1-Article

Text to Audio Converter for Learning Uzbek

Artikova M, Talipova O

This article gave an example of creating a program for translating text into audio. In the conditions of the modern language situation, the national school faces an important task - the formation of the ability to solve various communicative tasks in certain areas and situations of communication by means of language. The ultimate goal of teaching languages at school is the practical mastery of them. Achieving this goal is directly dependent on the use of effective methods, techniques and teaching aids. Now in the public domain there are many speech synthesizers that allow you to convert text into voice. They are developed by a variety of companies and independent users, constantly improving reading algorithms, making the tool even more accurate during its work. For successful playback and additional settings, you will need special software, the functionality of which allows you to implement the task. The program in this article is very similar to one of the google translator functionality. Unfortunately, due to the limited base of google languages, the translator cannot translate some words of the Uzbek language into audio. This article describes the technology and process for creating this software. The software was created in the form of a web application.

2-Article

Study of the Dependence Electrical Resistance of Bioactive Points in the Meridians on the Amount of Glucose in the Blood

Alijon Guliev

This article proposes the use of a non-invasive method based on measuring the electrical resistance of bioactive points located in the meridians of the human body in diagnosing diabetes. Using analysis of variance, it was found that the existing meridians in the human body are not equally important for diagnosing diabetes, shows that each meridian is actually related to some internal organs in the body. An analysis of variance was carried out, and the influence of one or more factors affecting a random variable in complex situations was assessed by carefully studying various sources that cause differences in the values of a random variable over the same time interval. Practical use has been made, and as a result, it has been found out that there are many cases when only one factor affects the final result, and this factor takes a finite number of values. Numerous computational experiments were carried out on the basis of software products, various situations were tested, and the results were analyzed using a single-factor analysis of variance. It is noted that if a person has a disease before his symptoms are detected, all pathological signs first reach the meridians along the nerve fibers, and then reach the points already existing in these meridians. As a result, the disease leads to a meridional imbalance, and there are sharp changes in the sum of biophysical indicators at points located in the meridians.

3-Article

Monitoring of the hydrogeological characteristics of aquifers by automated measuring systems

Djumanov Jamoljon, Khushvaktov Saydulla, Kuchkorov Temurbek, Anorboyev Erkin

In this paper presented, the general structure of the proposed control system device, the capabilities and differences of the proposed device are expressed with graphical and tabular data. The studies consider the measurement of groundwater level or depth to water, which is critical for identifying long-term trends, as well as changes in levels, including the decline in groundwater levels. Automated measurements for aquifer management were tested, and the need for a regime and timely and accurate data for assessing the state of groundwater in order to manage and prevent adverse situations was identified. It is noted that the main source of information on the hydrological loads acting on aquifers and their impact on the recharge, storage, and discharge of groundwater are measurements of the level, temperature, and salinity of groundwater. The studies mainly discussed research work on the use of automated devices with several benefits and considered their further improvement. The device was tested, both synchronously with other devices, and operate autonomously and do not require frequent visits of the observer to the well, certain economic advantages were also evaluated and it is planned to use the device in dangerous and hard-to-reach places. Has presents ways of creating remotely monitoring of elements of the hydro regime characteristics of aquifers in observation wells based on proposed device of hardware and software systems.

4-Article

MODELS AND ALGORITHMS FOR REMOTE MONITORING OF THE CONDITION OF OVERHEAD POWER LINES

D.A.Davronbekov, A.F.Khayrullaev

Electricity theft is the biggest problem today, which causes huge losses to electrical panels. In addition, to hide these losses, the price will eventually increase. If we can prevent these thefts, we will be able to save a lot of energy. By tracking the electricity used, you will find out where the best opportunity to save energy is. More than one-third of the electricity generated is lost due to electricity theft, power loss and inefficiency of the distribution system. Prohibited or illegal use of electricity not only has an economic impact, but also hinders the design and modeling stages of the power system. This paper proposes a real-time remote transmission line condition monitoring system to ensure the prospects of safe operation of overhead power lines. The proposed system is designed for low-voltage transmission lines (0.4 - 6 kV) and provides information about the condition of the line between the distribution transformer and the consumer meter. Many households indulge in different forms of electricity theft and illegal tampering of electric metering devices. These lead to distribution system faults and overload as well as loss of revenue by the distribution companies. In particular, the application of sensor network in the remote monitoring system, the use of the modern Arduino Mega microcontroller offers the possibility of more accurate and reliable monitoring of the characteristics of overhead power lines.

5-Article

The main characteristics of optical fiber amplifiers for optical transmission systems

Siddikov Isomiddin Khakimovich, Fayzullayeva Barno Bakxadirovna

The article is dedicated to an overview of the various options for optical fiber amplifiers, the principles of their operation, advantages and disadvantages. The article presents drawings of various characteristics of amplifiers and a diagram of a structure with a combined DFB - laser and optical amplifier, and a functional diagram of a FOA - a fiber optic amplifier. When transmitting optical signals over fiber communication lines over long distances, due to scattering and absorption, which attenuate the information signal, it is necessary to use repeater amplifiers in series. We know that there are two types of repeaters – optic-electronic, based on the conversion of “light - electronic signal - light", and fully optical, which have recently become widespread. The main limiting factors in fiber optic transmission systems are attenuation, dispersion, and non-linear optical effects. This chapter deals with devices - optical amplifiers, which compensate for optical power losses that occur in optical fibers, connectors, passive splitters, etc.

6-Article

Algorithm and program for solving the gas filtration problem in dynamically interconnected porous media

Elmira Nazirova, Аbdugani Nematov, Mohiniso Mahmudova, Khurshida Karabaeva

The article examines the actual problem of developing gas fields in order to increase gas extraction from productive layers and to determine the main indicators of the research object. This article presents an analysis of scientific works related to the problem of mathematical modeling of gas filtration processes in multilayer porous media. For a comprehensive study of the considered process, a mathematical model of the process of gas filtration in a non-homogeneous three-layer porous medium with weakly conductive interlayers and the process of their dynamic interaction was developed based on the basic laws of hydromechanics. Of course, in the numerical modeling of such a complex process, the finite difference method was used, and due to the fact that the equation is nonlinear with respect to the pressure function, the quasi-linear method and iterative methods were used to solve it. The article is devoted to the algorithm and program for solving the boundary value problem for calculating the main indicators of gas deposits in a three-layer porous medium, it presents the mathematical model of the boundary value problem of gas filtration in a dynamically interconnected porous medium, the block diagram of the solution algorithm and the program, the interface of the program was developed in the Delphi programming language. MatLab programming environment was used to visualize the results and present the graphs in 2D and 3D form. On the basis of the mathematical tool proposed in the work, computational experiments were carried out on different values of layer permeability coefficients in porous media with a weakly conductive layer, and the results were presented in the form of graphic objects and an analysis of their calculations was presented.

7-Article

SURVEY ON METHODS AND MODELS FOR REDUCING FALSE ALARM RATE OF IDS

Gulomov Sh. R., Bozorov S. M.

This paper presents the importance of intrusion detection systems to detect network attacks, limitations of existing systems and problems with high false alarm rates and reducing it. As summary of analyzed related works defects of researchers has been expressed. At the main part the base requirements to build intrusion detection systems with low false alarm rate are also shown, analyzed and invited most significant ones. Presented models with continuous and differential patterns for decrease false positives. Therefore, some intrusion detection systems methods and models like multilayer perceptron, backpropagation, fuzzy logic have been analyzed step by step to reduce the false alarm rate and improve detection accuracy. Also described effects of choosing right datasets, importance of features in dataset and their selection to train the model, clustering and classification algorithms to false alarm rate of intrusion detection. In the conclusion achieved that using hybrid architectures for build intrusion detection systems may occur detection accuracy and resource consumption.

8-Article

Mathematical modelling of nonlinear processes of magneto-elastic thin plates of complex configuration and computational experiments

Shohruh Safarov, Zarif Qodirov

In this article, a mathematical model based on the Hamilton-Ostrogradsky variational principle is presented. Using the Kirkhgoff-Lyav hypothesis, transform the mathematical model in 3D form into his 2D model. The tensor view is used to determine variational representations of potential and kinetic energy, as well as changes in work done by external forces, Cauchy relations, Hooke's law, Lorentz forces, and Maxwell's electromagnetic forces. In this case, the influence of the electromagnetic field on the deformation stress state of the magnetoelastic plate is taken into account. The result was a mathematical model in the form of a system of higher-order differential equations with special derivatives with initial and boundary conditions on the displacements. To solve this problem, we developed computational algorithms, created practical software tools, performed computational experiments, and analyzed the results obtained.

9-Article

ALGORITHM FOR BUILDING AND TRAINING A NEURAL NETWORK TO SOLVE THE PROBLEM OF SPEECH RECOGNITION

Kamolov Nodirjon Ma’murjon o’g’li

This article considers the problem of speech recognition using the example of word recognition learned through the reverse error distribution method. The article presents the practical results of training a constructed neural network with training samples of different sizes. One of the promising areas to solve speech recognition problems is the use of artificial neural networks. Neural networks are widely used to solve various classes of speech recognition problems, taking into account their ability to generalize. A method to solve the speech recognition problem is considered using the example of recognizing individual words of a limited dictionary using a direct propagation neural network trained by the error back propagation method. The practical results of training the constructed neural network at different sizes of the training sample are presented.

10-Article

DETECTION OF TRANSIT DATA (TIME-OF-FLIGHT -TOF) FOR ULTRASONIC FLOW METER MEASUREMENT.

H.N. Zaynidinov, A.M. Rasulov, M.A. Kuchkarov, X.X. Mamirov, T.B. Boltaniyozov

In the present paper, an algorithm to compute an accurate TOF is proposed. The form of the electric signal received by the transducer is obtained from an acoustically-forced underdamped oscillator model, and the analytical solution of the model is proposed as a reference wave. In order to validate the effectiveness of this procedure, an ultrasonic flowmeter system is designed and tested in a flowmeter calibration test rig. It is demonstrated that the use of the presented scheme overcome the average method limitations, and turns out to be a convenient solution in a wide range of conditions. Robust measurements of near-zero flow values are acquired, which allow the achievement of a high dynamic range. The error curve of the proposed system have been obtained, revealing that the absolute value of the relative errors are lower than 2% within all the spectrum of flow rates considered (from 0.2 to 150 m3/h). Results demonstrate that the algorithm provides high-precision measurements within a wide dynamic range. The algorithm is portable and versatile: it can be adapted to different types of transducers without the need of additional measurements, allowing to adjust parameters on-the-fly for an optimal performance of the ultrasonic flowmeter system.

