops.
The analysis of the nature of the armed struggle of the last decades shows a sharp increase in the need of the troops (forces) for information provision, in particular in the provision of topogeodesic information (geospatial information) of the armed forces, the timeliness and reliability of which affects the course and results of military operations. One of the types of geospatial data that is needed by the military both in peacetime and in a special period are topographic maps, they are the main source of information about the terrain and the objects on it. Nowadays, the state of topographic information on most nomenclature sheets of topographic maps of the territory of Ukraine does not meet the modern requirements of the troops.
The successful implementation of the task of updating topographical information is facilitated by the use of the most acceptable and economically justified composition of forces and devices of topographical support of troops. However, the experience of geodetic support during the operational training of troops (forces), the operation of the United Forces in the east of the country and the large-scale armed aggression by the Russian Federation showed that the currently available forces and devices of geodetic support of the troops are not capable of fulfilling the entire scope of the assigned tasks This especially applies to operational commands. Therefore, in order to successfully carry out changes in the system of topographic support regarding the necessary composition of forces and devices of this support, appropriate scientific researches were conducted. This, in turn, will provide an opportunity to improve existing methods and develop recommendations regarding the necessary composition of forces and devices of topographic support for the troops of the operational command. The mentioned recommendations will make it possible to justify the composition of the forces and devices of topographical support of the troops of the operational command by the persons planning the operations of the groups of troops (forces).
Keywords: operational command, topographic support, timeliness and reliability of information, geospatial data, topographic data.  During the research of the use issue of samples of weapons and military equipment in operations, a significant correlation was established between the predicted effectiveness of such use and the characteristics of the quantitative and qualitative composition of samples of weapons and military equipment. Taking this into account, it is clear that while preparing any operation, it would be desirable to have such a basis of the composition of samples of weapons and military equipment, which is well-founded and can only be adjusted in the conditions of a certain operation. The objects of the research are the samples of weapons and military equipment that are the part of the groups of troops (forces). However, the results of the analysis show that the existing methods for substantiating the composition of samples of weapons and military equipment need improvement. First of all, this concerns the determination of the basic (support) version of the composition of samples of weapons and military equipment (WME). As in the existing methods, the basic (support) version of the composition is not determined, it is chosen by comparing the composition of one's troops and the enemy's troops according to their combat potential. This approach does not provide an opportunity to compare the groups, taking into account the specifics of the use of their striking equipment and to create the necessary balance of forces at all stages of the operation.
Taking into account the above, we conducted the researches that made it possible to determine that with certain proportions, characteristic of the organizational structure of the absolute majority of military formations, there is a close to linear relationship between the number of their personnel and WME and combat potential.
Based on the research results, an improved method of determining the basic (support) composition of samples of weapons and military equipment in operation is proposed.
Keywords: samples of weapons and military equipment, predicted efficiency of use, composition of troops, multicriteria optimization methods. The object of research is the amplitude-frequency characteristic of sound insulation of multi-layer heterogeneous structures. One of the most problematic places is the calculation methodology, due to the existence of standards such as ISO, DBN, which have some differences between them, and software, the calculation methodology of which is not fully known, due to the copyright of the latter.
Using the modeling method in the INSUL software complex, the amplitude-frequency characteristics of the sound insulation of the partition walls between the halls were obtained. On-site measurements were carried out on a real object identical to the simulated fence constructions presented by the MULTIPLEX cinema in the DAFI shopping center in Dnipro (Ukraine). The calculation methodology of the INSUL software is completely unknown. According to information from open sources, it can be concluded that linear modeling of the amplitude-frequency characteristic is used. The measurements were carried out according to DBN B.1.1-31:2013. Data obtained theoretically, during modeling, and experimentally are compared with each other. The expediency of taking into account the results obtained with the help of the INSUL software when writing a new methodology for the state standard or for improving the latter is considered. Comparing real and simulated amplitudefrequency characteristics, it is possible to conclude that the INSUL software complex is not accurate at high frequencies. This phenomenon may be related to the inherent resonances of the existing partition, or to the quality of the installation, which can be equated to a setup error, due to the lack of possibility of influencing the quality of the construction process of the measurement object. A possible error occurs when the soundproof partition is violently connected to the existing structures of the building, or due to the presence of resonating technical elements in the cinema hall (air ducts, metal structures of the stage and seats, etc.). This error is not influential due to the not significantly small wavelength in the frequency of 3000-8000 Hz. Therefore, as a conclusion, the INSUL software can be used in the development of the methodology of state standards for sound insulation in buildings.
Keywords: soundproofing, soundisolation, building regulations, architectural acoustics, building acoustics, building structures, soundproofing structures. This paper is the first result of an experimental study of manufacturing deformation monitoring equipment using a position sensor, an efficient solution for contractors in bridge monitoring during either process of construction or operation in the context of the increasing number of large bridges. The object of the research is bridgework. And the work is aimed at achieving two tasks, which include meeting the need of the surveyors in real-time monitoring and warning during the construction process with a suitable price and reducing the reliance on manufacturers in supplying and operating monitoring systems.

THE FIRST PHASE OF RESEARCH AND
The study used a KTR position sensor, a Linear Variable Differential Transformer, and an Arduino board for ISSN 2664-9969 embedded coding. Data from the KTR sensor are acquired, processed, and controlled by a program written in Python. The results of the study are compared to the true observations from a Mitutoyo Palmer which has 0.01 mm accuracy in a laboratory.
The initial results from the laboratory show prospects for applications in reality, and accuracy is suitable for technical measurements. This is also a good choice for surveyors in acquiring continuously monitoring data at high accuracy for bridge monitoring in general and structure in particular. The study is fundamental to expanding the number of channels from 1 to 8 or 16 for monitoring a full cross-section. At the same time, cable connection mode will be developed to Wi-Fi or Bluetooth mode for online observation. The results of the study confirm the scientificity, and feasibility of the solution. This solution can be applied for either bridge monitoring or other monitoring fields and can be produced by contractors. The cost of monitoring projects will be significantly reduced and there will be no more disruption of monitoring projects after suppliers leave as a new system will be quickly added or replaced.