Olesia Yakovenko
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Olesia Yakovenko has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Olha Kryvokulska
National University "Kyiv Aviation Institute
4 shared publications - Yaroslav Marchenko
National University "Kyiv Aviation Institute
2 shared publications - Yevhenii Yakymchuk
National University "Kyiv Aviation Institute
2 shared publications Anna Ilienko
State University "Kyiv Aviation Institute
1 shared publication- Anna Ilyenko
State University “Kyiv Aviation Institute
1 shared publication - Olena Dubchak
National University "Kyiv Aviation Institute
1 shared publication
Latest Publications
- Open Access
- 24 September 2026
- Published by Borys Grinchenko Kyiv University in Cybersecurity: Education, Science, Technique
- Vol. 2 (34) , 757-765
NEXT GENERATION PHISHING: THE EVOLUTION OF SOCIAL ENGINEERING IN THE CONTEXT OF DIGITAL TRANSFORMATION
This article analyzes current trends in phishing, including the use of artificial intelligence technologies to automate the creation of phishing content, increase the level of personalization, and optimize attack scenarios. It examines the influence of the human factor as a key vulnerability that determines the effectiveness of social engineering. A quantitative analysis of the dynamics of phishing attacks in recent years is conducted, demonstrating an exponential increase in the number of incidents and indicating the active use of automated tools by cybercriminals. The distribution of channels used to carry out phishing attacks -specifically email, SMS, voice communications, and social networks - was examined separately, allowing for the determination of their relative effectiveness and risk level. A multifactorial mathematical model for assessing the probability of a phishing attack’s success is proposed, which takes into account the level of security of the information system, the effectiveness of the social engineering methods used, the characteristics of the distribution channel, as well as user behavioral patterns. Unlike existing approaches, the model has been extended by introducing a time-dependent adaptation function, which allows for changes in attack parameters over time. Additionally, elements of probability theory are used to model attack intensity, and the information entropy approach is employed to assess uncertainty in the selection of phishing channels. The results confirm that modern phishing is a complex, multidimensional threat that combines technical, behavioral, and informational aspects. The proposed model allows for the formalization of the risk assessment process, enables the prediction of attack effectiveness, and can be used to develop adaptive cybersecurity systems. The practical significance of the study lies in the applicability of the results to information security management systems, as well as in raising user awareness of modern social engineering methods.
- 24 September 2026
- Published by Borys Grinchenko Kyiv University in Cybersecurity: Education, Science, Technique
- Vol. 2 (34) , 757-765
Open AccessNEXT GENERATION PHISHING: THE EVOLUTION OF SOCIAL ENGINEERING IN THE CONTEXT OF DIGITAL TRANSFORMATION
This article analyzes current trends in phishing, including the use of artificial intelligence technologies to automate the creation of phishing content, increase the level of personalization, and optimize attack scenarios. It examines the influence of the human factor as a key vulnerability that determines the effectiveness of social engineering. A quantitative analysis of the dynamics of phishing attacks in recent years is conducted, demonstrating an exponential increase in the number of incidents and indicating the active use of automated tools by cybercriminals. The distribution of channels used to carry out phishing attacks -specifically email, SMS, voice communications, and social networks - was examined separately, allowing for the determination of their relative effectiveness and risk level. A multifactorial mathematical model for assessing the probability of a phishing attack’s success is proposed, which takes into account the level of security of the information system, the effectiveness of the social engineering methods used, the characteristics of the distribution channel, as well as user behavioral patterns. Unlike existing approaches, the model has been extended by introducing a time-dependent adaptation function, which allows for changes in attack parameters over time. Additionally, elements of probability theory are used to model attack intensity, and the information entropy approach is employed to assess uncertainty in the selection of phishing channels. The results confirm that modern phishing is a complex, multidimensional threat that combines technical, behavioral, and informational aspects. The proposed model allows for the formalization of the risk assessment process, enables the prediction of attack effectiveness, and can be used to develop adaptive cybersecurity systems. The practical significance of the study lies in the applicability of the results to information security management systems, as well as in raising user awareness of modern social engineering methods.
- Open Access
- 22 April 2026
- Published by State University "Kyiv Aviation Institute" in Ukrainian Scientific Journal of Information Security
- Vol. 32 (1) , 40-45
Сучасні загрози кібербезпеці критичної інфраструктури в умовах гібридної війни
У статті систематизовано сучасні загрози кібербезпеці критичної інфраструктури в умовах гібридної війни на прикладі України. Проведено аналіз еволюції кібератак у 2015–2025 роках — від шкідливого ПЗ BlackEnergy та Industroyer до wiper-кампаній, фішингових атак із використанням штучного інтелекту та цілеспрямованих впливів на державні реєстри. Ви-значено найбільш уразливі сектори критичної інфраструктури, проведено систематизацію типів загроз і запропоновано комплексні заходи підвищення кіберстійкості на основі українського та міжнародного досвіду.
