System research and information technologies

Hybrid system of computational intelligence based on bagging and group method of data handling

Yevgeniy Bodyanskiy, Oleksii Kuzmenko, Helen Zaichenko, Yuriy Zaychenko — Fri, 29 Mar 2024 00:00:00 +0200

The paper considers the problem of short- and middle-term forecasting in the financial sphere. To solve this problem, a hybrid system of computational intelligence based on the group method of data handling (GMDH) and bagging, as well as an algorithm for its training, is proposed. The odd stacks of the hybrid system are formed by ensembles of parallel connected subsystems. ARIMA and the GMDH-neo-fuzzy hybrid network were chosen as such subsystems. The proposed system does not require a large training data set, automatically determines the number of stacks during training, and provides online operation. The experimental investigations were conducted using the proposed hybrid system, as well as separately using ARIMA and GMDH-neo-fuzzy. The accuracy of the predictions obtained is compared, based on which the feasibility of using the proposed hybrid system is substantiated.

Statistical methods of feature engineering for the problem of forest state classification using satellite data

Yevhenii Salii, Alla Lavreniuk, Nataliia Kussul — Fri, 29 Mar 2024 00:00:00 +0200

Timely detection of forest diseases is an important task for their prevention and spread limitation. The usage of satellite imagery provides capabilities for large-scale forest monitoring. Machine learning models allow to automate the analysis of these data for anomaly detection indicating diseases. However, selecting informative features is key to building an effective model. In this work, the application of Bhattacharyya distance and Spearman’s rank correlation coefficient for feature selection from satellite images was investigated. A greedy algorithm was applied to form a subset of weakly correlated features. The experiment showed that selected features allow for improving the classification quality compared to using all spectral bands. The proposed approach demonstrates effectiveness for informative and weakly correlated feature selection and can be utilized in other remote sensing tasks.

Evaluation of the thermal regime of the cathode operation of a high-voltage glow discharge electron gun, which forms a ribbon electron beam

Fri, 29 Mar 2024 00:00:00 +0200

The article discusses the various methods of estimating the surface cathode temperature of the high-voltage glow discharge electron gun, which forms a ribbon electron beam with a linear focus. Numerical estimations have been made to design the cathode assembly of an industrial gun. It is shown that the most effective way to make approximate estimates of the temperature of the cathode surface in high-voltage glow discharge electron guns for various technological purposes is to use arithmetic-logical ratios for modeling the geometry of the cathode assembly and locus functions for estimating the temperature distribution. The accuracy of such estimates, made using the heat balance equation, was 5–10%, sufficient at the initial stage of designing an electron gun. It is shown that using the SolidWorks CAD software complex for designing high-voltage glow discharge electron guns is effective only for solving the complex engineering design tasks and preparing the corresponding technical documentation. The results of the theoretical research published in the article are of interest to a wide range of specialists engaged in developing electron beam equipment and its implementation in industrial production.

Designing Petri nets with strong place and transition anticipation for real-valued functions

Vitalii Statkevych — Fri, 29 Mar 2024 00:00:00 +0200

We propose extending the classic Petri nets and considering D. Dubois’s strong anticipation in two ways. We propose to add a new term into a transition rule that contains a real-valued function of a new marking in a certain place (strong place anticipation) or of a new marking in the input place of a certain transition (an example of strong transition anticipation). Any integer constraints are not applied either to the weight function or to the marking in contrast to the classic Petri nets (as in continuous Petri nets). The execution of the mentioned nets is investigated, and important properties are stated. Several examples of reachability graphs are given, and differences from classic Petri nets are formulated. We also investigate the conditions of the equality of the markings, which are obtained by firing the sequences of transitions t_jt_k and t_kt_j.

Intelligent optimal control of nonlinear diabetic population dynamics system using a genetic algorithm

Abdellatif El Ouissari, Karim El Moutaouakil — Fri, 29 Mar 2024 00:00:00 +0200

Diabetes is a chronic disease affecting millions of people worldwide. Several studies have been carried out to control the diabetes problem, involving both linear and non-linear models. However, the complexity of linear models makes it impossible to describe the diabetic population dynamic in depth. To capture more detail about this dynamic, non-linear terms were introduced into the mathematical models, resulting in more complicated models strongly consistent with reality (capable of re-producing observable data). The most commonly used methods for control estimation are Pantryagain’s maximum principle and Gumel’s numerical method. However, these methods lead to a costly strategy regarding material and human resources; in addition, diabetologists cannot use the formulas implemented by the proposed controls. In this paper, the authors propose a straightforward and well-performing strategy based on non-linear models and genetic algorithms (GA) that consists of three steps: 1) discretization of the considered non-linear model using classical numerical methods (trapezoidal rule and Euler–Cauchy algorithm); 2) estimation of the optimal control, in several points, based on GA with appropriate fitness function and suitable genetic operators (mutation, crossover, and selection); 3) construction of the optimal control using an interpolation model (splines). The results show that the use of the GA for non-linear models was successfully solved, resulting in a control approach that shows a significant decrease in the number of diabetes cases and diabetics with complications. Remarkably, this result is achieved using less than 70% of available resources.

