Murat Dener

Bridging the gap between predictive accuracy and interpretability in web application security, this study benchmarks machine learning (ML) and deep learning (DL) models for intrusion detection using the publicly available FWAF dataset. We evaluate six widely used classifiers—Logistic Regression, Decision Tree, Random Forest, XGBoost, 1D-CNN, and LSTM—under a highly imbalanced setting (benign-dominant traffic). All models are trained on identical data using a unified pipeline with stratified train/validation/test splits (64%/16%/20%) and class-weight balancing to ensure a fair comparison. Models are assessed using accuracy, precision, recall, F1-score, and imbalance-aware measures (Macro-F1, PRAUC, and MCC). To improve transparency, we integrate Explainable Artificial Intelligence (XAI) techniques—specifically SHAP (SHapley Additive ExPlanations) and LIME (Local Interpretable Model-Agnostic Explanations)—to quantify feature contributions and highlight decision patterns relevant to WAF operation. An ablation study confirms that attack-indicator features (is_xss, is_lfi, is_oci, is_sqli) are derived independently from labels via regex matching, with their removal causing only marginal performance degradation (1–2% Macro-F1). Experimental results show that Random Forest achieves the strongest overall performance (99.3% accuracy, 0.949 Macro-F1, 0.935 PR-AUC), while the feature-based LSTM provides comparable results (99.2% accuracy, 0.937 Macro-F1, 0.928 PR-AUC). Ultimately, the study emphasizes the value of balancing performance with interpretability, empowering Security Operations Centers (SOC) to validate automated decisions and foster trustworthy AI-driven web application firewalls

The manual classification of diplomatic documents by security sensitivity level is labor-intensive and inconsistent at scale. This study proposes SACHN-DeBERTa-v3-Large, a hybrid classification architecture integrating a DeBERTa-v3-Large backbone with a Security-Aware Gate, Prototype Classification Layer, and Supervised Contrastive Projection Head. A rule-based preprocessing pipeline removes inline classification markers prior to training, ensuring that models learn semantic content rather than surface-form annotation artifacts. Two diplomatic corpora are evaluated: the WikiLeaks Cable Classifier (9005 documents, three classes) and a novel Foreign Relations of the United States (FRUS) dataset (24,706 documents, four classes) constructed for this study. A controlled multi-LLM comparative study evaluates LLaMA-3-8B-Instruct, Qwen2.5-14B-Instruct, and Mistral-7B-Instruct-v0.3 under identical QLoRA fine-tuning conditions. SACHN-DeBERTa-v3-Large achieves 96.12% accuracy and 92.57% macro F1 on WikiLeaks, and 92.11% accuracy and 91.02% macro F1 on FRUS, surpassing all baselines by at least 10 percentage points in accuracy and 26 points in macro F1 (McNemar’s test, p < 0.001). Under the primary QLoRA protocol (r = 16), no evaluated LLM exceeded 57.16% accuracy; extended experiments with r = 64 and 8-bit INT8 quantisation yielded a best result of 64.48% (Mistral-7B-Instruct, WikiLeaks), confirming that the performance gap relative to SACHN-DeBERTa-v3-Large remains structural. Post-hoc SHAP and LIME analyses confirm that predictions are grounded in domain-specific semantic content, validating the classification marker removal methodology.

Network Access Control, a critical element in protecting IT and OT environments, plays a vital role in determining who or what can access network resources and what features the device that will provide this connection must have. The primary function of NAC is to enforce security policies at the layer 2 and layer 3 of the enterprise network, control access to network resources, and prevent unauthorized access to the network. It acts as a security guard by examining every device that attempts to connect to the network and ensuring compliance with established security policies and protocols. In an age where insider threats pose a significant risk, NAC’s ability to manage, monitor and restrict network access based on user credentials and device compatibility is invaluable. Blockchain technology, renowned for its robust security features such as decentralization, immutability and transparency, enhances the capabilities of NAC solutions. When integrated with NAC, blockchain offers a new layer of security and trust. It ensures that the access control and monitoring activities are not only well-documented and well-controlled but also immutable and transparent. This integration significantly reduces the chances of unauthorized access and internal threats, thereby fortifying the network’s defense against advanced persistent threats (APT) and sophisticated cyber attacks. This study proposes the integration of NAC within the Hyperledger Fabric network architecture. Both NAC and Hyperledger Fabric are compatible for private networks so blockchain’s decentralized structure and its ability to maintain an immutable record of network access events complement NAC’s robust access control mechanisms. This integration is designed to enhance cybersecurity, establish trust, and improve accountability. Furthermore, this study is set to make a contribution to the academic literature and ongoing research in the realm of cybersecurity.

