46870 research outputs found
Sort by
The Effect of Grid Search-Based Hyperparameter Tuning on Multinomial Naïve Bayes Model Performance in Classifying Stress Levels Among University Students
No full textStress among university students is an increasingly prominent issue, given the high levels of academic, social, and emotional pressures they encounter. Early identification of stress indicators is essential to prevent the development of more severe mental health issues. This study investigates the effect of hyperparameter tuning using grid search on the performance of a Multinomial Naïve Bayes (MNB) model for classifying student stress levels. The dataset was collected via web scraping from social media platforms where students express their emotions and experiences. Text preprocessing and feature extraction were performed using the TFIDF method, and stress levels were categorized into three classes: No Stress (0), Mild Stress (1), and High Stress (2). The MNB model was trained and evaluated using k-fold crossvalidation, with performance assessed via accuracy, precision, recall, and F1-score. Results indicate that grid search-based hyperparameter tuning significantly improved classification performance, increasing accuracy from 72.3% to 79.6%. These findings highlight that even simple models like Naïve Bayes can benefit substantially from systematic hyperparameter optimization, particularly in the context of stress detection from student-generated text data
What if passengers experienced cyberattacks in a level 5 automated and connected shuttle? Impact of cyberattack type and contextual factors on perceived safety:A scenario-based study
Full text linkConnected and Automated shuttles are increasingly accessible on public roads. These vehicles, will be characterized by extensive connectivity, enabling them to make driving decisions and share information. This connectivity will increase exposure to malicious attacks, which could lead to accidents, thereby causing psychological or physical harm to passengers. However, the effects of such attacks on passengers remain understudied. This study explored the effect of the type of cyberattack (target, consequences, alert message source) and contextual factors (social presence, traffic density) on perceived safety. A total of 420 participants were exposed to scenarios in which a character had a ride in an automated shuttle under different conditions and rated their level of perceived safety. Results highlighted that the target of the attack, consequences of the attack, alert message source and traffic density influenced perceived safety, except for social presence. Perceived safety scores were the greatest when the target of the attack was the passenger’s mobile phone, when the consequences of the attack were mild, when the alert message was provided by the vehicle’s antivirus, when the traffic density was low, with few vulnerable road users crossing the road. Cluster analyses highlighted three passenger profiles: alarmed, pondered and relaxed. This cluster formation was linked to sociodemographic characteristics, anxiety levels as car passengers and subjective cybersecurity knowledge. This research contributes to understanding passengers’ reactions to cyberattacks in public transport. Future studies should focus on cyberattacks’ impact on trust, acceptance, and on the specific information to provide passengers with during such events.</p
Review of Gender equality in the professional workplace: pioneering radical action for women’s empowerment
No full textGender equality in the professional workplace: pioneering radical action forwomen’s empowerment, by Karen R. Vollum-Dix, Leeds: Emerald Publishing Limited,2025, 184 pp., ISBN 978 1 835 49157 7, £80.00 (hardcover), ISBN 978 1 8354 9156 0(e-book
The First X-Ray Polarimetry of GRS 1739–278 Reveals Its Rapidly Spinning Black Hole
Full text linkWe present a joint spectropolarimetric analysis of the black hole X-ray binary GRS 1739–278 during its 2025 mini-outburst, using simultaneous observations from IXPE and NuSTAR. The IXPE data show a polarization degree (PD) of PD = (2.3% ± 0.4%) and polarization angle (PA) of PA = 62° ± 5° in the 2−8 keV range. The model-independent analysis reveals that PD increases from ∼2% at 2 keV to ∼10% in the 6–8 keV band, while PA remains stable across the IXPE band within statistical uncertainties. Broadband spectral modeling of the combined IXPE and NuSTAR datasets shows that hard Comptonization contributes negligibly in this soft-state observation, while a substantial reflected component is required in addition to the thermal disk emission. We then model the IXPE Stokes spectra using the kynbbrr model. The best fits indicate that high-spin configurations enhance the contribution of the reflected returning radiation, which dominates the observed polarization properties. From the kynbbrr modeling, we infer an extreme black hole spin of a 0.994 0.003 0.004 = + and a system inclination of i 54 4 = 8 ° ° + ° . Owing to the large contribution from the returning radiation, the observed polarization direction is nearly parallel to the projected system axis, the position angle of which is predicted at 58° ± 4°. Our results demonstrate that X-ray polarimetry, combined with broadband spectroscopy, directly probes the geometry and relativistic effects in accretion disks around stellar-mass black holes