11-Article

OVERVIEW OF SUPER-RESOLUTION IN IMAGE PROCESSING AND THEIR APPLICATIONS

A.Yusupov, Kh.Kh.Nosirov, M.M.Arabboev, Sh.A.Begmatov, M.A.Rikhsivoev, S.M.Saydiakbarov, Z.Kh.Khamidjonov, S.A.Vakhkhobov

In recent years, the interest in image super-resolution is increased sharply developing the field of research in Image processing. Image super-resolution is a technology that increases the resolution of an image, typically by upscaling it to a higher resolution than the original. This can be achieved using machine learning algorithms such as convolutional neural networks (CNNs) that have been trained on large datasets of high-resolution images. The goal of image super-resolution is to enhance the quality of low-resolution images, making them more precise and detailed. In this paper, image super-resolution concepts, types, and applications were described, also was given brief information about super-resolution frameworks by authors. Super-resolution frameworks are computational approaches or algorithms designed to enhance the resolution and quality of images or videos. These frameworks leverage various techniques, including traditional image processing methods and advanced deep learning techniques, to generate high-resolution versions of low-resolution inputs. They aim to reconstruct fine details, improve image clarity, and enhance overall visual quality. Super-resolution frameworks can be based on convolutional neural networks (CNNs). These frameworks play a crucial role in various applications such as digital photography, imaging, and video processing, where higher resolution and better image quality are desired.

12-Article

A CONVOLUTIONAL NEURAL NETWORK MODEL BASED ON WAVELETS FOR CANCER TYPE CLASSIFICATION IN LUNG TOMOGRAPHIC IMAGES

Zaynidinov H.N., Jureayev U.S.

The paper proposes a wavelet convolutional neural network (WCNN) architecture for cancer diagnosis and classification from lung tomographic images. The results show the effectiveness of the WCNN model in classifying lung images, which can improve the accuracy of lung cancer diagnosis and treatment. The proposed model consists of four main steps, namely, preprocessing, feature extraction, classification, and visualization. A Daubechies wavelet transform was used to perform the discrete wavelet transform. In the proposed model, 3rd order Daubechies wavelet decomposition is applied to the images. The results show that the proposed method gives good results with 99.9% accuracy.

13-Article

An Overview of Video Stabilization Algorithms

Khabibullo Nosirov, Shakhzod Tashmetov, Elnur Norov

Handheld cameras have been widely used for video recording purposes. Such videos are frequently employed for different applications, ranging from personal to professional use. For individuals afflicted with high-frequency tremor, it is imperative to establish stabilization. The present medium of visual communication that records moving images and sound, commonly known as video, has become a ubiquitous form of media consumption in today's society. Its prevalence extends beyond traditional forms of entertainment, such as movies and television shows, and has been increasingly utilized in various fields including education, business, and social media. The utilization of video as a means of delivering information has demonstrated a significant impact on the way individuals perceive and comprehend various subjects, thereby becoming an essential tool in the education and training of learners. Therefore, it is crucial to understand the importance of video as a medium of communication and its impact on individuals in different contexts. The main focus of this paper also to analyze current Video Stabilization algorithms to provide wide point of view on this field.

14-Article

The Implementation of Machine Learning and Deep Learning Algorithms for Crop Yield Prediction in Agriculture

Nodir Rahimov, Dilmurod Khasanov

. In most Asian countries, since economy of the country rely on agriculture, in such countries, the agricultural system is one of the most important sectors. Crop yield prediction is a crucial task in agriculture that can help farmers make informed decisions and optimize their crop production. Accurate predictions can help farmers better plan their resources and reduce waste, ultimately leading to higher profits and a more sustainable agricultural industry. This article presents a comprehensive study on the utilization of machine learning and deep learning techniques to predict the crop yield in agriculture, implemented and compared some AI algorithms based on a given dataset. To this end, dynamic analyses data have been collected for crop yield prediction and used to construct a regression prediction model using a multivariate regression (MR), a deep neural network (DNN), multiple linear regression (MLR), gradient boosting regressor tree (GBRT) to analyze a range of agricultural factors that impact wheat crop yields. These factors include soil moisture, temperature, rainfall, and crop growth stages. The model is trained on a large dataset of wheat crop yields and corresponding agricultural factors, allowing it to learn patterns and make accurate predictions. The experiments conducted on the dataset demonstrate the effectiveness of the proposed model.

15-Article

A comprehensive review of the use of data mining algorithms in facial recognition systems for payment systems

Irgasheva D.Y, Agzamova M.Sh.

The use of electronic payment systems has become more widespread with the increase in online transactions, and the need for secure and reliable authentication methods has become a critical issue. The use of facial recognition based on unique individual biometric features is a potential solution. However, the effectiveness of facial recognition can be enhanced through the use of data mining techniques to analyze and classify facial features. Facial recognition in low-quality datasets poses a significant challenge for biometric identity verification in payment systems. In recent years, researchers have been working on improving the accuracy and reliability of facial recognition under these challenging conditions by combining adaptive loss functions and data mining methods. This article presents a study that explores the use of data mining algorithms for facial recognition in payment systems. The study highlights the importance of using data mining algorithms in facial recognition systems to enhance accuracy and security. By analyzing the work of various authors in this field, this study aims to contribute to the development of more effective and efficient facial recognition systems for payment systems.

16-Article

Machine Learning Approach for Content Delivery Network Slicing

Nazirjon Nabijanovich Khasanov

This paper explores content delivery networks and the principles of their construction. Issues to consider when implementing content delivery networks and problems that can be solved using machine learning are analyzed. The principles of network slicing in content delivery networks were studied, and a scheme for applying machine learning was proposed.

17-Article

Numerical modeling of a doubly nonlinear parabolic reaction-diffusion process

Zarnigor Fayzullayeva, Sadullayeva Shakhlo Azimboyevna

In this article, the limitation of the rate of heat propagation, spatial localization properties of mathematical models of heat propagation and diffusion processes described by the system of nonlinear equations of parabolic type in two-component environment are defined and properties of nonlinear mathematical model of convective migration rate depending on time and damping coefficient in two-component medium were studied. The effect of these factors on the studied heat transfer and diffusion processes is shown. Analyzing the solution based on the principle of comparison, it was determined that the one-way localization of the solution and the effect of convective transfer occur under the influence of convective transfer in two dimensions. The main calculation scheme of diffusion processes in moving media characterized by the system of parabolic-type nonlinear equations was built, the algorithm and methods and software were developed and the solution was visualized

18-Article

Mathematical and informational model of calculation of key indicators in development of gas fields with complex configuration

Mohiniso Mahmudova, Talipova Ozoda

The article is devoted to the construction of a mathematical and informational model of gas fields with a complex configuration. The geological and hydrodynamic parameters of the mine, the configuration of the filtration area, the wells and their discharge, the initial state of the pressure area, the permeability of the layer, the porosity, and the working hours of the wells were taken into account in the construction of the information model. On the basis of the created information model, a general scheme of the database was developed, which took into account the configuration of the filtration area, wells and their flow rates, pressure values, layer permeability, porosity, and the start and end time of well control. At the same time, a program was created based on a numerical model and a calculation algorithm, with the help of which a series of computational experiments can be conducted on a computer to study the process of oil and gas filtration in a porous medium. The result of the software is presented in a visual form in 2D and 3D graphics.

19-Article

ANALYSIS OF CUBIC SPLINE FUNCTIONS IN BIOMEDICAL SIGNAL RECOVERY

H.N. Zaynidinov, B.R. Azimov, M.M. Abduganiev

this article discusses the issues of digital processing and restoration of biomedicine signals to the electroencephalogram signal (EEG), and 21 sensors in the EEG apparatus are placed on the brain, sensors are called, their contact types, the use of bipolar contact in determining signs of the disease, the processes of interpolation of signals, separation of parts indicating signs of the disease into mass headquarters are studied. In the course of work, the spline functions Natural and Hermite were chosen as the mathematical model most convenient for digital processing of EEG signals. Based on the constructed mathematical model, an analysis of spline functions in the restoration of electroencephalogram (EEG) signals is given.

20-Article

COMPARATIVE ANALYSIS OF AI METHODS FOR ATHLETES TRAINING

M.Rikhsivoev, M.Arabboev, Sh.Begmatov, S.Saydiakbarov, Kh.Aliyarov, Kh.Nosirov, Z.Khamidjonov, S.Vakhkhobov

Athlete monitoring is essential for improving training, preventing injuries, and enhancing performance. With the advancement of wearable sensors and artificial intelligence (AI), advanced systems have been developed to track key performance indicators (KPIs). This review paper focuses on using AI techniques in athlete monitoring applications, including machine learning, deep learning, and natural language processing usage cases. These approaches help model workload, recovery, fitness, and injury risk indicators. This review paper also examines the various data sources used in athlete monitoring, such as biometric measurements, performance metrics, GPS and accelerometer data, training logs, medical records, sleep monitoring, nutrition and hydration, psychological monitoring, and environmental factors. These sources provide comprehensive insights into an athlete's physiological and psychological states. The collected data can be analyzed to identify patterns, trends, and correlations using various modeling approaches such as machine learning, predictive analytics, statistical analysis, data fusion, and real-time monitoring. This facilitates real-time feedback and personalized training programs tailored to individual athletes’ needs. AI-driven athlete monitoring systems have shown promising results in preventing injuries, enhancing performance, optimizing recovery, and providing personalized recommendations based on individual athletes’ data.

21-Article

SAFETY OF ROBOTIC SYSTEMS, THE LEVEL OF IMPORTANCE OF DEVICES FOR INFORMATION SECURITY AND PERFECT SECURITY SYSTEMS

Kodirov Zarif Zafarovich

This article discusses algorithms and tools for information security through software applications with artificial intelligence and security through devices and technologies. Intelligently designed safe robots help people in various areas of life and they work for people. Regular monitoring, maintenance, and updates are essential to ensure that robotic systems remain safe throughout their operational life. This includes software updates, hardware inspections, and periodic risk reassessments. The safety of robotic systems is a multidimensional challenge that requires collaboration among engineers, researchers, regulators, and users. As technology advances, ongoing efforts to improve safety standards and practices will play a crucial role in the responsible integration of robots into various domains.

22-Article

Digital processing of geophysical signals by cubic basic splines

Muslimjon Kuchkarov

The article discusses the increased accuracy and efficiency in resolving signals using the spline method of processing and retrieval. At the same time, algorithms have been developed in the processing of digital signals. Based on the generated algorithms, a graph of cubic cubic splines in constant steps was constructed. A table comparing the results based on the algorithms, the experimental values, and the processed results based on the analytical functions is provided.

23-Article

Principles Of Integration And Development Of Functional Structures.

Bakhodir Muminov, Ubaydullo Arabov

Currently, there are many approaches to the implementation of interactive services in information systems. And their integration with central government services is among the important issues being discussed. In this case, we can mention such approaches to information integration as enterprise application integration, inter-organizational integration, business process integration, information integration and data extraction, transformation and loading. In particular, the principles of integration of interactive services are outlined, such as uniformity of programming technology, similar architecture of computer networks and uniformity of database structure. Here is a description of the functional structure of a software package for integration based on interactive services. This structure identifies such main components as the structure of the organization's information systems, the structure of the implementation of the interactive service, data and resources, objects for integration, as well as sending and receiving information

24-Article

MODELING OF DETERMINING THE SIZE OF CAVITATION DAMAGE IN A FLOW

O‘ktam Ibragimovich Begimov

The article presents modern methods of cavitation erosion of metal and protection against cavitation, and considers safety issues. The general hypothesis of cavitation erosion, erosion of a material, the occurrence of cavitation erosion due to impact, the impact of a flow hitting the surface of a material and causing cavitation bubbles, a jet resulting from the explosion of bubbles hitting the surface of a material and bursting, and the occurrence of inertial force is considered. The patterns of erosion of steel material in a hydraulic structure are considered and graphs are obtained that describe the erosion of steel material using formulas or semi-empirical formulas expressing bending, shear, axial movement, stretching or compression of a sample of this material. Material as a result of the application of cyclic symmetrical forces. The microstructure of blocks has various variations.

