Gulnur Aksholak

Ms. Gulnur Aksholak

orcid

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 February 2026 in Computers
Verified
Journal Article
Open access
PDF file
GAGulnur Aksholak
Gulnur Aksholak
ABAgyn Bedelbayev
Agyn Bedelbayev
RMRaiymbek Magazov
Raiymbek Magazov
KKKaplan Kaplan
Kaplan Kaplan
Network intrusion detection has challenges that fundamentally differ from language and vision tasks typically addressed by Transformer models. In particular, network traffic features lack inherent ordering, datasets are extremely class-imbalanced (with benign traffic often exceeding 80%), and reported accuracies in the literature vary widely (57–95%) without systematic explanation. To address these challenges, we propose a controlled experimental study that isolates and quantifies the impact of tokenization strategies on Transformer-based intrusion detection systems. Specifically, we introduce and compare three tokenization approaches—feature-wise tokenization (78 tokens) based on CICIDS2017, a sample-wise single-token baseline, and an optimized sample-wise tokenization—under identical training and evaluation protocols on a highly imbalanced intrusion detection dataset. We demonstrate that tokenization choice alone accounts for an accuracy gap of 37.43 percentage points, improving performance from 57.09% to 94.52% (100 K data). Furthermore, we show that architectural mechanisms for handling class imbalance—namely Batch Normalization and capped loss weights—yield an additional 15.05% improvement, making them approximately 21× more effective than increasing the training data by 50%. We achieve a macro-average AUC of 0.98, improve minority-class recall by 7–12%, and maintain strong discrimination even for classes with as few as four samples (AUC 0.9811). These results highlight tokenization and imbalance-aware architectural design as primary drivers of performance in Transformer-based intrusion detection and contribute practical guidance for deploying such models in modern network infrastructures, including IoT and cloud environments where extreme class imbalance is inherent. This study also presents practical implementation scheme recommending sample-wise tokenization, constrained class weighting, and Batch Normalization after embedding and classification layers to improve stability and performance in highly unstable table-based IDS problems.

10 April 2026
Verified
Conference Paper
Closed access
RMRaiymbek Magazov
Raiymbek Magazov
FUFatima Uralova
Fatima Uralova
GAGulnur Aksholak
Gulnur Aksholak
KAKuanysh Abeshev
Kuanysh Abeshev
Linux servers exposed via Secure Shell (SSH) remain a major target in command-driven cyber intrusions, yet conventional low- and medium-interaction honeypots often provide unrealistic responses and limited session state, causing attackers to terminate sessions before later attack stages can be observed. This paper proposes an intelligent SSH honeypot based on large language models that maintains context-aware terminal interaction while simultaneously supporting semantic analysis of attacker command streams. The system combines an SSH front-end, an LLM processing layer, centralized logging, and an embedding-based analytics pipeline for clustering commands and mapping them to attack stages such as reconnaissance, exploitation, and persistence. The language model is fine-tuned for Linux command understanding and realistic response generation, and the proposed method is evaluated against token-based and ngram baselines. Results from a six-month distributed deployment with approximately four million interaction events show that the LLM-enhanced approach improves command recognition performance to an F1-score of 0.90 and increases attacker engagement by extending session depth and duration. The findings indicate that LLM-assisted honeypots can provide more realistic deception and richer threatintelligence telemetry from Linux-based cyber incidents.

14 July 2026 in Future Internet
Verified
Journal Article
Open access
PDF file
RMRaiymbek Magazov
Raiymbek Magazov
FUFatima Uralova
Fatima Uralova
KAKuanysh Abeshev
Kuanysh Abeshev
GAGuldana Akhmedi
Guldana Akhmedi
GAGulnur Aksholak
Gulnur Aksholak
Secure Shell (SSH) intrusions against Linux servers remain a dominant vector of opportunistic and targeted cyber incidents, yet operational honeypots treat attacker commands either as isolated lookup keys for static templates or as input to a single uniform language model. Two research directions partially address this gap: LLM-driven response generation raises interaction realism but incurs a fidelity–latency trade-off, while semantic command analytics classifies attack stages from embeddings yet applies a single fine-tuned model uniformly across sessions. This work introduces the phase-adaptive model routing framework (PAMR), which explicitly couples both views. A lightweight online phase estimator infers the current attack stage—reconnaissance, exploitation, or persistence—from a sliding window of recent commands via compact semantic embeddings; a router then dispatches each command to one of several heterogeneous LLM backends according to estimated phase, command complexity, and a confidence-weighted latency budget. The routing decision is formulated as a constrained optimization problem and realized at runtime as an O1 lookup in a precomputed dispatch table. PAMR is evaluated on a controlled, reproducible benchmark of 412 stage-annotated command sessions that combines representative Linux command–response pairs with synthesized attacker traces; we explicitly state that this is a laboratory benchmark rather than live attacker traffic, and we scope our claims accordingly. Relative to uniform-model baselines, PAMR reduces mean response latency by approximately 38% against an API-hosted high-capacity backend and by approximately 45% against a locally hosted mid-capacity backend, while keeping token-level response-fidelity metrics close to the high-capacity baseline (cosine similarity ≈ 0.39 vs. 0.40) and maintaining an online stage-classification macro-F1 above 0.87. We further provide a first-order analytical treatment of the timing side-channel that any backend-routing architecture introduces, and we frame the contribution as a latency/token-fidelity trade-off, leaving validation of operational realism against live adversaries to future work.