- 22 April 2026
- Published by State University "Kyiv Aviation Institute" in Ukrainian Scientific Journal of Information Security
- Vol. 32 (1) , 40-45
Open AccessСучасні загрози кібербезпеці критичної інфраструктури в умовах гібридної війни
У статті систематизовано сучасні загрози кібербезпеці критичної інфраструктури в умовах гібридної війни на прикладі України. Проведено аналіз еволюції кібератак у 2015–2025 роках — від шкідливого ПЗ BlackEnergy та Industroyer до wiper-кампаній, фішингових атак із використанням штучного інтелекту та цілеспрямованих впливів на державні реєстри. Ви-значено найбільш уразливі сектори критичної інфраструктури, проведено систематизацію типів загроз і запропоновано комплексні заходи підвищення кіберстійкості на основі українського та міжнародного досвіду.
- Open Access
- 26 March 2026
- Published by Borys Grinchenko Kyiv University in Cybersecurity: Education, Science, Technique
- p. 711
INTELLIGENT TECHNOLOGIES IN CYBERSECURITY: ANALYSIS OF THE POTENTIAL AND CHALLENGES OF THE APPLICATION OF ARTIFICIAL INTELLIGENCE
The article provides a systematic analysis of the opportunities and challenges of applying artificial intelligence (AI) in the field of cybersecurity. The feasibility of integrating AI technologies into modern cyber defense strategies in the face of increasing complexity and intensity of cyber threats is substantiated. The main areas of AI use are identified, in particular, the automation of anomaly detection, forecasting cyber threats, accelerating incident response, and the formation of adaptive protection systems. Practical cases of implementing AI solutions are analyzed, which confirm the increase in the accuracy of threat detection and reduction of response time. Special attention is paid to the risks and vulnerabilities of AI-oriented systems, in particular, threats such as adversarial attacks and data poisoning, as well as the use of AI by attackers to implement AI-assisted attacks. The prospects for the development of AI technologies in 2025–2026 are considered, including the introduction of autonomous intelligent agents (agentic AI) in operational cyber defense centers. It is shown that the transition to autonomous protection models requires increased control, the use of hybrid approaches (AI combined with human expert supervision) and the improvement of regulatory and ethical mechanisms. The results of the study allow us to form a comprehensive vision of the evolution of AI-oriented cyber defense systems and justify the need for a balanced combination of technological capabilities and risk management.
- 26 March 2026
- Published by Borys Grinchenko Kyiv University in Cybersecurity: Education, Science, Technique
- p. 711
Open AccessINTELLIGENT TECHNOLOGIES IN CYBERSECURITY: ANALYSIS OF THE POTENTIAL AND CHALLENGES OF THE APPLICATION OF ARTIFICIAL INTELLIGENCE
The article provides a systematic analysis of the opportunities and challenges of applying artificial intelligence (AI) in the field of cybersecurity. The feasibility of integrating AI technologies into modern cyber defense strategies in the face of increasing complexity and intensity of cyber threats is substantiated. The main areas of AI use are identified, in particular, the automation of anomaly detection, forecasting cyber threats, accelerating incident response, and the formation of adaptive protection systems. Practical cases of implementing AI solutions are analyzed, which confirm the increase in the accuracy of threat detection and reduction of response time. Special attention is paid to the risks and vulnerabilities of AI-oriented systems, in particular, threats such as adversarial attacks and data poisoning, as well as the use of AI by attackers to implement AI-assisted attacks. The prospects for the development of AI technologies in 2025–2026 are considered, including the introduction of autonomous intelligent agents (agentic AI) in operational cyber defense centers. It is shown that the transition to autonomous protection models requires increased control, the use of hybrid approaches (AI combined with human expert supervision) and the improvement of regulatory and ethical mechanisms. The results of the study allow us to form a comprehensive vision of the evolution of AI-oriented cyber defense systems and justify the need for a balanced combination of technological capabilities and risk management.
- Open Access
- 25 December 2025
- Published by State University "Kyiv Aviation Institute" in Ukrainian Scientific Journal of Information Security
- Vol. 31 (3) , 207-214
Аналіз та класифікація вебзагроз у системах з інтелектуальними помічниками
У статті досліджено трансформацію моделей вебзагроз в умовах переходу від традиційних вебдодатків до розподілених систем з інтегрованими інтелектуальними помічниками. На відміну від наявних підходів, що зосереджені переважно на технічних вразливостях (SQL-інʼєкції, міжсайтовий скриптинг, порушення автентифікації), у цій роботі обґрунтовано, що інтелектуальний помічник виступає новою поверхнею атак, для якої класичні методи моделювання загроз є принципово недостатніми. Показано, що ключовий ризик зміщується з рівня синтаксичної валідації вхідних даних на рівень семантичного контролю намірів, меж повноважень і контексту агрегації. На основі порівняльного аналізу традиційних і AI-орієнтованих систем виокремлено пʼять категорій нових загроз: маніпуляція намірами користувача, витік через агрегацію даних, некоректна робота моделі, надмірні повноваження агента та порушення контексту доступу. Запропоновано чотирирівневу концептуальну модель аналізу загроз, орієнтовану на інтелектуальні компоненти, яка охоплює рівні взаємодії з користувачем, інтерпретації запиту, виклику сервісів і доступу до ресурсів. Результати підтверджують необхідність розширення існуючих стандартів моделювання загроз (зокрема STRIDE) і можуть бути використані при проєктуванні захищених інформаційних систем з AI-компонентами.