25-Article

DATA STORAGE IN BUSINESS ANALYTICS AND ETL PROCESSES

Muminov Bakhodir, Eshankulov Khamza, Murodova Rano

the purpose of this article is to introduce business intelligence and the concept of data warehousing, highlighting important aspects for business and responding to the exponential growth of technological change. Since the data warehouse is a vital element, the article explains the basic definitions, concepts, technical architectures and structured models of data and data warehouses in general. Its goal is to provide a comprehensive understanding of the role of data warehousing in business intelligence. Extract, transform, and load (ETL) processes are an integral part of a data warehouse because they allow data to be extracted from various sources, converted into a standardized format, and loaded into a central repository for subsequent analysis.

26-Article

FACTORS AFFECTING EMPLOYMENT OF GRADUATES.

Ma’ruf Kucimov

The need to create algorithms for the employment of a university graduate is one of the modern trends in the labor market. The answer to these challenges is the development of the necessary employability competencies among graduates. As a competence, the employability of graduates means the acquisition of knowledge, skills in various areas of employment and the development of new methods of adaptation in the labor market. Preparing graduates for the realities of the modern labor market is the task and concern of the university, since the indicator "level of employment of university graduates" is one of the main criteria for the effectiveness of training and an indicator of its competitiveness.

27-Article

DIVIDING SOCIAL NETWORKS INTO TWO COMMUNITIES: A MARKETING PERSPECTIVE

Dilshodbek Zakhidov

Explores the concept of partitioning customer-centric social networks into two distinct communities to optimize marketing strategies. It delves into the application of graph algorithms to identify discrete communities within broader social networks. The article illustrates the practical application of these concepts through examples of customers categorized based on their product preferences. The article emphasizes the potential for improved personalization and efficiency in marketing campaigns through such strategic targeting tactics. It also highlights the role of Maximum Likelihood Estimation (MLE) in estimating the probability of an individual being categorized into a specific group based on their behaviors, thus making targeted marketing more precise.

28-Article

TECHNIQUES TO PREVENT SQL INJECTION AND CROSS-SITE SCRIPTING

Khujakulov Toshtemir Abdikhafizovich, Gaipnazarov Rustam Takhiritdinovich, Majidova Yulduz Daniyarovna

The article discusses a method for preventing SQL injections and cross-site scripting using the Knuth-Morris-Pratt string matching algorithm. SQL injection and cross-site scripting remain a major threat to data-driven web applications. There are more and more cases where hackers gain unlimited access to the internal database of web applications in order to steal, edit and destroy confidential data. Therefore, measures must be taken to curb the growing threats of SQL injection and XSS attacks. This study presents a technique to detect and prevent these threats using the Knuth-Morris-Pratt string matching algorithm. The algorithm was used to match the user's input string with a stored injection string pattern to detect any malicious code. The implementation was carried out using the PHP scripting language and the Apache XAMPP server. The security level of the method was measured using various test cases of SQL injection, cross-site scripting (XSS), and code injection attacks. The results obtained showed that the proposed technique is able to successfully detect and prevent attacks, register an attack record in the database, block the system by its mac address, and also generate a warning message. Thus, the proposed technique turned out to be more effective in detecting and preventing SQL injections and XSS attacks.

29-Article

The model of creation of virtual museum geometric shapes divided into segments and the algorithm of geometric intersection of objects

Djumanov J. Kh., Xudayberganov T.R

The article tries to solve the problem of inter-section with a plane by constructing their ex-act geometric dimensions using a new geo-metric model based on the creation of virtual images of original museum exhibits based on engineering geometry and computer graphics. The algorithm of intersection of three-dimensional shapes is described.

30-Article

Model for transmitting URLLC packets in 5G networks

Amirsaidov Ulugbek

5G New Radio (NR) will support a new Ultra Reliable Low Latency Communications (URLLC) service designed to support small packets transmission with very strict latency and reliability requirements. Grant-free (GF) scheduling has been proposed for URLLC instead of the traditional high-latency grant-based (GB) scheduling adopted in 4G Long Term Evolution (LTE) networks. Although GF scheduling on shared resources provides lower latency, the ability to achieve the URLLC reliability requirement may be defaced due to the increased probability of collisions. In this paper, the optimal configuration of the modulation-code scheme and the number of retransmissions of Uplink Grant-Free URLLC packets are determined. Analytical expressions have been developed for calculating the service characteristics of URLLC traffic depending on the quality of the communication channel, modulation code scheme (MCS) parameters and the number of resource blocks required to transmit a packet after encoding it.

31-Article

Towards Identifying Interactions Between Human and Object Using Transformer-Based Models

Abdurashidova Kamola Turgunbaevna, Abdukhakimov Fayzulla Kudratulla ugli, Chorshanbiyeva Sevinch Akramovna

Understanding the subtle interactions between individuals and objects is critical for a variety of applications, including robotics, computer vision, and human-computer interaction. This paper looks at the research landscape for using transformer-based models to discover and understand these relationships. Existing approaches based on their methodologies, datasets used, and application fields have been explored thoroughly. The paper identifies several existing SOTA (state-of-the-art) methodologies that use transformer-based models to detect interactions among people, and objects as well.

32-Article

Computational Algorithms For Solving Boundary Value Problems Of Magnetoelastic Plates With Complex Form

Fakhriddin Nuraliyev, Muhammadbobur Mirzaakhmedov, Isaeva Nazim

This article presents a method of constructing a sequence of coordinate functions (solution structures) using the R-function method of V.L. Rvachev (Rvachev Function Method, RFM) to solve numerically the problem of the boundary of magnetoelastic plates of complex structural form in the process of thermoelectromagnetic-elastic deformation according to the given boundary conditions and constructing an algorithm with the joint use of the Newmark method, Bubnov-Galerkin variational and variational forms, Gaussian numerical methods

33-Article

FIRST-LEVEL KINSHIP IDENTIFICATION ALGORITHM

Zaynidinov X.N, Kuchkarov M.A, Yoqubov M.Gʻ, Tosheva D.M

In today's world, the identification of first-degree kinship is considered crucial in issues such as family, inheritance, marriage, and confirming paternity or maternity. Despite the fact that research has been conducted by several influential foreign institutions in this field, sufficient algorithms have not been developed for national software products in our country to address this issue. This article examines the issue of first-degree kinship from several perspectives: first, compiling a state registry based on the genome; second, situations requiring the identification of first-degree kinship; and third, the algorithm for the specific case of "Parent+Parent+Child" combination, which is characteristic of first-degree kinship. An algorithm has been developed based on this combination to identify first-degree kinship, and results have been obtained through its implementation. Based on these results, the probability of paternity (or maternity) has been calculated as a percentage indicator. The obtained results contribute to the development of the national genome database and enable the utilization of algorithms for identifying first-degree kinship.

34-Article

METHODS OF SPATIAL AUGMENTED REALITY TECHNOLOGY IN MOBILE APPLICATIONS

Artikova M.

The paper presents an overview of methods and technologies used to develop mobile applications using augmented reality (AR). The main focus is on spatial methods that allow for the effective integration of virtual objects into the user's real-world environment. The methods used in developing a mobile application based on marker-based and spatial augmented reality technologies are identified and analysed. In particular, the use of Simultaneous localisation and mapping (SLAM, Simultaneous localisation and mapping) method is justified, which being an integral part of the augmented reality experience, provides geometric positioning of the augmented reality system, extends the real environment by adding virtual objects based on the localisation and structure of the environment. Visual interaction with the surrounding world based on augmented reality technologies is considered, and current trends and prospects of AR technologies in mobile applications are also discussed.

35-Article

Design of Warehouse Management Information System of production enterprise

Javlon Gulyamov

The issue of managing reserves in the warehouse is considered urgent, and the provision of timely and sufficient production enterprise departments with them ensures the absence of production interruptions. The fact that stocks are collected in the warehouse more than necessary leads to the freezing of the enterprise's funds, unnecessary expenses. Therefore, it is advisable to digitize the management of reserves, use modern solutions. The article presents the issue of designing the information system of technological processes of the warehouse of the islab production enterprise, in which it is proposed to identify reserves by QR code when entering and leaving the warehouse. The article developed a project of Information System modules for accounting for the turnover of reserves in the warehouse.

36-Article

INFORMATION LOGIC MODEL AND ALGORITHM FOR DETERMINING THE H-INDEX OF NATIONAL SCIENTIFIC JOURNALS

Nazirova E. Sh, Erkinov Sh.Sh, Hajiyev S.A.

IDEF0-model of the process of submission and publication of an article by a user (author) to the uzjurnal.uz platform, the IDEF0-model of the process of registering and submitting an article from the platform, the IDEF0 model of the tasks of the main and separate magazine administrators from the platform is developed. developed. Using this model, the process from the arrival of an article to its publication on the uzzhurnal.uz platform is described. Journal users are divided into basic admin, journal admins, users, authors, editors and reviewers, and the performance tasks of each role are revealed using the above IDEF0 models.

37-Article

FEATURE SELECTION IN MULTI-OUTPUT CLASSIFICATION USING RECURSIVE FEATURE ELIMINATION: A CASE STUDY ON MEDICAL DATA

Sharipov D.K., Saidov A.D.

This study investigates the use of Recursive Feature Elimination (RFE) for feature selection in multi-output classification, particularly with a medical dataset. The classification task requires predicting several related binary outcomes simultaneously, making feature selection vital for reducing dimensionality and enhancing model performance. RFE is applied to each target variable, and common features across all targets are identified to refine the feature set while preserving prediction accuracy. A RandomForestClassifier, paired with a MultiOutputClassifier, is used to assess the predictive power of the selected features. The results indicate that RFE improves classification efficiency by removing redundant and irrelevant features, resulting in a model that is both more interpretable and computationally efficient. This method shows potential for optimizing multi-output medical classification tasks, providing advantages in healthcare data analysis.

38-Article

Algorithm and software for calculating the main indicators of а gas fields

Gulstan Artikbaeva

The article presents the structure of the modules of a software complex that provides a visual representation of numerical results obtained during computational experiments. It addresses the development of a software complex for monitoring and forecasting based on the visual presentation of numerical results obtained during computational experiments on the main filtration process parameters in the development of single- and double-layer gas fields. The focus is on designing software interfaces that ensure convenient interaction for users conducting experiments, as well as creating a software complex that visually presents numerical results in 3D graphics. The use of mathematical modeling methods, as well as modern computer technologies, in solving issues related to the design, management, and research of gas and gas condensate fields simplifies problem-solving and requires less time. This enables the acceleration of the design and research of gas and gas condensate fields.