"Smart city technologies are being developed in many cities around the world to improve people's well-being, social life and economic prosperity with the help of digital transformation and technology. The need for cities to provide supportive services and inclusive technologies for all people is also increasing with the increasing urban population and smart city applications. Current technologies play an important role in the development of smart cities. Thanks to these technologies, data can be collected and analyzed in real time. As a result of the analysis, more effective and efficient decisions can be made about the city. In our country, the number of local and national companies should be increased in terms of smart cities and support should be received from local and national companies when establishing smart cities. Because it is expected that the applications that will collect important data throughout the country will belong to that country in terms of both hardware and software. This situation both contributes to the full independence of the country and the information that can create intelligence is kept within the country. Smart cities have three basic benefits: cost, time and sustainability. Consumption in a smart city can be optimized and savings can be achieved. Since problems can be detected early, costly repairs and maintenance are eliminated. In addition, since processes are automated, decision-making becomes easier with real-time data and time is saved. Sustainability is an important aspect of smart cities. When energy and water usage are optimized and waste is reduced, a more environmentally friendly city emerges. Priority should be given to accelerating urban transformation and innovation when establishing smart cities. Among the technologies required for smart cities, big data, artificial intelligence, internet of things, cyber security, blockchain, digital twin, cloud computing, etc. should be used effectively. Smart cities cannot be established only with technology companies, all stakeholders must come together for this. Citizen participation is extremely important in these processes. There are many challenges encountered when creating smart cities. These challenges are divided into administrative, managerial and technological groups. In this study, many points to be considered when creating smart cities are explained with examples and suggestions are given. It is evaluated that the study will be beneficial to smart city stakeholders."

Transformer is a kind of deep neural network that relies on the technique of self-attention and used initially in the field of natural language processing. Scientists use transformer for computer vision (CV) applications because of its good data representation capabilities. Transformer-based models yield similar performance or surpass other network architectures, including convolutional and recurrent neural networks, in a variety of visual benchmarks. In this work, we investigate the methods for video anomaly detection (VAD) using vision transformer models in the recent literature. The main topics we explore comprise vision transformers used in CV applications with a special focus on VAD methods leveraging transformer architecture. We also briefly present anomaly detection methods based on transformers. Additionally, we address the advantages, challenges and current limitations of the transformer architecture as well as potential solutions to address the technical challenges. In the concluding section of this study, we offer avenues for further investigation concerning the use of vision transformers in VAD tasks.

In recent years, Internet of Things (IoT) systems are used in Industrial Internet of Things (IIoT) systems due to their widespread use in industrial sectors, providing convenience to users in SCADA systems, like other domains. In addition to the diverse technological advancements discussed, the inherent symmetry within the network structures of SCADA systems utilized in the IIoT echoes a fundamental balance sought in digital frameworks. However, along with the advantages of IIoT systems, there are also disadvantages, one major drawback being their vulnerability to attacks. It has been observed that advanced methods such as artificial intelligence, unlike traditional detection techniques, are more successful at detecting attacks on IIoT systems used in SCADA systems. The proposed model was developed to detect cyberattacks on SCADA systems using machine learning and deep learning models. The SCADA network traffic consists of over 7 million rows and has a size of 627 MB. Attack network traffic refers to the type of traffic aimed at causing damage to the system. The attack traffic in this study includes five different attacks. Normal traffic is the type of traffic that facilitates the system’s usual communication. Prepared network traffic is not a different type of traffic. Prepared network traffic, as named, is the state of the traffic dataset that has been made ready for analysis with models. The prepared network traffic was examined using eight machine learning models, including the CART, Decision Tree, KNN, Logistic Regression, Naive Bayes, Random Forest, SVM, and XGBoost models, as well as seven deep learning models, namely, CNN, GRU, LSTM, MLP, RNN, CNN-LSTM, and LSTM-CNN. During the evaluation of the models, performance parameters such as the accuracy, F-score, precision, and recall were considered, and the results are presented accordingly. Upon examining the dataset with various models, the highest outcomes were achieved using the MLP model. The investigation utilizing the MLP model resulted in an accuracy of 99.95%, a precision of 99.63%, a recall of 99.49%, and an F-score of 99.56%. These values were obtained with a batch-size combination of 100 and 20 epochs. By addressing cyberattack detection in SCADA systems used in the IIoT within a big data environment, the study encompasses a multidisciplinary approach, touching upon cybersecurity, big data analytics, AI, information security, and IoT-related concerns, all of which are focal points within the scope of the journal. This breadth and depth of coverage make the study highly relevant and aligned with the diverse interests of the journal.