Operational risk estimation using system analysis methodology

Petro Bidyuk, Oxana Tymoshchuk, Liudmyla Levenchuk — Fri, 29 Mar 2024 00:00:00 +0200

Financial risks are considered today as popular research topics due to the existing practical necessity for the use of their mathematical models, estimates of possible loss in many areas of human activities, forecasting, and respective managerial decisions in financial and other spheres where capital, obligations, stocks, bonds, and other activities are circulating successfully. Financial processes today exhibit sophisticated forms of evolution in time that require the application of sophisticated modeling, risk estimating, forecasting, and decision-making/support methods, techniques, and procedures. The system analysis approach is applied to solving such problems as a unique and universal research methodology. The financial risks, specifically the operational ones in the study considered, are classified as nonlinear and nonstationary processes that require appropriate methods for analysis and a rather sophisticated analytical description to estimate and forecast possible loss. The results of operational risk analysis are achieved in the form of systemic methodology, models constructed with statistical data, regression analysis, and Bayesian techniques, and estimated loss with the models. The models and system analysis approach proposed for analyzing financial processes are suitable for practical applications, provided the users have appropriate statistical data and expert estimates.

Study on the profitability of agricultural enterprises in Ukraine during the russian military invasion of Ukraine

Olga Tsesliv, Tamara Dunaieva, Julia Yereshko, Oleksandr Tsesliv — Fri, 29 Mar 2024 00:00:00 +0200

This paper examines the effectiveness of grouping agricultural enterprises according to the wheat harvested area and assesses their profitability. We have developed linear and non-linear regression equations to predict the income for said groups of enterprises. The methodology is designed for cases when future market prices are probabilistic in nature. With the help of the developed methodology, it is possible to calculate the necessary production volumes in the conditions of price fluctuations. We have used the Goldfeld–Quandt parametric test to test the model for heteroscedasticity. Calculations show that agricultural holdings are indeed inefficient, and preference should be given to enterprises with medium crop areas. Application of the Lagrange multipliers method when solving the problem of agricultural enterprise optimization makes it possible to increase profitability. The case of price risk, when future market prices are not deterministic, is considered. Therefore, it is necessary to be guided by two criteria when making managerial decisions: to maximize the expected total net income and to minimize the variance of the total net income.

Potential applications of internet of things: A comprehensive analysis

Mahadevappa Punitha, Puranic Math Rekha — Fri, 29 Mar 2024 00:00:00 +0200

Internet of Things (IoT) is the amalgamation of hardware, like sensors and trackers, which monitor several parameters of the environment or physical objects, and software that processes all the data gathered by hardware. Globally, the IoT market is anticipated to reach 53.8 billion USD by 2025. This enhancing demand is due to its innate ability to automate, which drives several industries to adopt IoT. In addition, minimum memory cost, processing, and storage with an increase in Big Data (BD), cloud, and conjunction of industrial networks and the internet are the added factors for the increase in IoT development. Due to this significance, IoT has applications in numerous areas like medical management, farming, wearable technology, smart energy meters, smart cities, etc. The applications are not limited to the examples mentioned above. Considering this, existing studies have considered different applications and attempted to execute them. As different applications have been focused on by these studies, the present review intends to provide a compilation of potential applications of IoT as considered by conventional research between 2018 and 2022. The study also intends to explore the advantages and disadvantages of different IoT applications (deliberated by conventional studies) through tabular analysis. Further, this review emphasizes IoT’s major key challenges and countermeasures to resolve its security issues. Finally, the study affords recommendations that will assist all IoT experts in bringing IoT products with enhanced security into the market.

Digital medical image encryption approach in real-time applications

Izz Abboud, Muaayed Al-Rawi, Nasir Al-Awad — Fri, 29 Mar 2024 00:00:00 +0200

Patient information and medical imaging data are now subject to stringent data security and confidentiality standards due to the proliferation of telemedicine techniques and medical imaging instruments. Because of the problems described above, as well as the possibility of data or information being stolen, this brings up the dilemma of transmitting data on medical images via an open network. In the past, potential solutions included the utilization of methods such as information concealment and image encryption. Nevertheless, attempting to reconstruct the original image utilizing these approaches may result in complications. In the process of this paper, an algorithm for safeguarding medical images based on the pixels of interest was established. Detection of image histogram peaks for the purpose of calculating peaks in medical images pixels of interest in medical image that have had their threshold values processed. The threshold is shown by taking the average of all the peaks in the histogram. After that, a Sudoku matrix is used to assign values of interest to each of these pixels. The proposed method will be assessed by a variety of statistical procedures, and the outcomes of these analyses will be compared to previously established standards. According to the findings, the suggested method has superior security performance compared to other image encryption methods already in use.

Cardiomyopathy prediction in patients with permanent ventricular pacing using machine learning methods

Fri, 29 Mar 2024 00:00:00 +0200

Pacing-induced cardiomyopathy is a notable issue in patients needing permanent ventricular pacing. Identifying risk groups early and swiftly preventing the ailment can reduce patient harm. However, current prognostic methods require clarity. We employed machine learning to develop predictive models using medical data. Three algorithms — decision tree, group method of data handling, and logistic regression — formed models that forecast pacing-induced cardiomyopathy. These models displayed high accuracy in predicting development, signifying soundness. Factors like age, paced QRS width, pacing mode, and ventricular index during implantation significantly influenced predictions. Machine learning can enhance pacing-induced cardiomyopathy prediction in ventricular pacing patients, aiding medical practice and preventive strategies.