3 December 2024 in NEWS OF THE NATIONAL ACADEMY OF SCIENCES OF THE REPUBLIC OF KAZAKHSTAN
Verified
Journal Article
Closed access
GAGulnur Aksholak
Gulnur Aksholak
ABA. Bedelbayev
A. Bedelbayev
RMR. Magazov
R. Magazov
Our study aims to analyze and evaluate modern machine learning methods for detecting malware, a critical challenge given the increasing complexity and volume of cyber threats. We explore various algorithms, including Support Vector Machines, Random Forest, Logistic Regression, Decision Trees, and K-Means Clustering, comparing their effectiveness in identifying and classifying malware. Our methodology combines static, dynamic, and memory-based analysis techniques, offering a comprehensive approach to understanding malware behavior. Key findings reveal that Decision Trees and Random Forests demonstrate impressive accuracy in both binary and multi-class classification tasks. We also highlight novel methods such as the Self-Organizing Incremental Neural Network (SOINN), which effectively handles evolving malware threats. The integration of static and dynamic analysis methods deepens insights into malware behavior. This research underscores the importance of advancing machine learning techniques to enhance cybersecurity measures against evolving global malware threats, offering valuable insights for future research directions.

22 November 2018
Verified
Conference Paper
Closed access
SKSerik Kulmamedov
Serik Kulmamedov
GAGulnur Aksholak
Gulnur Aksholak
The article discusses cryptographic algorithms based on elliptic curves over finite fields. It is shown that elliptic curves allow us to construct examples of finite abelian groups for cryptographic purposes. By evaluating the field, you can easily increase the cipher strength. A variant of the software implementation of this approach is considered. A software-implemented protocol can encrypt messages, generate a digital signature, and after transmitting the message, decrypt them on the recipient's side. The evaluation of the cryptographic strength of the described protocol was made.

20 March 2025 in NEWS OF THE NATIONAL ACADEMY OF SCIENCES OF THE REPUBLIC OF KAZAKHSTAN
Verified
Journal Article
Closed access
GAGulnur Aksholak
Gulnur Aksholak
ABA.A. Bedelbayev
A.A. Bedelbayev
RMR.S. Magazov
R.S. Magazov
In an era marked by rapid technological evolution, especially within cloud computing and container orchestration, the security of systems like Kubernetes is paramount. In this study, we investigate the security architecture of Kubernetes, emphasizing both its strengths and vulnerabilities. We analyze the core components—such as the master node and worker nodes—highlighting their roles in maintaining scalability and resilience in containerized environments. Our research further explores the various security challenges, including misconfigurations and vulnerabilities identified in recent CVEs (Common Vulnerabilities and Exposures), and the potential impact these issues have on Kubernetes deployments. A significant portion of the analysis is dedicated to recent developments in Kubernetes security, such as the introduction of Kubernetes 1.29 and the Kubernetes Gateway API. These advancements are crucial for strengthening the security posture of Kubernetes deployments, particularly in managing network traffic and enforcing access controls. A critical part of our analysis is the comparative evaluation of security tools, including Kube-bench, Sonobuoy, and Kube-hunter. These tools are assessed based on criteria such as usability, customization options, performance overhead, and integration capabilities. Our findings suggest that while these tools provide essential security features, there is still a gap in comprehensive, automated solutions that can address the dynamic and complex nature of Kubernetes environments. We conclude by emphasizing the importance of continuous research and the implementation of advanced security measures to maintain a high level of protection for Kubernetes clusters. The study offers valuable insights into the current state of Kubernetes security and suggests future directions for research, including the development of AI-driven security solutions that can proactively detect and mitigate threats. By focusing on these areas, we aim to contribute to the ongoing efforts in securing cloud-native applications against modern cyber threats.