39-Article

INTEGRAL METHODS OF SOLVING INVERSE PROBLEMS OF THE DYNAMICS

Sagatov Miraziz Vorisovich

Based on the analysis and systematization of the inverse problems of dynamics, the study of the properties and features of the types of dynamic models under consideration, an approach is proposed for the development of appropriate methods of mathematical modeling based on the use and implementation of integral models in the form of Volterra equations of the I and II kind, their functional capabilities are determined in the study of various classes of problems , and also formulated the features that affect the choice of methods for their numerical solution. Methods for obtaining integral models are proposed, which are the basis for constructing algorithms for solving inverse problems of dynamics for a fairly wide class of dynamic objects. Integral methods for the identification of dynamic objects have been developed, which make it possible to obtain stable non-optimization algorithms for calculating the parameters of mathematical models. Recurrent methods of parametric identification of transfer functions of dynamic objects with an arbitrary input action are proposed (the obtained parameters of the transfer functions are also coefficients of the corresponding differential equations, which makes it possible to obtain equivalent mathematical models in the form of integral equations).

40-Article

MODEL OF A MULTI-AGENT SYSTEM FOR MONITORING AERONAUTICAL INFORMATION TELECOMMUNICATION NETWORK

Eshmuradov Dilshod, Turaeva Nasiba, Saidova Gulchehra, Muradova Alevtina

The purpose of the work in this article is to develop models and algorithms to ensure the effectiveness of monitoring the telecommunications network of information transmission in air navigation systems, allowing to systematize and improve the efficiency of the work of employees of the Department of the data transmission network of the aviation industry of Uzbekistan. The relevance of the development of this model is to use intelligent agents to help operators quickly enough to recognize a potential problem and help prevent a wide range of real failures. The development of a model of a network monitoring system based on multi-agent technology was completed and the function of each agent in the system was described. During the development, the formation and stages of operation of the multi-agent technology were studied. The practical results of the study of the work are that the proposed software and hardware for monitoring the transmission of aeronautical information in the airport telecommunications networks were successfully implemented and the efficiency of the network has increased significantly.

41-Article

INSTRUMENTAL TOOLS OF THE PLATFORM FOR DETERMINING THE H INDEX OF SCIENTIFIC JOURNALS

Elmira Nazirova, Shokhrukh Erkinov, Hajiyev Sukhrob

The h-index has emerged as a pivotal metric for evaluating the scientific impact and productivity of journals, authors, and institutions. In the context of scientific journals, the h-index is instrumental in determining their influence and standing within the research community. This paper explores the critical tools and platforms available for calculating the h-index of scientific journals, emphasizing their methodology, accessibility, and accuracy. Prominent tools, including Web of Science, Scopus, and Google Scholar, are analyzed for their strengths and limitations in h-index determination. We highlight the role of citation databases, their indexing algorithms, and coverage of disciplines in influencing the h-index calculation. The study also examines open-access platforms and emerging AI-powered tools, which aim to democratize access to bibliometric analysis. Through a comparative analysis, the paper identifies key considerations for selecting the appropriate platform based on journal scope, citation frequency, and data granularity. Furthermore, challenges such as data incompleteness, citation biases, and variations in calculation methodologies are discussed. By synthesizing insights on available tools and their applications, this paper provides a comprehensive guide for researchers, editors, and policymakers to assess and leverage the h-index effectively in the realm of academic publishing.

42-Article

Issues of developing services for mobile communication companies in the trend of m-government

Aripova Dilafruz

This article examines measures to enhance the service life of devices by implementing a comprehensive preventive maintenance plan to ensure reliability and usability of mobile communication infrastructure in the "M-Government" environment of mobile communication companies. The aim is to provide high-quality services to users while increasing and supporting the customer base of service providers. This is achieved by studying customer relationships based on an analysis of customer satisfaction, retention, loyalty, acquisition, and attrition criteria. The importance of providing analytical information about each of the key indicators influencing customer relationships to the management of mobile companies is also highlighted

43-Article

A Bidirectional Notation of Korean and Uzbek Alphabet

Jeongyong Byun, Yakubov Maqsadxon, Nodir Egamberdiyev, Yuldashov Raxmon, Naim Nodira, Bekmuxammedov Bunyodbek

This article explores the challenges and solutions in developing a bidirectional system for Uzbek and Korean notation to accommodate the growing trade and administrative interactions between Uzbekistan and Korea. It highlights the limitations of the 24 Korean alphabets in representing Uzbek sounds accurately and proposes a solution using Hunminjeongeum Haerye and Unicode Hangul Jamo. The study analyzes Uzbek phonetic characteristics and their relation to Hangul, introducing eight consonant and four vowel rules for improved representation of Uzbek sounds in Korean. The research evaluates the effectiveness of these rules by applying them to Uzbek place names, personal names, and product names written in Hangul. It concludes that the proposed method enhances phonetic accuracy and identifies areas for further improvement, aiming for a notation system comparable to the International Phonetic Alphabet (IPA).

44-Article

The importance of student image recognition programs in education management

Azimov Bunyod, Ahmadaliyev Boburjon

Currently, the use of artificial intelligence systems to prevent human errors is developing day by day. The automated method of determining attendance by face recognition using intelligent systems is of great importance. I implemented this using the Python program. In addition, I tested the method of using the wavelet method to increase the accuracy and reliability of this system in test control. I explained the software sequence (algorithm) of implementing this process and my thoughts on using the necessary libraries.

45-Article

The role of artificial intelligence in digital transformation

Fayzullayev Shakhzod, Yuldashov Rahmon

Artificial Intelligence (AI) has emerged as a transformative force driving digital transformation across industries. This article examines the pivotal role of AI in reshaping business processes, enhancing operational efficiency and fostering innovation in the digital economy. The study focuses on how AI technologies such as machine learning, natural language processing, and computer vision contribute to automating decision-making processes, improving customer experiences and enabling data-driven strategies. Using a mixed-methods approach, this research investigates the adoption trends, challenges, and opportunities associated with AI implementation in organizations. Key findings highlight that businesses leveraging AI gain competitive advantages by streamlining operations and identifying new revenue streams. However, the study also underscores barriers such as ethical concerns, data privacy issues and the skills gap that hinder widespread AI integration.

46-Article

HUMAN OBJECT DETECTION FROM IMAGE USING COMPUTER VISION TECHNOLOGIES

Khalilov Sirojiddin, Khamdamov Utkir, Omonullayeva Farangiza

Currently, identifying human objects from video images around the world is somewhat difficult for scientific researchers engaged in computer vision. More effective methods for identifying a car, phone, cup and other objects in images are proposed. It is important to identify human objects from the images obtained by surveillance cameras in public places. Computer programs perform this process from scratch, meaning that the system is fed various images as input and after comparisons with a pre-stored database, the system identifies the person. In this paper, an approach for identifying different types of human faces through image processing has been developed. The Canny operator was used to determine the shapes of the objects in the input image.

47-Article

Water quality classification using KNN algorithm and real-time monitoring

Khudoyorkhan Jamolov, Djumanov Jamoljon, Rajabov Farkhat

Water quality refers to the extent to which water satisfies specific standards and requirements for various uses, including drinking, agriculture, industrial processes, and environmental conservation. Water quality parameters include various pH, TDS, temperature, and water level parameters that can affect water properties and safety such as water quality. To classify current and future water quality, machine learning-based classification models such as the algorithm, were calibrated using historical data and tested on independent datasets to evaluate classification accuracy. The result of this study is to use 4 attributes which are water quality parameters.

48-Article

Enhancing Land Cover Change Detection through Advanced Algorithms and Multi-Sensor Data Integration

Kuchkorov Temurbek, Saydullayev Humoyun, Khamzaev Jamshid

This thesis evaluates various change detection algorithms for land cover analysis using remote sensing data, aiming to enhance accuracy and efficiency in environmental monitoring. It reviews algebra-based, classification-based, and advanced statistical techniques, highlighting their applications, strengths, and limitations. The study emphasizes the integration of machine learning and multi-sensor data fusion to address current challenges in the field, advocating for hybrid models that improve detection capabilities across diverse landscapes.

49-Article

The role of machine learning algorithms in image classification

Raximov Abdug`ofur

Image classification is one of the most important and widely used areas of artificial intelligence (AI) and machine learning (ML) technologies today. The importance of analyzing and using images in various applications such as medicine, automotive, security, social media and entertainment is increasing. The image classification process usually consists of dividing the data in the images into specific categories or classes, and machine learning algorithms play an important role in this process. These algorithms allow you to extract complete data from images, analyze them and divide them into the correct classes.

50-Article

History Of Development and Basic Concepts Of Quantum Computing Theory

Quantum computing has emerged as a groundbreaking field driven by the need to address intricate challenges in physics, mathematics, and algorithmic theory. The interplay between quantum mechanics, computational theory, and algorithmic innovation has catalyzed the development of this domain. Initiated by R. Feynman`s vision in 1982, which highlighted the limitations of classical systems in simulating quantum phenomena, quantum computing has paved the way for a new era of computational efficiency. Subsequent advancements, such as D. Deutsch`s quantum Turing machine and groundbreaking algorithms by P. Shor and L. Grover, have established the unparalleled capabilities of quantum systems.This study explores the fundamental concepts of quantum computing, such as qubits, quantum registers, and quantum operations, emphasizing their practical applications in cryptography, materials science, optimization, and artificial intelligence. Challenges such as noise resilience and error correction are also discussed, highlighting ongoing efforts to create stable and reliable quantum systems. The research underscores the transformative potential of quantum technologies in advancing computation, securing data, and innovating across various scientific and industrial domains.

51-Article

Object Detection: AI Algorithms, Methods, and Results

Saydazimov Javlonbek,Atadjanova Nozima

Object detection is a fundamental task in computer vision, combining image classification and localization to detect multiple objects in a scene. With advances in artificial intelligence (AI), object detection algorithms have transitioned from traditional handcrafted techniques to sophisticated deep learning methods. This paper explores various object detection algorithms, methodologies, and their performance on benchmark datasets, providing insights into their strengths, limitations, and future potential.

52-Article

Adaptive hybrid filtering algorithms with an artificial neural network as part of an adaptive filter

Sevinov Jasur,Zaripova Shahlo

— The paper presents algorithms for adaptive hybrid filtering with an artificial neural network as part of an adaptive filter. This approach is one of the options for the practical application of adaptive hybrid filtering methods. Adaptive hybrid filtering algorithms with an artificial neural network as part of an adaptive filter allow you to train a neural network and adjust the weighting coefficients for filtering procedures to the statistical characteristics of the measured and observed indicators.

53-Article

Using artificial intelligence technologies in higher education management

Turg‘unov Abrorjon, Aliev Oybek, Nosirova Namunabonu, Marks Matyakubov

The rapid evolution of artificial intelligence (AI) technologies presents transformative opportunities in higher education management. This study explores how AI can enhance decision-making, streamline administrative processes, and improve resource allocation in universities and colleges. Key applications include predictive analytics for enrollment management, intelligent systems for personalized student support, and AI-driven tools for optimizing faculty workload distribution. The research also highlights the potential of AI to improve operational efficiency by automating routine tasks, such as scheduling, grading, and communication. Ethical considerations, including data privacy and equity, are critically examined to ensure responsible implementation. By integrating AI technologies strategically, higher education institutions can improve their adaptability, efficiency, and capacity to meet diverse stakeholder needs in a rapidly changing educational landscape. This paper provides a framework for leveraging AI effectively, along with case studies demonstrating its impact on institutional performance and student outcomes.