Criminal activities are a critical obstacle to socioeconomic development and must be controlled. However, human surveillance-based control methods are prone to error, raise legal concerns, and necessitate the development of more robust alternatives. This study aims to contribute to the development of strategies for reducing and preventing crime by ensuring the optimal allocation of police resources to locations at the right time. To achieve this goal, crime datasets from three of the most metropolitan cities in the USA—San Francisco, Chicago, and Philadelphia—were subjected to comprehensive preprocessing and exploratory data analysis. The analysis identified the most reliable and dangerous months, days, and hours in terms of the frequency of criminal incidents, the most common types of crimes, and the police districts with the highest crime rates. Crime-type prediction models were developed using machine learning algorithms, including XGBoost, CatBoost, random forest (RF), decision tree (DT), multilayer perceptron (MLP), K-nearest neighbors (KNN), Gaussian Naive Bayes (GNB), and logistic regression (LR). Additionally, time series analyses were conducted in 10, 22, and 22 different police districts for the three datasets, respectively, using deep learning models such as long short-term memory (LSTM) and bidirectional long short-term memory (BLSTM) and statistical methods such as Holt–Winters exponential smoothing (HWES), Prophet, and seasonal autoregressive integrated moving average (SARIMA). The primary objective was to accurately predict future high-crime hot spots. Furthermore, crime trend forecasts for the next 5 years were made using the best models, based on the lowest root-mean-squared error (RMSE) values obtained through statistical methods. By combining traditional machine learning methods, deep learning approaches, and statistical techniques, this study analyzed criminal incidents from various perspectives, including crime-type prediction, regional crime prediction, trend forecasting, and exploratory data analysis. The results obtained are expected to contribute to the development of proactive policing strategies.

The incorporation of artificial intelligence (AI) into neuroscience has the potential to significantly enhance our comprehension of brain function and facilitate more effective diagnosis and treatment of neurological disorders. Artificial intelligence (AI) techniques, particularly deep learning and machine learning, offer transformative solutions by improving the analysis of complex neural data, facilitating early diagnosis, and enabling personalized treatment approaches. A bibliometric analysis is a method that employs quantitative techniques for the examination of scientific literature, with the objective of identifying trends in research, evaluating the impact of influential studies, and mapping the networks of collaboration. In light of the accelerated growth and interdisciplinary scope of AI applications in neuroscience, a bibliometric analysis is vital for mapping the landscape, identifying pivotal contributions, and underscoring emerging areas of interest. This study aims to address this need by examining 1,208 studies published between 1983 and 2024 from the Web of Science database. The analysis reveals a notable surge in publications since the mid-2010s, with substantial advancements in neurological imaging, brain-computer interfaces (BCI), and the diagnosis and treatment of neurological diseases. The analysis underscores the pioneering role of countries such as the United States, China, and the United Kingdom in this field and highlights the prevalence of international collaboration. This study offers a comprehensive overview of the current state and future directions of AI applications in neuroscience, as well as an examination of the transformative potential of AI in advancing neurological research and healthcare. It is recommended that future research address the ethical issues, data privacy concerns, and interpretability of AI models in order to fully capitalize on the benefits of AI in neuroscience.

Smart Grids are an area where next-generation technologies, applications, architectures, and approaches are utilized. These grids involve equipping and managing electrical systems with information and communication technologies. Equipping and managing electrical systems with information and communication technologies, developing data-driven solutions, and integrating them with Internet of Things (IoT) applications are among the significant applications of Smart Grids. As dynamic systems, Smart Grids embody symmetrical principles in their utilization of next-generation technologies and approaches. The symmetrical integration of Wireless Sensor Networks (WSNs) and energy harvesting techniques not only enhances the resilience and reliability of Smart Grids but also ensures a balanced and harmonized energy management system. WSNs carry the potential to enhance various aspects of Smart Grids by offering energy efficiency, reliability, and cost-effective solutions. These networks find applications in various domains including power generation, distribution, monitoring, control management, measurement, demand response, pricing, fault detection, and power automation. Smart Grids hold a position among critical infrastructures, and without ensuring their cybersecurity, they can result in national security vulnerabilities, disruption of public order, loss of life, or significant economic damage. Therefore, developing security approaches against cyberattacks in Smart Grids is of paramount importance. This study examines the literature on “Cybersecurity with WSN in Smart Grids,” presenting a systematic review of applications, challenges, and standards. Our goal is to demonstrate how we can enhance cybersecurity in Smart Grids with research collected from various sources. In line with this goal, recommendations for future research in this field are provided, taking into account symmetrical principles.