Scientific Journal of Astana IT University
Not a member yet
250 research outputs found
Sort by
ANALYSIS OF MACHINE LEARNING METHODS FOR PREDICTIONS OF STOCK EXCHANGE SHARE PRICES
Get PDFSince the stock market is one of the most important areas for investors, stock market price trend prediction is still a hot subject for researchers in both financial and technical fields. Lately, a lot of work has been analyzed and done in the field of machine learning algorithms for analyzing price patterns and predicting stock prices and index changes. Currently, machine-learning methods are receiving a lot of attention for predicting prices in financial markets. The main goal of current research is to improve and develop a system for predicting future prices in financial markets with higher accuracy using machine-learning methods. Precise predicting stock market returns is a very difficult task due to the volatile and non-linear nature of financial stock markets. With the advent of artificial intelligence and machine learning, forecasting methods have become more effective at predicting stock prices. In this article, we looked at the machine learning techniques that have been used to trade stocks to predict price changes before an actual rise or fall in the stock price occurs. In particular, the article discusses in detail the use of support vector machines, linear regression, and prediction using decision stumps, classification using the nearest neighbor algorithm, and the advantages and disadvantages of each method. The paper introduces parameters and variables that can be used to recognize stock price patterns that might be useful in future stock forecasting, and how the boost can be combined with other learning algorithms to improve the accuracy of such forecasting systems
NEURAL NETWORK MODEL OF SOIL MOISTURE FORECAST NORTH KAZAKHSTAN REGION
Get PDFDealing with agriculture, it is valuable to know an amount of moisture in a soil and to know how to forecast the stored soil moisture within particular period. Forecasting the stored soil moisture works for planning an extent and structure of crop production areas and adjustment of plant-growing programs. Having known about an amount of moisture in one-meter soil and the depth of precipitation in a vegetation season shall help farmers to determine a seeding time, type of fertilizers depending on soil quality and to work out an irrigation schedule as well. In this regard, over the last few years some vigorous activities applied to machine training methods of the weather forecast have been launched in the world. The goal of present research is to develop an artificial neuron network which shall afford an opportunity to figure out a stored soil moisture prior to outgoing to winter in a short-term. North Kazakhstan Region agrometeorological measuring stations for the period from 2012 to 2022 were used in the course of the neuron network training. The Levenberg-Marquardt algorithm aimed at non-linear regression models optimization was chosen for network training. The algorithm includes sequential approximation of initial parameter values to a local optimum. The mean squared error (MSE) function and the correlation coefficient ensure accuracy and precision of forecasts. As a result, 7 neural networks under MATLAB environment using the Levenberg-Marquardt algorithm, with different input and output data, and with different number of learning iterations came to realization. Following analysis of the results, the choice was fallen on the ANN9 best network offering minimum error function and actual data maximum correlation. The neural network obtained fits for use to make efficient decisions in the North Kazakhstan region agricultural sector in the short term
MODELING OF A DIGITAL COMMUNICATION SYSTEM WITH INTERFERENCE-RESISTANT CODING OVER DELAYED MULTIPATH CHANNELS FOR A FIXED WIRELESS ACCESS SYSTEM
No full textIn this study, an analysis of the performance of a digital communication system was conducted to improve the efficiency of the communication channel. The system was de- signed and evaluated in the MatLab/Simulink environment, specifically for the transmission of binary data in a multipath channel with static fading. One of the main advantages of this system is its ability to provide high noise immunity, even in the presence of noise, interference, and signal delays. The performance of the system was investigated in terms of bit error rate (BER) over both an additive white Gaussian noise (AWGN) channel and a multipath Rayleigh fading channel. The study also considered the impact of inter-symbol interference (ISI) and explored different parameters for a static channel, resembling a wireless LAN system based on IEEE 802.11 (as defined by Rec ITU-R M.1225). This study uses a solution to suppress inter-symbol interference using cascading coding (convolutional turbo codes with Reed Solo- mon code), Fast Fourier Transform (FFT) and Inverse Fast Fourier Transform (IFFT), amplifiers at the transmitter side and normalizer at the receiver side, by regulating the amplification ratios of transmitter and receiver sides. Modeling of a digital communication system for a wide-band fixed wireless LAN system (for Indoor office and Outdoor to indoor systems) has been performed. The modeling outcomes show that the applied method provides a good per- formance improvement in channels with ISI with static fading.