54-Article

Different of A Study based Face Recognition Algorithms

Uralova Iroda, Boboqulov Abbos

In recent years, the biometrics has achieved a great attention on a world level. A Biometric System operates by getting biometric information from a personal that extracts a feature set from the data which is acquired, and helps in comparing this feature set against the template stored in the database. There are biometric technologies which could either be physiological or behavioral. Face Recognition is having the importance to provide biometric authentication with easy image acquisition that can be used for online and offline applications. There are number of existing approaches for biometric facial recognition and classification. This paper gives a review on some of the common and reliable approaches which include PCA, LDA, SVM, SIFT, SURF, etc. Face recognition is one of the most exciting and prominent applications of deep learning applied in the domain of visual computing. Not only can it showcase the power of deep learning but also it can highlight some of the concerns of using a methodology whereby a black-box approach is utilized for solving problems whose solutions have real-life consequences.

55-Article

Statistical analysis of big data in databases using fractal theory

Anarova Shahzoda, Qarshiyeva Soliha, Ismoilova Maftuna

This paper reviews the integration of fractal theory in improving statistical analysis within large-scale databases. It points out the increasing complexity of modern datasets, which are usually characterized by an irregular pattern and self-similarity across different scales. The authors introduce fractal theory as a strong mathematical method to capture these properties, thus allowing for more precise analysis and modeling of nonlinear, high-dimensional data structures. The article points out the superiority of the fractal-based approach in comparison with the traditional statistical method when sparse, incomplete, or unstructured data arises. An example shows fractal dimensions improve the accuracy of multidimensional feature extraction in a database, enabling proper decision-making related to finance, healthcare, or logistic areas. The paper is concluded with remarks on how fractal-based methodologies promise to provide the solution needed by these increasing exponential data records and advocate that more detailed research in these techniques will be carried out for broader applications. This work would be beneficial to researchers in the field of data science, database management, and applied mathematics.

56-Article

Models and Algorithms to Improve the Efficiency of Information Exchange Processes in Digital Transformation

Egamberdiev Nodir, Sodiqov Vali, Yadgorov Feruz,

this paper explores two models aimed at enhancing the efficiency of information exchange in digital transformation: Data Routing Optimization and Real-Time Data Synchronization. Model 1 optimizes data transfer speed by reducing routing delays, achieving a 5% improvement in transmission time. Model 2 focuses on ensuring real-time data consistency across distributed systems, reducing synchronization delay by 4%. Both models are evaluated based on system resource utilization, scalability, and real-time performance. The findings demonstrate significant improvements in network efficiency, offering valuable insights for optimizing information exchange in digital transformation processes.

57-Article

Development and Application of IoT-Based Environmental Monitoring Systems for Enhanced Ecological Management and Sustainability

Kuchkorov Temurbek, Yusupov Agzamjon, Khamzaev Jamshid

This research is dedicated to the development of a new smart environmental monitoring system in the Republic of Uzbekistan. This system provides rapid and accurate responses to changing environmental conditions by monitoring the quality of atmospheric air, water resources and soil composition. The system enables real-time observation and deep analysis of various ecological changes and weather conditions, allowing for a more accurate assessment of environmental status and the anticipation of ecological risks. Additionally, the application of this system in urban and rural areas plays a crucial role in preventing various ecological problems and managing them effectively, which assists in the planning and implementation of environmental protection measures more efficiently.

58-Article

Analysis of load balancing algorithm in information systems

Rakhimov Mekhriddin, Ergashev Shakhboz, Karimberdiyev Jakhongir

In this article will show comparative analysis of common load balancing algorithms. The primary purpose of load balancing analysis is to ensure that the distribution of network traffic or computing workload across multiple servers or resources is efficient, fair, and resilient. Load balancing is crucial in maintaining high availability, performance, and reliability in systems such as web applications, cloud services, and data centers. Here’s a detailed look at the key purposes of analyzing load balancing strategies. Analyzing load balancing algorithms involves evaluating their performance based on various metrics and considerations. There are several established load balancing algorithms, each with unique characteristics and suitable use cases. In this article analaysed some of the most commonly used ones.

59-Article

Development of computer modeling of the process of harmful substances dispersion in the atmosphere

Sharipov Daler, Xurramova, Ra`no

One of the actual challenges in environmental modeling, which garners quite big scientific interest, is the development of mathematical models of active aerosols propagation in the atmosphere. This paper deals with a computational model for estimating, forecasting, and visualizing the spatial and temporal evolution of active aerosol emissions propagation in the atmospheric boundary layer. The study employs numerical methods for solving gyrodynamic problems. The developed model consist the govering semi-empirical equation of air pollutants atmospheric dispersion as well as initial and mixed boundary conditions. In order to solve the stated problem there was used finite differences method and tridiagonal matrix algorithm. The problem was approximated using an implicit difference scheme. Based on the proposed computer model there were performed calculations to predict air pollution concentration over industrial regions of Uzbekistan and to optimize the placement of new production facilities, considering local specific environment conditions with the aim of minimizing the potential anthropogenic impact.

60-Article

General formulation of the optimization problem for the elements of elastic structures

Yakubov Sobir, Daminova Barno

This paper considers the classical formulation of the problem of optimal design of engineering structures. The results of solving optimal design problems determine the shapes, internal properties and operating conditions of structures that deliver the extremum of the selected characteristics of structures, taking into account additional restrictions.

61-Article

A Cyber Threat Detection Model Based on the Intellectual Integration of IoC, Log and Dark Web Monitoring Data in Cyber Intelligence Operations

Ganiyev Adukhalil, Tojimatov Dostonbek

In today`s era, as the complexity and scale of cyber threats continue to grow, traditional threat detection methods are often insufficient to ensure efficiency in real-time operations. This study proposes a threat detection model that integrates IoC (Indicators of Compromise), log files, and Dark Web data to enhance cyber intelligence operations. The model includes three main stages: data integration, threat analysis, and ensuring real-time visualization and alerting.

62-Article

ALGORITHM FOR DIGITAL PROCESSING OF SEISMIC SIGNALS IN DISTRIBUTED SYSTEMS

Kuchkarov Muslimjon

The problem of misallocation of cache lines in processors is studied based on the specific features of memory organization in multi-core and multi-processor systems of distributed systems. Incorrect allocation of cache lines leads to sequential execution of tasks that should be executed in parallel on several cores. It is difficult to determine the problem, and its presence can cause a sharp decrease in the performance of the entire system. In many seismic studies, scientists` efforts are aimed at finding reliable signs of seismic hazard. Abrupt changes in one or another parameter are called abrupt effects, emissions or anomalies, through which the principle of prediction can be implemented - this means predicting the place, strength and time of a future seismic event. The use of distributed systems in the implementation of predictions with the help of machine learning, digital processing of large volumes of digital signals, ensures fast and high-quality determination of results. To increase the speed, it is necessary to develop parallel algorithms. Synchronization of common parallel streams is necessary to solve parallel computing problems and to ignore cache memory in parallel programming.

63-Article

MYSQL-BASED MODELING AND H-INDEX COMPUTATION FOR ENHANCING BIBLIOMETRIC EVALUATION ON THE UZJURNAL.UZ PLATFORM

Nazirova E.Sh, Erkinov S.S, Khajiyev S.A.

This study presents a database-driven, algorithmic approach to evaluating the scientific impact of national journals hosted on the UzJurnal.uz platform. A relational database architecture was developed using MySQL to store and process structured bibliometric data including article metadata, citation records, authorship, and editorial workflows. IDEF1X modeling techniques were applied to visualize the relationships among users, articles, and administrative roles, enabling scalable and traceable publication management. A dedicated H-index computation algorithm was implemented to automate impact analysis, based on citation frequency and journal-level output. The system demonstrates compatibility with diverse server environments and offers extensibility for integrating additional metrics such as the G-index and i10-index. Results show that the platform can provide accurate and transparent bibliometric evaluations, fostering data-informed decision-making in national research ecosystems.

64-Article

METHODS OF BUILDING RECOMMENDATION SYSTEMS IN E-COMMERCE

Narziev Nosir Bakhshilloevich , Olimova Mukhlisa Vohidjon qizi

The purpose of this article is to study the main approaches and algorithms for building recommendation services in online trading systems. Recommendation systems optimize the sales process by identifying interesting and relevant products or services for users and offering them personalized recommendations, and serve to increase sales in e-commerce systems. In this article, the capabilities of the main techniques of recommendation systems, such as collaborative filtering, content-based approaches and hybrid models have been analyzed. A comparative analysis of the effectiveness of these methods and algorithms is carried out, their advantages and disadvantages are identified. The paper concludes by suggesting theoretical and practical frameworks for future advancements in recommender systems.

65-Article

Theoretical concept of the concept of video and video

Maftuna Bekmirzayeva

As time progresses, the demand for digital data is increasing. As a graphic engineer, our task is to make a proposal that meets these requirements. We can provide information and data in various forms. The most optimal type is video. This article covers the concept, features, standards, and video quality of video. This article explores the fundamental concepts of video, its key features, industry standards, and factors affecting video quality, such as resolution, compression, and frame rate. Additionally, we discuss various video formats, their advantages, and their applications in different fields, including streaming, broadcasting, and digital marketing. The rapid advancement of video processing technologies, such as artificial intelligence (AI)-powered enhancements, real-time rendering, and high-efficiency codecs, further expands the possibilities of video usage. By understanding these aspects, we can optimize video content to meet diverse needs across different platforms and applications while ensuring high quality and efficiency in data transmission.

66-Article

Information support for solving functional issues of automated management of corporate information exchange (on the field of tax authorities)

Askaraliev Odilbek

The article studies the issues of information support for solving functional issues of automated management of corporate information exchange (on the example of tax authorities), the design of its control structure and its use in the management system. The importance of using the accumulated experience to support decision-making in the organizational management system is emphasized. In addition, relevant recommendations are developed for the implementation of intelligent support for decision-making in integrated management systems, as well as service technologies. For the efficient operation of information exchange in the automated information management system, information about the state of the managed object, the external environment and the received control effects must be correctly presented. The operation of the Unified Computer System (UCS) for Data Processing in Tax Authorities requires conditions such as the optimal volume of information coming from various control units, the distribution of information flows by time and area.

67-Article

Organizational and information support of NоSQL data structures in integrated information systems

Askaraliyev O.U., Normatova M.N.

This article presents an analysis of the organizational and information support of the NoSQL data structure in integrated information systems. For experimental studies, an architecture was developed based on virtual machines simulating the operation of a web service for data collection with a classic three-tier architecture. It was proved that differences in the structure of an unstructured database can affect the reliability and efficiency of the entire information system. The article proposes a methodology for assessing the effectiveness of the data structure of corporate information systems using the MongoDB database management system (DBMS), a methodology for information and organizational support of the database.

