This article examines a hybrid model for detecting phishing attacks in information and communication systems by combining URL features, email text, behavioral signals, and local domain characteristics. The study treats phishing not only as fake-link detection, but as a multilayer cybersecurity problem involving language cues, user psychology, and network anomalies. Traditional machine learning, deep learning, CNN+BI-LSTM, stacking, and explainable AI methods were compared. Results show that a multi-source integrated model is more effective against smishing, zero-day domains, and messages generated with generative AI
| Mualliflar | Jumaniyozov, Firuz, Kuchkarov, Taxir |
|---|---|
| Jurnal | Al-Farg'oniy avlodlari |
| Nashr sanasi | 2026-04-28 |
| Son | 2 |
| Betlar | 9-14 |
| Til | Rus |
ИКТ, кибербезопасность, фишинг, гибридная модель, CNN, Bi-LSTM, Random Forest, stacking, explainable AI, анализ URL, домен .uz, AKT, kiberxavfsizlik, fishing, gibrid, model, CNN, Bi-LSTM, Random Forest, stacking, explainable AI, URL tahlili, .uz domen, ICT, cybersecurity, phishing, hybrid model, CNN, Bi-LSTM, Random Forest, stacking, explainable AI, URL analysis, .uz domain
This article analyzes the issues of developing a communication management system based on artificial intelligence (AI) in the field of public services. During the research, the role of AI technologies in automating…
The article presents the development of an intelligent control structure for a mechatronic module designed to display 3D graphics. The main objective is to achieve high precision in controlling the module’s spatial…
Infrastructure for training artificial intelligence (AI) models requires high bandwidth and minimal latency. In developing countries, particularly Uzbekistan, the high cost of traditional InfiniBand networks poses a…
This article considers the problem of tracking state variables in nonlinear control objects under conditions of unknown harmonic disturbances affecting the output signal. The main goal of the research is to develop an…
This paper provides a theoretical and practical analysis of the proposed multi-agent architecture for information systems based on a cognitive and modular approach. The study explored the possibilities of organizing the…
Image-based data, in particular, require significant computational resources, slowing down processing. This study investigates the efficiency of parallel image filtering on large-scale data using GPU. Filtering was…
This article examines the mathematical modeling of heat and mass transfer processes under turbulent flow conditions. Flow characteristics are analyzed based on the Reynolds-Averaged Navier-Stokes (RANS) approach…
This study presents a comprehensive simulation analysis of germanium (Ge) diffusion into silicon (Si) substrates with a resistivity of 100 Ω·cm (n-type) in the temperature range of 800–1200 °C using the Synopsys…
This article analyzes the effectiveness of malware detection methods using artificial intelligence technologies. With the rapid development of modern information and communication technologies, the number of malicious…
The scarcity of labeled data remains a fundamental limitation of supervised learning models. This study proposes an autoencoder-based unsupervised approach for detecting anomalies in network traffic. Experiments…