Material circularity in the UK’s foundation industries
Full text linkAs the UK advances toward its 2050 net-zero target, moving beyond energy efficiency to comprehensive resource management is essential. This study evaluates how circular economy principles can reshape the use of aluminium, lead, steel, glass, and paper in the UK's foundation industries. Using a circularity index, we assess current performance and identify improvement pathways, quantifying the potential of enhanced recycling and material reuse. Results show that circularity strategies alone could reduce emissions by 42% and energy use by 17%, with reductions rising to 69% and 56% respectively, when combined with best-practice energy intensities. While aluminium and steel offer the largest gains, barriers remain for lead, glass, and paper due to technical and quality constraints. These findings highlight the opportunities and the complexities of industrial circularity, providing evidence to guide policymakers and industry leaders in accelerating the transition to a more sustainable and resource-efficient economy.</p
An ultra-precise fast fourier transform – part 2
Full text linkAn earlier paper [1] describes the application of Prism Signal Processing to the Fast Fourier Transform (FFT), which generates high precision estimates of the frequency, amplitude and phase of spectral peaks. The current paper describes improvements to the Prism FFT. These include: a simplified calculation; applicability to shorter FFT window lengths (e.g. 1024 samples); improved performance against the Cramer Rao Lower Bound (CRLB), typically delivering root mean square errors of 2.2σ for frequency and 1.5σ, for amplitude and phase, where σ is defined as the square root of the corresponding CRLB. The method also delivers significantly reduced spectral leakage. MATLAB code implementing the Prism FFT is provided as an appendix
Curating a Corpus:A Three-Phase Model
No full textWe describe in this Research Note processes and protocols for curating a corpus of texts for analysis, from collection of data to readiness for analysis. We offer example case studies working with texts in different genres and languages, and using different tools, to illustrate general principles for corpus curation. Rather than a comprehensive guide for researchers interested in corpus linguistics methods, we offer a conversational starting point, supplementing our overview of three phases (collection, cleaning, and pre-processing) with authentic experiences from our own diverse research. Further, we reflect on the pedagogical implications associated with corpus linguistics, as well as the challenges and next steps in corpus curation and analysis in the age of generative AI. Our experiences show how common curation phases can be applied to different studies and contexts, and the considerations that arise when doing so.</p
Distributed Coordination for Heterogeneous Non-Terrestrial Networks
Full text linkTo achieve global coverage and ubiquitous connectivity, the non-terrestrial network (NTN) has been regarded as a key enabler in the sixth generation (6G) network, which includes uncrewed aerial vehicles (UAVs), high-altitude platforms (HAPs), and satellites. Since the unique characteristics of various NTN platforms strongly affect their implementation and lead to a highly dynamic and heterogeneous NTN scenario, achieving distributed coordination remains an important research direction. However, the explicit and systematic analysis of the individual layers' challenges and corresponding distributed coordination solutions in heterogeneous NTNs has not been proposed yet. Therefore, in this article, we summarize the unique characteristics of each NTN platform, identify communication challenges within individual layers, and propose potential delay-tolerant or delay-sensitive coordinated solutions accordingly. We further analyse the feasibility of leveraging multi-agent deep reinforcement learning (MADRL) algorithms to achieve the proposed coordinated solutions. Finally, we present a case study of the joint scheduling and trajectory optimization problem in heterogeneous NTN, where a two-timescale multi-agent deep deterministic policy gradient (TTS-MADDPG) algorithm is developed to validate the effectiveness of distributed coordination