68-Article

Analysis of accuracy of local spline functions using analytic functions

Abduganiev Mukhriddin

After many scientific studies in the world, we can say that the use of spline functions in working with digital signals is widely used and good results are achieved. Nowadays, there are many ways to wish digital signals. Examples of them include Lagrange and Newton interpolation methods, spline function, Fourier series and exponential function approximation methods, orthogonal methods for solving systems of linear equations, discrete wavelet transforms and a number of other interpolation methods. During the research work, cubic Bezier spline function and interpolation Natural spline function were selected for equal intervals. This article describes in detail the process of interpolation of functions to approximate local cubic spline functions to given analytic functions.

69-Article

Numerical Simulation of Non-Stationary Heat Conduction in Axisymmetric Bodies

Polatov Askhad, Sapayev Shikhnazar, Matyakubov Marks

The distribution of the temperature field of elements of three-dimensional structures is examined in the article and the influence of non-homogeneities on the temperature field in three-dimensional axisymmetric structures is studied. Based on the variational formulation of the problem, a finite element model for solving axisymmetric problems of heat distribution in three-dimensional bodies was built based on the finite element method; a computational algorithm of the solution and the software were implemented. Shape functions corresponding to linear triangular elements are used in the finite element solution to the problem. They ensure its continuity at the boundary with neighboring elements since this distribution is linear along any side of the triangle and with the same change in value at the nodes, the same changes will occur along the entire inner boundary. A discrete finite element mesh model is a set of parameters that consist of the total number of finite elements and mesh nodes, arrays of their node coordinates, and node numbers of finite elements. To verify the reliability of the results obtained, a computational experiment was conducted, related to the study of the effect of the increase in the number of finite elements on the convergence of solutions. Analysis of the experimental results confirms the numerical convergence of temperature values as the finite element mesh is refined. At the initial stage of the study, the influence of temperature distribution for homogeneous structural elements was studied.

70-Article

Numerical Modeling Of Boundary Value Problems For Parabolic Differential Equations

Nematov Abdugani, Sadikov Rustam, Bakhriddinov Akbar.

This article examines numerical methods for solving boundary value problems for parabolic differential equations, which are widely used in modeling physical processes such as the filtration of oil, gas, and water in porous media, as well as heat transfer processes. The primary focus is on the differential-difference method, which is characterized by its high accuracy and ease of programming. The theoretical aspects of the method are presented, including algorithmic implementation, the use of the differential sweep method, and an analysis of the stability of the computational process. Special attention is given to the advantages of the differential-difference approach, such as the ability to automatically fulfill internal conditions during phase transitions, ensure absolute computational stability, and simplify programming. The article also describes the solution algorithm, which includes a longitudinal-transverse scheme that enables efficient problem-solving using parallel computations. Additionally, numerical calculation results are provided, including comparisons with analytical solutions that demonstrate the method's high accuracy. It is shown that the developed algorithms can be easily adapted for solving other classes of problems, making them a universal tool for numerical modeling in computational mathematics. The conclusions are supported by graphical results and error analysis, confirming the reliability of the proposed approach.

71-Article

MODELING BUBBLE RECOVERY PROCESSES IN A VIDEO IMAGEHalimov Shahzod

Halimov Shahzod

- This study investigates the gas motion through the froth of a prototype bi-dimensional flotation cell. Using advanced video and image analysis techniques, surface velocity data obtained with a Visiofroth system were validated by comparing them with manual tracking results of bubble motion on the froth surface. Images for manual tracking were captured using a high-speed video camera. Additionally, discharge velocity profiles were analyzed as a function of froth height over the launder lip under different froth Crowder angles and superficial air rates. An increasing froth discharge velocity from the lip level to the top of the froth was observed. Air recovery was estimated using the velocity profile and compared with values derived from surface velocity estimation. Results indicated differences from 24% to 47% in air recovery when froth flow area at the lip was considered. However, both approaches displayed similar trends under varying JG and froth Crowder angles. Using top-of-froth velocity alongside the froth flow area at the Visiofroth location resulted in air recovery values closer to those obtained from the discharge velocity profile at the lip.

72-Article

Construction Of Higher-Order Spline Functions, Digital Processing And Modeling Of Signals, And Their Comparative Analysis

Zaynidinov Hakimjon, Bakhramov Sayfiddin, Kobilov Sirojiddin, Kuchkarov Muslimjon

In this work, numerical processing of signals using high-order spline models in the direction of the approximation theory of mathematical modeling, which plays an important role in the development of science, is considered one of the important and urgent issues, therefore, third-order, fifth-order, seventh-order Ryabenky spline models selected. We can also say that high order Ryabenky spline models are effective and reliable mathematical models in digital signal processing. As an example, the functions obtained as an experiment and the initial values of the gastroenterological signal were obtained. Approximation of gastroenterological signals and functions, numerical processing of signals by obtained values, and results of comparative analysis are presented. From the results of the comparative analysis, it was analyzed that the results obtained in this study in digital processing of signals are better approximated by the seventh-order Ryabenky spline model among the results of digital processing of signals based on third-order, fifth-order, and seventh-order Ryabenky spline functions. The higher-order Ryabenky spline functions discussed in this paper can also be used in digital processing and processing of various medical and other signals.

73-Article

MODELING GEOFILTRATION PROCESSES IN IRRIGATED CULTURAL AREAS AND SIMULATING THE WATER ABSORPTION PROCESS

Haknazarova Dilobar¸ Tadjikhodjaeva Elvira, Akhmedova Iroda, Murodullaev Bakhtiyor

In this work, a detailed analysis of the physical-geological and melioration features of the territory was carried out when modeling the geofiltration process on irrigated sown areas. Taking into account the above-ground and underground influencing factors in modeling, based on existing information models and scientific literature, a new mathematical model is proposed. In this process, special attention was paid to the phenomenon of mass transfer, as well as the components of velocity in the direction of the liquid flow, surface tension, and molecular diffusion processes were introduced, and a mathematical model was formed. Based on the proposed mathematical model, effective computational algorithms have been developed, and a software package for predicting groundwater levels has been created using them. This software package is aimed at accurate modeling and forecasting of geofiltration processes and includes a number of complementary software tools. Computational experiments were conducted, and the results of the experiments were analyzed. These results are aimed at a deeper study of geofiltration processes and ensuring optimal regulation of groundwater levels during irrigation. The data obtained through computational experiments carried out using the software package confirmed the effectiveness of the process and showed that this process can be used to improve melioration conditions and predict groundwater levels.

74-Article

Defuzzification of Intuitionistic Fuzzy Sets and its Application

Charos Khidirova, Alevtina Muradova

This paper presents a comprehensive study of defuzzification methods designed specifically for intuitionistic fuzzy sets (IFS), which extend traditional fuzzy sets by incorporating an additional dimension of uncertainty – hesitation. The research analyzes and compares three main classes of defuzzification techniques: centroid-based, score function-based, and hesitation-driven methods. Each approach is evaluated in terms of its ability to transform fuzzy data, described by membership, non-membership, and hesitation degrees, into meaningful crisp outputs that support more accurate decision-making. A significant contribution of this study is the application of the centroid-based defuzzification method in the medical domain, specifically for cardiovascular disease prediction. The paper outlines a framework where patient data is modeled using IFS and converted into a single risk score, enhancing clinical decision-making by quantifying uncertainty in diagnostic parameters. The results demonstrate that incorporating hesitation into defuzzification processes can yield more nuanced, reliable outcomes compared to traditional methods. The findings underscore the importance of selecting appropriate defuzzification strategies depending on the decision context and highlight the potential for applying IFS-based models in other complex and uncertain environments. The paper concludes with suggestions for future research, including hybrid techniques and broader applications in decision-support systems.

75-Article

CREATING OPEN-CORRES PARALLEL CORPUS BASES FOR MACHINE TRANSLATION

Nazirova Elmira, Abidova Shakhnoza, Uzakova Mamura, Usmanova Kamola

This study focuses on the development of an open-source parallel corpus database for machine translation among the Uzbek, Turkish, and Karakalpak languages, which all belong to the Turkic language family. The paper presents the linguistic and technological aspects of creating an open-platform translator, incorporating lexical and terminological databases, and highlights the importance of a constantly evolving terminological base for the Uzbek language. Through the use of linguo-mathematical models, the research analyzes complex word formation in agglutinative languages and proposes a functional software interface that supports accurate translation based on morphological structure. The translation model includes modules for corpus selection, data cleaning, lexical resource integration, and machine learning. Practical implementation and testing of the "Parallel Translator" program demonstrate its ability to analyze and translate complex word forms effectively, contributing to the advancement of computational linguistics and machine translation for underrepresented languages.

76-Article

Algorithm and program for solving the gas filtration problem in dynamically interconnected porous media

Normakhmad Ravshanov, Аbdugani Nematov, Shikhnazar Ismailov, Mohiniso Mahmudova

This article examines the mathematical modeling of the thermal impact process on an oil reservoir when water vapor is pumped into it. The problem of heat propagation in a porous medium is being investigated, taking into account heat losses through the roof and bottom of the layer, which leads to a more realistic description of temperature fields. The model is based on a system of differential equations with phase transitions and takes into account the thermal anisotropy of rocks. Numerical methods using finite-difference schemes and the distillation method were used to solve the Stefan problem. A computational experiment was conducted, the results of which showed a significant influence of heat loss on the temperature field and the efficiency of oil displacement. The obtained data were compared with previously developed models, key differences were identified, and refined calculation approaches were proposed. The work has applied significance for increasing the oil recovery of highly viscous oil reservoirs.

77-Article

MATHEMATICAL MODEL FOR EVALUATION OF TRILINGUAL ELECTRONIC TRANSLATION

Nazirova Elmira, Abidova Shakhnoza, Uzakova Mamura, Boymurodov Farrukh

The paper considers the issue of improving the mathematical model for a hybrid method of machine translation evaluation. Today, electronic translation remains a popular tool. The development of technologies and adaptation to linguistic and cultural features makes it indispensable in education, business and everyday life. The scientific article is devoted to the study of improving the model for evaluating electronic translation. The authors paid special attention to the analysis of existing methods for evaluating electronic translation, such as BLEU, METEOR, ROUGE was improved on the basis of which a new mixed model for evaluating electronic translation. An exact, also given example of calculating the assessment with an incoming text in the Uzbek language, which is translated into the Kyrgyz and Karakalpak languages. As a result, assessments were obtained using the METEOR, ROUGE methods. In addition, digital results of evaluating electronic translation based on the hybrid method were obtained. An analysis of the comparison of the translation platforms GOOGLE Translate, PROMT, Yandex and DEEPL after the implementation of the above methods is given.

78-Article

THE POTENTIAL OF VIRTUAL AND AUGMENTED REALITY TECHNOLOGIES FOR THE PROMOTION OF LITERARY TOURISM IN UZBEKISTAN

Artikova Muazzam, Eldor Sayfiyev, Fozil Jo‘raboyev

The article explores the potential of virtual reality (VR) and augmented reality (AR) technologies in the context of developing literary tourism in Uzbekistan. It presents a project aimed at creating a digital resource that comprehensively reflects the country's literary heritage through the development of a multilingual electronic platform and a mobile application based on VR and AR technologies. The study examines both conceptual and practical aspects of the digitalization of literary heritage, focusing on the creation of the "Literary Guide" electronic platform. The paper outlines technological solutions, including the implementation of virtual tours and the integration of marker-based and spatial AR technologies, and identifies key literary landmarks—specifically, the house-museums of prominent writers in Tashkent and Samarkand—covered by the project. The importance of innovative digital tools is emphasized for the promotion of cultural assets, enhancement of tourist experiences, and stimulation of post-pandemic development in the tourism sector. The results presented may serve as a foundation for further research and the adoption of similar initiatives in the field of cultural tourism.

