Quantum Machine Learning (QML) leverages the transformative power of quantum computing to explore a broad range of applications, including optimization, data analysis, and complex problem‐solving. Central to this study is the using of an innovative intrusion detection system leveraging QML models, with a preference for Quantum Neural Network (QNN) architectures for classification tasks. The inherent advantages of QNNs, notably their parallel processing capabilities facilitated by quantum computers and the exploitation of quantum superposition and parallelism, are elucidated. These attributes empower QNNs to execute certain classification tasks expediently and with heightened efficiency. Empirical validation is conducted through the deployment and testing of a QNN‐based intrusion detection system, employing a subset of the CIC‐DDoS 2019 dataset. Notably, despite employing a reduced feature set, the QNN‐based system exhibits remarkable classification accuracy, achieving a commendable rate of 92.63%. Moreover, the study advocates for the utilization of quantum computing libraries such as Qiskit, facilitating QNN training on local machines or quantum simulators. The findings underscore the efficacy of a QNN‐based intrusion detection system in attaining superior classification accuracy when confronted with large‐scale training datasets. However, it is imperative to acknowledge the constraints imposed by the limited number of qubits available on local machines and simulators.

A transformative hybrid approach is proposed, combining Quantum Machine Learning (QML) with both traditional and meta-heuristic feature selection algorithms to overcome the complexities and limitations of conventional credit card fraud detection methods. In this study, advanced data balancing techniques such as SMOTE-ENN (Synthetic Minority Over-sampling Technique–Edited Nearest Neighbors) and Random Under Sampler (RUS) are employed on the imbalanced European Cardholder Dataset to address class imbalance and enhance model resilience. Core feature selection algorithms—both traditional methods like K-Best and meta-heuristic techniques including Particle Swarm Optimization (PSO), Ant Colony Optimization (ACO), and Atom Search Optimization (ASO)—are systematically evaluated, each paired with the Variational Quantum Classifier (VQC) for classification. Remarkably, the PSO+VQC combination achieves an accuracy rate of 94.54%, underscoring the efficacy of integrating meta-heuristic algorithms with VQC to manage complex, high-dimensional data in fraud detection. These findings highlight QML and meta-heuristic algorithms’ potential to surpass conventional methods, delivering superior accuracy and efficiency in critical, data-intensive applications such as financial fraud detection.

Doğal afetler ve insan kaynaklı tahribatlar sebebi ile zarar gören tarihi yapılar ilk haline en yakın bir görünüme kavuşacak şekilde restore edilmektedir. Aynı zamanda mimarlık öğrencileri bu yapılar üzerinde restorasyon süreçlerini öğrenmek için araştırmalar yapmakta ve bilgi toplamaktadır. Fakat erişebilirliği kısıtlayan pandemi, hastalık, tehlike ve olumsuz hava gibi koşullar öğrencilerin bu yapılara erişmelerini engellemektedir. Bu çalışma, mimarlık öğrencilerinin restorasyon derslerinde tarihi yapıları inceleme sırasında yaşadıkları erişim zorluklarını Sanal Gerçeklik (SG) teknolojisinden faydalanarak gidermeyi ve özellikle öğrencilere uzaktan erişimle, restorasyon dersindeki ölçme tekniklerini sanal ortamda uygulamalı olarak öğretmeyi amaçlamaktadır. Çalışma kapsamında, restorasyon öncesi veri toplama işlemi tamamlanmış olan Tarihi Safranbolu Tabakhane Binası gerçek ölçüleri ile modellenmiş ve sanal gerçeklik ortamına aktarılmıştır. Öğrencilerin bu binada etkileşimli olarak beraber çalışabilmeleri için geleneksel ölçme araçları olan metre, su hortumu, şakül ve yazı tahtasının sanal versiyonları sisteme entegre edilmiştir. Böylece mimari restorasyon alanında literatürde yer alan diğer çalışmalardan farklı olarak, birden fazla öğrencinin fiziksel olarak farklı ortamlarda olmalarına rağmen aynı ortamdaymış gibi eş zamanlı olarak birbirleriyle etkileşimde bulunduğu, sesli ve görsel iletişim kurabildikleri sanal bir eğitim ortamı oluşturulmuştur. Kullanıcı deneyi sonrası yapılan anketlerde, öğrencilerin geliştirilen yönteme genellikle olumlu bakış açısı sergiledikleri ve anketteki toplam 14 ifadenin puanlamasında sistemin %88,62’lik bir puan ortalamasına ulaştığı gözlemlenmiştir. Sonuçlar, sanal gerçekliğin yüz yüze eğitime ulaşımın olmadığı durumlarda, restorasyon derslerine yönelik uygulamalı bir eğitim fırsatı sunduğunu ortaya koymaktadır.

Phase picking in seismology is the first step of signal processing and locating a seismic event. At the beginning of seismological research, it is straightforward to manually pick P- and S-wave arrival times because the number of seismic stations is relatively small. Recently, instrumentation has improved, and the gap is lower than before. This widespread instrumentation creates problems for users who still have to pick manually, especially in national networks such as those in Türkiye. We used pre-trained deep learning models, trained on different seismic datasets, to estimate P- and S-wave arrival times for earthquakes in the Marmara Region, NW Türkiye. We compared these times with manual readings collected from the Ministry of Interior, the Disaster and Emergency Management Presidency, and the Earthquake Department (AFAD). The results indicate that PhaseNet produces arrival-time estimates that are largely consistent with expert manual readings, demonstrating its potential to substantially reduce analyst workload in large-scale seismic monitoring systems. The mean absolute error ranges from 6 to 14 seconds, and the number of total picks varies between 75,000 and 140,000 for both the P-wave and S-wave. This project has been supported by the Scientific and Technological Research Council of Türkiye (TÜBİTAK) under grant 124E294, and the results have been shared on the website (https://quakemlab.sakarya.edu.tr)

The rapid expansion of the Internet of Things (IoT) has increased the risk of large-scale Distributed Denial-of-Service (DDoS) attacks. In high-availability IoT environments, the operational costs of false positives and false negatives are asymmetric, whereas conventional deep learning models usually optimize static accuracy-based objectives. To address this, we propose CDRL-QNN, a cost-aware and chaos-driven reinforcement learning quantum neural network framework in which a parameterized quantum circuit serves as the action-value function approximator within a Deep Q-Network (DQN) agent. The framework incorporates asymmetric operational penalties through both the reward function and sample-wise weighted Bellman optimization, while a logistic-map-based deterministic perturbation mechanism is used to promote exploration under constrained quantum-circuit training conditions. Evaluated on a computationally constrained balanced subset of the CIC-DDoS2019 dataset, the proposed framework reduced false negatives from 49 to 33 without increasing false positives, improving recall from 0.9673 to 0.9780 and F1-score from 0.9738 to 0.9793 while lowering operational cost. These findings suggest that hybrid quantum representations can be integrated into cost-sensitive reinforcement learning pipelines for IoT intrusion detection under constrained experimental conditions.