The application of cascading coding (turbo code together with RS code), amplifiers, and FFT on the transmitter side, and IFFT on the receiver side will eliminate the effects of inter-symbol interference on digital signals in a multipath channel with static fading (as defined by Rec. ITU-R M.1225 for internal and external transmission systems) for wireless fixed systems
METHODS OF PROJECT-VECTOR MANAGEMENT OF EDUCATIONAL ENVIRONMENTS
Get PDFBased on the developed mathematical model of the project-vector space, the methods of determining the endpoints of the objects of the project-vector space (PVS) and the calculation of the trajectory of the movement to these points are proposed. It is shown that the problem of developing these methods is related to the definition of such projects (project integration), which will correspond to the maximum expansion of “Universal Projects” educational environment. Calculation of the trajectory of the movement in the project-vector space ensures the achievement of the objectives of the project with minimal time and financialresources. It is proposed to use the Monte Carlo method to calculate the options for the trajectory of PVS subjects. This distribution of probabilities when choosing the displacement of objects and subjects of PVS corresponds to the priority of subjects and the impact of objects on the displacement of these subjects. For the assessment of the magnitude of the impact on the movement of PVS subjects it was developed the structure of the expert table and the model of the calculation of the average expert assessment of such an impact. Implementation of the given methods will allow to dynamically evaluate the most important goals for all interestedparties of the project, as well as to develop ways to achieve them in real terms. In this case, the real conditions of the projects are described in the system of impact on the movement of objects and subjects of PVS in the expanding “Universal Projects”
MATHEMATICAL SUPPORT OF THE INFORMATION SYSTEM FOR DECISION SUPPORT IN THE SPHERE OF HEALTHCARE
Get PDFThe relevance of the topic is that currently modern medical information systems are aimed at providing management, economic and in some cases medical practice in the collection and processing of anamnestic data, including dental, ophthalmological, radiological, anesthesiological, resuscitation. This study goal is to develop models of basic principles and structural and functionalscheme of the decision support system as a tool that allows to model the process of “clinical thinking” of an endocrinologist in determining plans for drug treatment of diabetes based on a scenario approach and decision theory. In accordance with the set goal, the research tasks are formulated and solved, the essence of which is as follows:
- analyze existing medical decision support systems and identify shortcomings of such systems;- creating the criteria for developing medical decision support systems to improve usability;- develop acceptable algorithms to create a medical decision support system.Recommendations for modeling the functions of the doctor’s intelligence in the process of processing and evaluation of medical information using a personal computer are offered, which will improve the operation of existing medical information systems. The research was conducted within the framework of project management methodology and decision theory using information technology tools. The algorithms for the creation a medical decision support system proposed by the authors are based on the method of multicriteria ranking of alternatives, which is a tool for modeling “clinical thinking” and logical reasoning of an endocrinologist. Unlike existing medical information systems, it will not only collect, store and process information about patients,but also increase the efficiency of endocrinologist decisions regarding the prediction and development of a patient treatment plan
SIMPLIFIED ADAPTIVE TRIANGULATION OF THE CONTACT BOUNDARIES OF THE DAM MODEL
Get PDFTo numerically solve the system of integral equations, it is customary to establish a discrete grid within each integration area. In the context of 3D modeling, these areas correspond to surfaces situated in space. The standard discretization technique employed for the computational domain is triangulation. This study addresses the integral equation system pertinent to the electrical tomography of dams. The structural model encompasses an embankment dam, the upstream and downstream water bodies, the dam's base, and a potential leakage region on the upstream side. An alternative configuration may be encountered in specific scenarios with no water downstream. Consequently, the model may incorporate up to nine distinct contact boundaries. Accordingly, the system of integral equations comprises an equivalent number of equations. Effectively resolving this system through numerical methods necessitates applying triangulation techniques to these diverse surfaces. While mathematical packages like Matlab offer triangulation functions, they may not fully address the specific demands of the problem. Additionally, the grid resolution should be heightened in proximity to key elements such as the sounding line, the supply electrode, and the various contact lines within the medium. These considerations transform the triangulation task into a distinct subtask within the numerical simulation of the resistivity tomography problem. In this paper, we provide our specific approach to this problem. The simplification of the triangulation algorithm is rooted in the predominant utilization of the two-dimensional geometric properties inherent to the object under study. For most contact boundaries, the triangulation is constructed layer by layer with a gradual modulation in triangle dimensions as one progresses from one layer to the next, orthogonal to the axis of the dam. Concerning the surface corresponding to the leakage area, cylindrical coordinates are used for surface parameterization. This approach enables partitioning the surface into discrete strata, facilitating a systematic, layer-by-layer grid construction. Additionally, points at the intersections of contact boundaries are integrated into the pre-existing triangulation by applying a standard function within the Matlab package. In the future, the mathematical modeling based on the Integral Equation Method with adaptive discretization will help incorporate real-time computations into information systems related to monitoring hydraulic structures
DEVELOPMENT OF DAG BLOCKCHAIN MODEL