79-Article

MODELING THE DEVELOPMENT OF LEMMATIZATION ALGORITHMS FOR THE UZBEK LANGUAGE

Maksud S. Sharipov, Ogabek O. Sobirov

Lemmatization is a vital task in Natural Language Processing (NLP), essential for various applications. However, the Uzbek language, categorized as a low-resource language in NLP, lacks dedicated lemmatization systems. This deficiency hinders the development of advanced NLP tools for Uzbek texts. To address this, we present a structured approach using IDEF0 and IDEF1X models to design an Uzbek lemmatization system. The IDEF0 model details the system's functionality, outlining the lemmatization process, while the IDEF1X model illustrates the relationships between data components. Furthermore, we developed a rule-based system that leverages Uzbek morphology to accurately determine word lemmas. This research contributes a well-defined architectural framework and a practical rule-based solution for lemmatization, aiding in the advancement of NLP for Uzbek and similar low-resource, morphologically rich languages.

80-Article

Algorithm for Human Action Recognition Based on Images and Visual Data Using OpenPose

Fayzullayeva Z.I., Pirimqulova Z.A., Hojiyev S.N.

Human action recognition is a fundamental task in the field of computer vision and has become increasingly important in applications such as human-computer interaction, intelligent surveillance systems, virtual and augmented reality, and smart transportation. With the rapid advancement of deep learning techniques and cutting-edge algorithms, the effectiveness and accuracy of action recognition systems have significantly improved in recent years. In this study, we propose a real-time human action prediction model based on skeletal keypoints extracted from static RGB images using the OpenPose framework. The model processes spatial configurations of human joints to identify and classify actions with high precision. By leveraging Part Affinity Fields (PAFs) and confidence maps, our system successfully estimates human poses and tracks body movements efficiently. Experimental results demonstrate that the proposed approach outperforms several traditional methods, including DeeperCut (58.1%) and Convolutional Pose Machines (CPM) (55.2%), achieving a Percentage of Correct Keypoints (PCK) of 85.2%. This high accuracy indicates the model’s robustness and its potential for real-world applications that require efficient and accurate motion analysis in real-time scenarios. The results highlight the advantages of combining deep learning with skeletal keypoint estimation for advanced and scalable human action recognition systems.

81-Article

Hybrid Adaptive Ensemble with Dynamic Feature Selection for DDoS Attack Detection

Musaev Mukhammadjon, Rakhmatov Furkat

Distributed Denial of Service attacks continue to represent a significant challenge for intrusion detection systems due to their scale, diversity, and evolving nature. This paper introduces a novel hybrid adaptive ensemble with dynamic feature selection, which integrates three distinct components: Light Gradient Boosting Machine, one-dimensional convolutional neural network, and Isolation Forest. The proposed ensemble employs Shapley Additive Explanations to perform adaptive feature selection, which enhances classification performance while reducing computational complexity. Experimental evaluation using the CICIDS two thousand seventeen dataset demonstrates that the proposed model achieves an accuracy of ninety-nine point nine five percent, an F one score of ninety-nine point nine five percent, and a recall of ninety-nine point nine two percent. Comparative analysis with existing methods confirms that the proposed approach offers superior effectiveness, robustness, and scalability in detecting distributed denial of service attacks.

82-Article

Didactic potential of Artificial Intelligence in the context of digital transformation

Jahongir Modullayev

This article examines the didactic potential of integrating artificial intelligence (AI) technologies into the educational process. Today, advances in audio, video, telecommunications, and information technologies—when combined with AI—can significantly enhance teaching quality and effectiveness. First, we analyze the core concept of “didactics,” derived from the Greek term didaktós (“to teach”), as the theoretical framework for organizing instruction in a manner that maximizes learning outcomes. Next, we explore cutting-edge approaches for tailoring each student’s individual learning path based on data-driven insights, thereby optimizing instructional efficiency. Drawing on a comprehensive review of scholarly sources, this paper highlights the promise of AI-enabled tools and methods for delivering high-quality education and outlines their strategic value for future educators.

83-Article

3D FRACTAL MODELING GEOMETRY AND ALGORITHM IN INTERNAL DISEASE DETECTION

T.A. Khojakulov, A.A.Bahromov, R.T.Gaipnazarov

This article studies the use of modeling and algorithms based on 3D fractal geometry for early detection of internal diseases. Fractal models provide an opportunity to clarify the diagnostic process through accurate mathematical modeling of complex structures of the human body. The study developed a 3D medical visualization system using fractal dimension, geometric iterations, and diagnostic algorithms. This research explores the development and application of 3D fractal modeling geometry and algorithms for the detection of internal diseases. The primary goal is to enhance the accuracy and efficiency of diagnostic processes by integrating fractal analysis with advanced 3D geometric representations of internal organs and tissues. The study presents a mathematical framework for modeling biological structures using fractal geometry, which captures the complexity and self-similarity of physiological patterns. A novel algorithm is proposed to process medical imaging data and generate 3D fractal models that can be used to identify abnormalities and potential disease markers. Experimental results demonstrate the effectiveness of this approach in detecting irregularities associated with various internal diseases. The findings contribute to the fields of computational medicine, medical imaging, and artificial intelligence, offering a new pathway for early and precise diagnosis.

84-Article

SYSTEMATIC ANALYTIC-GEOMETRIC AUGMENTATION FOR ROBUST DEEP LEARNING

Azimov Rakhimjon, Azizbek Mahkamov

Robustness to geometric distortions and distribution shifts remains a major challenge in deep learning. Standard data augmentation methods, such as random rotations and translations, improve generalization but often fail to systematically cover the geometric transformation space. In this work, we propose an analytic-geometric data augmentation framework that employs controlled affine transformations—including fixed-angle rotations, shearing, and scaling—defined analytically for each training sample. We hypothesize that such systematic augmentation better exposes neural networks to the full spectrum of geometric variations, enhancing their ability to generalize and remain robust under previously unseen conditions. We conduct extensive experiments on MNIST, FashionMNIST, and CIFAR-10 (grayscale), training lightweight convolutional models with both standard and analytic-geometric augmentations. Evaluation on clean, noisy, and rotated test sets demonstrates that analytic-geometric augmentation consistently yields higher accuracy and significantly improved robustness across all scenarios. Notably, our method boosts performance under noise and out-of-distribution rotations by up to 30% compared to conventional augmentations. Our findings highlight the value of explicitly structured augmentation for building more robust and reliable computer vision systems, and suggest that analytic-geometric methods should be incorporated into standard deep learning pipelines for applications where robustness is critical.

85-Article

Development of a Medical Analytical System Based on Intelligent Data Analysis

Ruzibaev O.B., Doshanova M., Sabirov Sh.T.

Explores the design and implementation of a medical analytical system based on intelligent data analysis methods. It discusses modern approaches and machine learning algorithms applied to disease prediction and medical data analysis. The study presents experimental results obtained using various algorithms, including Naive Bayes, Decision Tree, Random Forest, and Gradient Boosting, on medical datasets. The experiments cover data preprocessing, feature selection, and classification quality assessment. Python, Django, and PostgreSQL were chosen as the software platform for system development, ensuring both flexibility and scalability. The primary objective of the system is to provide medical professionals with accessible analytical tools without requiring in-depth knowledge of machine learning. Experimental results demonstrate that ensemble methods, particularly Random Forest and Gradient Boosting, achieved the highest effectiveness.

86-Article

AI FOR URBAN MOBILITY: TRACKING PUBLIC SENTIMENT ON TASHKENT TRANSPORT VIA SOCIAL MEDIA

Bilolidin Khakimov

This paper investigates the use of Natural Language Processing (NLP) to monitor real-time public sentiment about Tashkent’s public transportation system through Uzbek (Latin and Cyrillic) and Russian social media posts. Recent policy changes have transformed the city’s transport network, but their public reception remains underexplored. Over a two-month period (May–June 2025), we collected 980 relevant public posts from social media. After preprocessing for mixed scripts, informal spellings, and code-switching, we applied sentiment analysis using a fine-tuned multilingual BERT (mBERT) model and topic modeling via Latent Dirichlet Allocation (LDA). Results indicate that negative sentiment dominated (68%), with complaints focused on service delays, overcrowding, malfunctioning payment devices, inaccurate real-time tracking, and frequent air conditioning failures in summer. Positive sentiment (9%) centered on the comfort of new articulated and electric buses and polite driver behavior. Our findings demonstrate the potential of NLP tools to provide continuous, low-cost passenger feedback in low-resource language environments, offering transport authorities actionable insights for targeted service improvements.

87-Article

Optimization of algorithms for interacting ROS2 and hardware interfaces

Uktamjon Madaminov

In this paper the methods for improving communication and efficiency in ROS2-based robotic systems are discusses. It highlights the growing complexity of industrial automation and the need for optimized algorithms to ensure smooth data exchange. It focuses on the use of the OSI model as a structured approach to understanding network communications in ROS2. The paper details how each OSI layer contributes to data transfer, reliability, and real-time performance in robotic applications. It also highlights the role of specialized libraries such as Fast DDS, Nav2, and OpenCV in improving system processing speed and responsiveness. Current study discusses algorithm optimization techniques including memory management, parallel processing, and middleware improvements to reduce latency and improve system efficiency. Through this model, it provides the opportunity to transfer data as well as more efficiently, quickly and reliably carry out their reception. The scientific article provides a separate article on these and similar aspects and is widely covered through detailed information.

88-Article

Employing Machine Learning in Building Intelligent Systems for the Protection of Electronic Resources

Musaev Mukhammadjon, Rakhmatov Furkat

Modern electronic resources and web applications are exposed to numerous cyber threats, including DDoS attacks, SQL injections, XSS attacks, Slowloris, and multi-vector attacks. This article presents an intelligent system for detecting and preventing cyber threats, developed based on hybrid traffic analysis methods. The system architecture utilizes SDN/NFV technologies for dynamic scaling and integrates machine learning algorithms: PNHAD (Poisson-Normal Hybrid Anomaly Detection), XGBoost, CRNN/LSTM, and the hybrid adaptive ensemble HAEDFS. Experimental studies on the open datasets CIC-IDS2017 and UNSW-NB15 demonstrated high attack classification accuracy (up to 99.95%) with a low false positive rate (<0.5%). The proposed solution provides adaptive and intelligent protection for electronic resources and can be recommended for implementation in critical infrastructures.

89-Article

INTELLIGENT SYSTEM FOR DISPATCHING THERMAL LOADS OF ELECTRIC POWER FACILITIES

Bakhrieva Khurshida

The article proposes a new approach to dispatching the heat load of electric power facilities based on the use of intelligent systems that employ big data analysis methods and artificial intelligence algorithms. The main objective of the work was to develop the architecture of an intelligent dispatching system that can ensure optimal distribution of the heat load based on predicted and current energy consumption data. The developed system used combined machine learning algorithms, including classification and regression methods, to analyze data on temperature conditions, heat source power and energy consumer states. The use of these methods allowed the system to adapt to changes in real time, predict heat loads and make decisions, minimizing energy losses and increasing the efficiency of energy infrastructure facilities. During the development of the system, the use of big data processing technologies was envisaged to analyze large arrays of information on the current state and demand for heat energy, which ensured more accurate forecasts and optimal distribution of heat flows. The system uses optimization methods that minimize energy costs and balance heat flows taking into account real changes in the load.