- 25 December 2025
- Published by State University "Kyiv Aviation Institute" in Ukrainian Scientific Journal of Information Security
- Vol. 31 (3) , 207-214
Open AccessАналіз та класифікація вебзагроз у системах з інтелектуальними помічниками
У статті досліджено трансформацію моделей вебзагроз в умовах переходу від традиційних вебдодатків до розподілених систем з інтегрованими інтелектуальними помічниками. На відміну від наявних підходів, що зосереджені переважно на технічних вразливостях (SQL-інʼєкції, міжсайтовий скриптинг, порушення автентифікації), у цій роботі обґрунтовано, що інтелектуальний помічник виступає новою поверхнею атак, для якої класичні методи моделювання загроз є принципово недостатніми. Показано, що ключовий ризик зміщується з рівня синтаксичної валідації вхідних даних на рівень семантичного контролю намірів, меж повноважень і контексту агрегації. На основі порівняльного аналізу традиційних і AI-орієнтованих систем виокремлено пʼять категорій нових загроз: маніпуляція намірами користувача, витік через агрегацію даних, некоректна робота моделі, надмірні повноваження агента та порушення контексту доступу. Запропоновано чотирирівневу концептуальну модель аналізу загроз, орієнтовану на інтелектуальні компоненти, яка охоплює рівні взаємодії з користувачем, інтерпретації запиту, виклику сервісів і доступу до ресурсів. Результати підтверджують необхідність розширення існуючих стандартів моделювання загроз (зокрема STRIDE) і можуть бути використані при проєктуванні захищених інформаційних систем з AI-компонентами.
- Open Access
- 1 January 2024
- Published by Borys Grinchenko Kyiv University in Cybersecurity: Education, Science, Technique
- Vol. 1 (25) , 318-329
PROSPECTS OF INTEGRATION OF ARTIFICIAL INTELLIGENCE IN CYBERSECURITY SYSTEMS
The article examines in detail the application of artificial intelligence (AI) in the field of cybersecurity, emphasizing its key role in predicting and preventing cyber attacks in the face of growing cyber threats. The study shows that 51% of large organizations already use AI to detect threats, and 12% were able to reduce cybersecurity costs thanks to its implementation. The authors propose a theoretical model that includes the stages of scanning the system for vulnerabilities, assessing the possible consequences of their exploitation, predicting attacks on based on the analysis of historical data and behavioral patterns of attackers, as well as the development of complex protection strategies. The theoretical model proposed in the article demonstrates how the integration of AI can significantly increase the effectiveness of cyber security systems, providing a deeper understanding of threats and more effective methods of their prediction and prevention. Special attention is paid to the latest AI technologies, such as OpenAI's ChatGPT, which is based on the principles of reinforcement learning with human feedback. The practical implementation of this model using the latest AI technologies, such as GPT-4, can be an important step in strengthening the protection of information systems against growing cyber threats. The article emphasizes the need for close cooperation between people and technology to achieve optimal results in countering cyber threats. Implementation of AI requires careful planning and constant adaptation of strategies to new challenges. Thus, the use of artificial intelligence in the field of cybersecurity opens up new opportunities for detecting, predicting and preventing cyber threats.
- 1 January 2024
- Published by Borys Grinchenko Kyiv University in Cybersecurity: Education, Science, Technique
- Vol. 1 (25) , 318-329
Open AccessPROSPECTS OF INTEGRATION OF ARTIFICIAL INTELLIGENCE IN CYBERSECURITY SYSTEMS
The article examines in detail the application of artificial intelligence (AI) in the field of cybersecurity, emphasizing its key role in predicting and preventing cyber attacks in the face of growing cyber threats. The study shows that 51% of large organizations already use AI to detect threats, and 12% were able to reduce cybersecurity costs thanks to its implementation. The authors propose a theoretical model that includes the stages of scanning the system for vulnerabilities, assessing the possible consequences of their exploitation, predicting attacks on based on the analysis of historical data and behavioral patterns of attackers, as well as the development of complex protection strategies. The theoretical model proposed in the article demonstrates how the integration of AI can significantly increase the effectiveness of cyber security systems, providing a deeper understanding of threats and more effective methods of their prediction and prevention. Special attention is paid to the latest AI technologies, such as OpenAI's ChatGPT, which is based on the principles of reinforcement learning with human feedback. The practical implementation of this model using the latest AI technologies, such as GPT-4, can be an important step in strengthening the protection of information systems against growing cyber threats. The article emphasizes the need for close cooperation between people and technology to achieve optimal results in countering cyber threats. Implementation of AI requires careful planning and constant adaptation of strategies to new challenges. Thus, the use of artificial intelligence in the field of cybersecurity opens up new opportunities for detecting, predicting and preventing cyber threats.