Get PDFIn this study the authors present an innovative approach to resolving scalability and efficiency challenges in blockchain technology through the integration of Directed Acyclic Graphs (DAGs). This approach helps to overcome the limitations of traditional blockchain systems, particularly in transaction processing. The classic blockchain has some problems as slow transaction processing and poor scalability. The authors offer Directed Acyclic Graph (DAG) as a scalable and energy-efficient alternative. The paper outlines the development of a DAG-based blockchain model, utilizing Python and Flask alongside the Ed25519 cryptographic curve. It conducts a comparative analysis of DAG with traditional consensus mechanisms like Proof of Work and Proof of Stake, underscoring the efficiency and security benefits of employment of DAG. The research methodology includes an extensive literature review and the construction of a practical model to demonstrate DAG's applicability in blockchain networks. Particularly notable is the exploration of DAG's potential in Internet of Things (IoT) ecosystems, addressing critical issues such as energy inefficiency and network communication challenges in existing consensus algorithms. The authors calculated the performance of the model and compared it with similar models on several evaluation criteria. The simulation results of our proposed model show an improvement in performance and security by minimizing end-to-end delay, time cost, energy consumption, and throughput. The model eliminates the limitations of classic blockchain systems, such as high latency and low scalability. It structures transactions and blocks as a DAG, which provides fast validation and high scalability without compromising security. The research demonstrates the transformative implications of DAG for advancing blockchain technology
DEVELOPMENT OF A DATABASE OF DIGITAL MULTICULTURAL CONTENT AND APPLICATION IN JOURNALISM LESSONS
Get PDFThe development of digital technologies stimulates changes in the field of education, including for students of the educational programs «Digital Journalism» and «Media Technologies». This paper presents the practice of using a database of digital multicultural content in journalism lessons. The relevance of this scientific publication is due to the need to integrate digital technologies into the educational process and the development of students’ skills in working with programming languages. The importance of using such content is described by the modern realities of the educational system, when all members of the academic community are involved in the process of forming national identity within the country and global globalization. Kazakhstan’s conditions demonstrate the penetration of cultures of various ethnic groups, religions, which is the reality of modern teachers and students. The author offers a model for creating a database of digital multicultural content. To do this, the multicultural content of three TV channels and online publications posted on digital services is being studied. The design and structuring of web portals, interfaces, and content are studied. The author conducts quantitative and qualitative research of media materials. A parallel is drawn between the amount of digital content published on television and on Internet resources. Through lexical and compositional analysis, trends and issues of digital multicultural content are identified. The information obtained during the study was organized into a database in the PostgreSQL programming language. As an experiment, the use of the database was tested in journalism lessons when performing exercises and conducting mini-studies. At the end of the training course, a survey of students was conducted on the effectiveness of using databases in the classroom. The students’ answers are analyzed and presented in a generalized form. The methodological methods of working with students described and analyzed in this article are important material for the development of the educational goals of future journalists
FORECASTING CUSTOMER FUTURE BEHAVIOR IN RETAIL BUSINESS USING MACHINE LEARNING MODELS
Get PDFThe ability to forecast customers’ future purchases, lifetime value, and churn are fundamental tasks in business management. These tasks become more complicated when the relationship between customers and business is not contractual. Therefore, the application of an appropriate method of customer analysis influences the efficiency of company cost management in interaction with their customers. The purpose of this paper is to compare existing solutions of customer lifetime value prediction and provide a new way to predict the future behavior of customers with consideration of the drawbacks of previous works. The method should have the following properties: use data that is available in any retail business; take into account that markets are constantly changing; be more precise than existing solutions. In this paper, we proposed the method of identifying customer churn provided a way to analyze customer behavior associated with churn or retention. In order to understand why customers churn, we used eleven customer behavioral metrics. The relationship of used metrics with churn was proved using churn cohort analysis. The results of training of logistic regression and neural network on prepared dataset showed that their forecast accuracy is in the healthy range for highly predictable churn. Based on predicted churn probabilities, we calculated the customer lifetime value in the future period. Our research results on customer behavior in the retail business confirm the hypothesis that customers who make many purchases are less likely to churn than customers who make few purchases. The main uniqueness of this work is the way of finding customer churn, as no such data was provided in the initial dataset. In addition, the minimum amount of data that most retail companies have was used. This enables the proposed methodologies to be applied to a large number of retail companies
DATA SECURITY, MODELING AND VISUALIZATION OF DATA FROM IOT DEVICES
Get PDFThe article describes the IoT infrastructure, the hardware of the IoT system, considers the issue of security of the chosen LoRa data transmission technology. Data was received from sensors for gas, temperature and humidity, atmospheric pressure, as well as the location of the end device. At the same time, the standardized security features of the selected LoRa technology for transmitting data from sensors to the server were investigated. The article deals with LoRa bi-directional secure communication line, the security function requires devices/end devices to be configured through the LoRa gateway. Security research is devoted to the development of a security mechanism to increase its resilience. The payload was formed with a hash of the last bytes, and the entire payload was encrypted with AES for integrity and confidentiality. A method for assessing and visualizing atmospheric air pollution is given on the example of the city of Almaty, Kazakhstan. The process of numerical modeling of the study of emissions of harmful substances into the atmosphere is based on a mathematical model formed by the system of Navier-Stokes equations, consisting of the continuity equation, as well as the equations of motion and the k-epsilon turbulence model. To test the numerical methods for processing mixing and chemical reactions, a test problem was chosen – a jet in a transverse flow. Three-dimensional numerical simulation has been implemented. The use of the Internet of Things (IoT) and the acquisition of big data made it possible to simultaneously observe the concentrations of several pollutants in the atmosphere, calculate this concentration and analyze the state of the surface air layer. Modeling allows forecasting the possible concentration of pollutants in certain areas at certain times of the year