Başkent University

We present a phenomenological study of rotating, charged black holes (BHs) in Einstein gravity coupled to a traceless (conformal) matter sector formed by ModMax nonlinear electrodynamics and a Kalb–Ramond (KR) two-form that spontaneously breaks local Lorentz symmetry. Starting from a family of obtained static, Schwarzschild-like solutions with a traceless KR sector, we construct the stationary, axisymmetric counterpart via the Newman–Janis algorithm. The resulting Newman-Kerr-like metric depends on four intrinsic parameters: the electric charge Q, the ModMax nonlinearity γ, the Lorentz-violation amplitude ℓ and the spin a. We analyse horizon structure and separatrices in parameter space, derive the null geodesic equations and obtain the photon capture boundary that defines the BH shadow. Using ray-tracing, we compute shadow silhouettes and a suite of shadow observables (areal radius, characteristic radius Rs, distortion δ, oblateness D) and show how γ and ℓ produce qualitatively distinct effects: γ acts as a screening factor for the electromagnetic imprint, while ℓ introduces angular-dependent metric rescalings that deform shadow shape beyond simple size rescaling. We confront model predictions with EHT angular-radius measurements for M87∗ and Sgr A∗ and derive conservative bounds on the combinations of (Q,γ,ℓ,a). Our results identify an effective charge combination Qeff2≃e−γQ2/(1−ℓ)2 and demonstrate that modest Qeff remains compatible with current EHT images while large Qeff is progressively disfavoured.

This study examines explosive price dynamics consistent with bubble-like behavior in clean and renewable energy ETFs, index, and equities compared to traditional energy counterparts, using the Generalized Supremum Augmented Dickey-Fuller (GSADF) test across four subperiods: Pre-Paris Agreement, Paris Agreement, COVID-19, and Russia-Ukraine War. Using daily price data for clean energy assets (ICLN, QCLN, CELS, ENPH, AEIS) and traditional energy assets (VDE, IXC, XLE, USEG), the analysis reveals pronounced prevalence of explosive price episodes in clean energy markets, with 76.55% of explosive price episodes in the Pre-agreement period and 72.66% during COVID-19, consistent with speculative dynamics. Traditional energy assets exhibit greater resilience, with limited explosive episodes, notably only USEG during the Russia-Ukraine War. The novel segmented subperiod approach, combined with GSADF’s precise bubble date-stamping of explosive episodes, distinguishes this study by offering granular insights into sector-specific vulnerabilities, unlike prior broader analyses. The findings indicate that selected clean energy instruments exhibited more frequent explosive price episodes during crisis periods than their traditional energy counterparts, a pattern that may inform risk assessment and further research relevant to SDG 7 (Affordable and Clean Energy) and SDG 13 (Climate Action) but does not by itself warrant specific investment or policy recommendations.

Purpose This study investigated the screening-related psychometric properties and screening accuracy of a single-item Visual Analog Scale (VAS) for identifying teachers at risk of voice disorders. Methods The VAS was developed as a rapid screening indicator of voice disorder risk through expert review and pilot testing. The final version of the VAS was structured as a 100 mm visual analog scale, scored on a 0–100 range, where higher scores indicate a worse perceived voice condition. It was administered alongside the Turkish Voice Disorders Screening Index (SIVD-TR) to 196 teachers working in Ankara, Turkey, to examine convergent validity evidence. After 7–10 days, a subsample of 30 participants completed the VAS again to assess test–retest reliability. In statistical analyzes, the psychometric properties of the VAS related to screening were examined within the scope of predictive utility, including test–retest reliability, known-groups discrimination between teachers with and without voice disorders, convergent relationships with the SIVD-TR, and risk classification via logistic regression. Receiver operating characteristic (ROC) curve analysis was used to evaluate screening accuracy. Results Test–retest reliability of the VAS was excellent (ICC = .946). VAS scores showed a strong correlation with SIVD-TR scores, supporting convergent screening validity. ROC analysis demonstrated excellent screening performance (AUC = .983). An optimal cutoff value of 37.5 yielded a sensitivity of .979 and a specificity of .919 for identifying teachers at risk of voice disorders. Logistic regression demonstrated that VAS scores significantly forecasted the risk of voice disorders. Teachers at risk had significantly higher VAS scores than those not at risk, supporting discriminative validity. Conclusion The single-item VAS showed strong evidence for being a reliable screening tool and was very accurate, making it useful for quickly identifying teachers who might be at risk for voice disorders.

Objective The growing prevalence of electric scooters (e-scooters) has contributed to changes in mobility patterns in various urban contexts; however, their use raises significant safety and behavioral concerns, particularly in regions with inadequate traffic infrastructure. Method This study explores the motivations, perceptions, and attitudes underlying shared e-scooter use through an exploratory approach employing Latent Dirichlet Allocation (LDA), a topic modeling technique. Responses from 57 participants, collected via 15 open-ended questions, were analyzed to uncover latent themes in user discourse. Results A three-topic LDA solution, selected mainly on the basis of interpretability, yielded three topics: (1) Perceived Risk and Reasons for Use, which reflected a tension between safety concerns and perceived convenience or enjoyment; (2) Positioning within the Traffic System, highlighting e-scooters as substitutes for walking or biking and their integration into urban traffic systems; and (3) User experience, highlighting a blend of utilitarian efficiency and recreational motivations. Conclusions The findings suggest that participants discussed e-scooters in relation to both practical and recreational benefits, while also expressing safety-related concerns. The study contributes to the limited psychological literature on e-scooter usage and offers exploratory insights into road safety and responsible micromobility behavior, while acknowledging that larger, and more representative samples are needed before drawing broader conclusions regarding policy and infrastructure improvements.

This study introduces a hybrid deep learning framework that integrates convolutional neural networks (CNNs) with Kolmogorov–Arnold networks (KAN) for fine-grained classification of eight Lactarius species using a curated dataset of 1.614 images. The proposed CNN–KAN architecture significantly outperforms seven state-of-the-art baseline models, ConvNeXt-Small, EfficientNetV2-Small, MobileNetV3-Small, RegNetY-400MF, ResNet-50, SqueezeNet1.1M, and ViT-Small, across all evaluation metrics. The model achieved an accuracy of 0.9877, F1-score 0.9877, precision 0.9879, sensitivity 0.9877, specificity 0.9982, MCC 0.9855, and AUC 0.9994, representing improvements of approximately 1–3