90-Article

The role and practical application of IDS systems in information security

Komil Tashev, Dilshod Mirsadikov, Omon Ruzimov

Intrusion Detection Systems (IDS) play a critical role in contemporary cybersecurity architectures by identifying and responding to malicious activities within a network. This paper provides a comprehensive overview of IDS, examining both signature-based and anomaly-based detection methods. Signature-based systems operate by matching traffic against known attack patterns, whereas anomaly-based systems learn normal behavior to detect deviations. Each approach has its strengths and limitations in terms of accuracy, adaptability, and resource consumption. The article further explores hybrid IDS models that combine both methods for enhanced detection capability. Real-world implementations are also discussed, including IDS deployment in enterprise networks, cloud-based infrastructures, and IoT environments. Practical case studies illustrate how IDS can thwart unauthorized access, detect data breaches, and provide forensic insights. As cyber threats grow increasingly complex, IDS solutions are evolving with machine learning and AI-driven techniques to improve responsiveness and minimize false positives. Understanding how these systems function and where they are best applied is vital for developing resilient cybersecurity strategies in both corporate and critical infrastructure sectors.

91-Article

Approaches to Building Semantic Networks for Machine Translation

Usmanova Kamola

This research delves into approaches for identifying semantic links in natural language by analyzing how symbols, concepts, and patterns are close to each other. Two computational techniques were applied: cosine similarity and TF-IDF. Cosine similarity, using a text-word matrix, assessed how often terms appear together within the corpus, while TF-IDF highlighted terms that are both frequent and significant in context. The outcomes were visualized as semantic networks, indicating weighted connections among words. To illustrate, in the sports domain, sportsmen (0.59), badminton (0.49), culture (0.48), and physical education (0.48) adduced the strongest associations with the main concept “sport.” These networks grouped terms into categories such as participants, sporting disciplines, related fields, cultural components, and mobility activities. The study ensures that integrating cosine similarity and TF-IDF effectively captures contextual meaning and supports the development of semantic models useful for machine translation, information retrieval, and natural language processing.

92-Article

The algorithm for identifying word adverbs or adverbials during translation from Uzbek to English

Usmanova Kamola

This paper presents an algorithm in order to identify adverbs and adverbials during translation from Uzbek to English for machine translation. It is irrefutable that the majority of people struggle with distinguishing them in a sentence. Hence, we set a goal to consider some important elements and develop methods so that it helps with machine translation software accuracy and efficiency. In this paper, an algorithm is given to identify adverbs and adverbials by comparing both language morphology and syntactic properties. A stemming and vector-based approach is used in the algorithm, addressing the issues of detecting and translating adverbs and adverbials. This helps the advancement of language use and mitigates language obstacles. By addressing the comparative analysis this method contributes to enhancing natural language understanding in computational systems. With this study it can be paved further research in order to improve semantic interpretation of adverbial elements which is important for translate accurately contextually.

93-Article

TImage Segmentation: Technologies, Algorithms, and Analytics

Fayzullayeva Zarnigor

Image processing and computer vision place great importance on image segmentation, as dividing an image into different regions helps in better understanding its content. Image segmentation may involve several stages. This field is applied across various domains, including medical imaging, autonomous vehicles, augmented reality, and image retrieval systems [1]. The segmentation process requires dividing the image into non-overlapping but meaningful regions, which is critical for analyzing natural scenes [2]. Despite many years of research, feature extraction and modeling remain key challenges. Deep learning technology is used at various stages, including collaborative, traditional, and semantic segmentation. Each stage includes its own specific algorithms and techniques [2]. Due to the diversity of human perception and the uncertainty in visual understanding, representing objects and meaningful regions remains a difficult task. New methods of image segmentation using deep neural networks provide more accurate results by leveraging various datasets.

94-Article

A Model for Synthetic Speech Detection Based on Emotional Variations Using Artificial Intelligence

Furkat Rakhmatov, Fakhriddin Abdirazakov, Azizbek Temirov, Baxodir Achilov, Gulzira Khaldarova, Ruslan Baydullayev

Today, the automatic recognition of emotions in human speech is gaining increased importance due to the growing role of speech interfaces in human computer interaction applications. In this context, the present study analyzes several methods for detecting synthetic speech and proposes a classification approach based on emotion recognition. Five primary human emotions — anger, boredom, happiness, neutral state, and sadness — are considered, and speech that does not correspond to any of these emotional categories is investigated as potentially synthetic.To achieve more accurate results, statistical methods were employed to combine various feature streams. For emotion detection in speech, a feature vector was constructed by combining the following acoustic features: 16 Linear Predictive Coding (LPC) coefficients, 12 Linear Predictive Cepstral Coefficients (LPCC), 16 Log Frequency Power Coefficients (LFPC), 16 Perceptual Linear Prediction (PLP) coefficients, 20 Mel Frequency Cepstral Coefficients (MFCC), and jitter.The proposed method is based on three classification techniques: Linear Discriminant Analysis (LDA), k-Nearest Neighbors (K-NN), and Hidden Markov Models (HMM). The experimental results demonstrate that the selected features are stable and effective in recognizing emotions and yield high performance based on two evaluation metrics.

95-Article

REAL-TIME ASSESSMENT OF HUMAN EMOTIONAL STATE FROM FACIAL IMAGES USING THE EFFICIENTNET MODEL AND THE FACEEMOCDS DATASET

Jurayev U.S., Kurbanov A.A.

The detection of emotional states from human facial expressions is a crucial research area in artificial intelligence and computer vision. This paper explores real-time emotion recognition using the EfficientNet-B0 model on the FaceEmocDS dataset. The FaceEmocDS dataset comprises 72,412 high-quality images, encompassing eight emotion classes: anger, contempt, disgust, fear, happiness, neutral, sadness, and surprise. Derived from multiple open-source datasets, it ensures diversity and balance after rigorous cleaning to remove low-quality or redundant samples. The model was trained via transfer learning, leveraging pre-trained weights from ImageNet, and fine-tuned over 30 epochs on the PyTorch framework. It achieved 74.32% accuracy on the validation set. To enhance generalization, data augmentation techniques were applied, including random rotations (up to 10 degrees), color jittering (adjusting brightness, contrast, and saturation), random resized cropping, horizontal flipping, and random erasing (with probability 0.7). Class weights were incorporated into the CrossEntropyLoss function to address imbalances, particularly in underrepresented classes like contempt and disgust. For real-time implementation, MediaPipe Face Detection was integrated for efficient face localization, followed by emotion classification with the EfficientNet-B0 model. The system operates at 20-30 frames per second (FPS) on standard hardware, making it suitable for dynamic environments. Test results yielded an accuracy of 73.18% and a weighted F1-score of 0.7327, with highest performance on happiness (93.08%) and lowest on contempt (63.05%). Confusion matrix analysis revealed common misclassifications, such as contempt with anger or disgust, highlighting areas for improvement. Compared to baseline models such as VGG and ResNet, this approach provides superior efficiency in terms of parameter count and inference speed while maintaining competitive accuracy.

96-Article

AI and bioinformatics-based methods for advancing smart camera security

Zulaykho Kobulova

The recent adoption of smart cameras in surveillance, smart homes and smart city facilities has cast serious questions on the issue of data protection and privacy. The standard security measures do not usually identify advanced threats and behavioural abnormalities and, consequently, one will need to consider more innovative options. This paper will suggest a combined model that will use bioinformatics tools and artificial intelligence (AI) and use them to improve the security of smart cameras. Using bioinformatics-inspired pattern recognition techniques, profile information of user and device behaviour is created and processed to identify deviations that can signal possible cyberattacks or unauthorised access. Machine learning and deep learning models are used as AI models to classify normal and abnormal activities in real time. The framework is set to handle large scale streams of video data and biometrics features and be scalable and flexible in various environments. The experimental analysis proves that the suggested solution has a tremendous positive impact on the accuracy of detection, the decrease in false alarms, and the increased resistance of smart camera systems to the changing threats. The paper reveals the opportunities of integrating bioinformatics and AI to create intelligent self-adaptive security in IoT-based surveillance systems.

97-Article

Interpolation Methods of Restoration and Comprehensive Processing of the Cardiosignal

Begmamat Dushanov, Narzillo Mamatov

ECG signals bear vital diagnostic information regarding the behavior of the heart but produce vast amounts of data that need to be compressed and accurately reconstructed. In this paper, a Python algorithm for reconstructing ECG signals using Hermite polynomial interpolation with Chebyshev nodes is discussed. The process uses optimally positioned nodes to provide precise segment-wise approximation and altration without losing important morphological details of the ECG wave. Python coding made use of SciPy and NumPy for error estimation and interpolation. In simulations with experiments using the MIT-BIH Arrhythmia Database, it was indicated that the new Hermite–Chebyshev algorithm revealed improved reconstruction accuracy, reduced mean square error, and improved clinically useful structure preservation in waveforms. The e训ciency of the algorithm holds great potential for real-time compression and reconstruction of ECGs with wearable devices and telecardiology

98-Article

Methods for optimizing the parameters of a linear programming problem (LPP)

Mirzayev Anvar Nazirovich, Alikulov Yolkin Kodirovich

this article presents a solution to a linear programming problem and a method that is important for optimizing parameter definition.In this article, we will consider the issues of studying the optimality of the formulation itself LPP. If there are disadvantages, change the task parameters in order to improve the production process and increase the income of the enterprise. We will limit ourselves here to considering these issues using specific examples, but the conclusions we draw will be general in nature and can be applied to a wide range of tasks

99-Article

Analysis of accuracy of local spline functions using analytic functions

Abduganiev Mukhriddin

- After many scientific studies in the world, we can say that the use of spline functions in working with digital signals is widely used and good results are achieved. Nowadays, there are many ways to wish digital signals. Examples of them include Lagrange and Newton interpolation methods, spline function, Fourier series and exponential function approximation methods, orthogonal methods for solving systems of linear equations, discrete wavelet transforms and a number of other interpolation methods. During the research work, cubic Bezier spline function and interpolation Natural spline function were selected for equal intervals. This article describes in detail the process of interpolation of functions to approximate local cubic spline functions to given analytic functions.

100-Article

Development Stages of the Electronic Archival System in Uzbekistan

U.M.Yusupov

This paper analyzes the development of the electronic archive system in Uzbekistan in two consecutive stages (2010–2019 and 2020–2024). The first phase (2010–2019) focused on infrastructure development of state archives, while the second phase (2020–2024) emphasized the digital transformation and implementation of electronic archival information systems. The study outlines institutional structures, statistical indicators, technological integration challenges, and service efficiency gains resulting from system digitization. The paper also presents the architecture of the Integrated Corporate Electronic Archive and evaluates its effectiveness in delivering online archival services.