ABSTRACT The resurgence of tick‐borne diseases necessitates predictive frameworks that integrate both high accuracy and ecological relevance. This study develops a comprehensive machine learning pipeline to forecast the occurrence of Ixodes ricinus , a principal tick vector in Europe, leveraging high‐dimensional climatic, environmental, and land‐use datasets. We assembled and cleaned regional occurrence datasets from the United Kingdom and wider European repositories, to create a harmonized database comprising over 27,000 verified occurance record. To represent local tick presence and reduce spatial bias, we transformed the point data into 20 km‐wide hexagonal grid cell duplicates. The framework that integrates hexagonal spatial binning, binary transformation, and spatially aware absence selection maintains a balanced 1:2 ratio to minimize sampling bias and spatial autocorrelation. Spatial interpretation was strengthened by adopting DBSCAN with geodesic (haversine) distance, which identifies density‐based clusters and noise points and avoids the Euclidean‐distance constraints inherent to K‐Means. Each observation was paired with dynamic environmental and land‐use variables, including monthly rainfall, NDVI, temperature, and annual land cover. Models were trained and evaluated using stratified fivefold cross‐validation and optimized through RandomizedSearchCV, ensuring efficient exploration of hyperparameter spaces. Comparative evaluation across Random Forest, CatBoost, Gradient Boosting, AdaBoost, and Support Vector Machine classifiers demonstrated high predictive accuracy, with Random Forest achieving an ROC–AUC of 0.941% and F1‐score of 0.882%. Incorporating spatial constraints and temporally aggregated features improved ecological realism and generalisation, addressing prior limitations in temporal dynamics and sampling bias. Feature importance analysis revealed NDVI, rainfall, and temperature as dominant predictors, aligning with ecological expectations. The study centres on tick occurrence, establishing a scalable and robust framework poised to support early warning systems and enable data‐driven surveillance of tick populations across Europe.

ABSTRACT Alzheimer's Disease (AD) is a progressive neurodegenerative disease diagnosed through cognitive impairment, and an early diagnosis is essential to improve treatment and care options. Current diagnostic approaches of AD, such as neuroimaging, cognitive assessments and biomarker research, are lengthy, vague and not sufficient to assess the early stages of AD. To address these problems, we introduced a novel deep learning model, ‘NeuroMixFormer’, which is based on a mixture‐of‐experts architecture for AD classification from MRI. The proposed multistage architecture employs a dynamic routing mechanism and four expert blocks per stage, each integrating dense connectivity with a spatial and channel attention module for feature extraction. To improve early feature learning, auxiliary classifiers are incorporated at intermediate stages of training. Evaluations on three datasets (ADNI, Mendeley and Kaggle Augmented Alzheimer's MRI) demonstrated the proposed model's superior performance over existing deep learning architectures and state‐of‐the‐art methods, achieving up to 99.48% accuracy on the Kaggle dataset, 90.28% on the Mendeley dataset and 99.86% on the ADNI dataset, respectively. Ablation studies confirmed the importance of dual‐attention mechanisms, and expert routing analysis showed clear specialisation patterns across AD stages, improving both classification accuracy and interpretability. These results underscore the effectiveness and generalisability of NeuroMixFormer in automated dementia detection, highlighting its potential to support early and precise AD diagnosis. However, the high computational cost and inference time associated with this high accuracy limit the practicality of the proposed approach in clinical settings.

Abstract The most prevalent type of cancer worldwide is mouth cancer. Around 2.5% of deaths are reported annually due to oral cancer in 2023. Early diagnosis of oral squamous cell carcinoma (OSCC), a prevalent oral cavity cancer, is essential for treating and recovering patients. A few computerized techniques exist but are focused on traditional machine learning methods, such as handcrafted features. In this work, we proposed a fully automated architecture based on Self‐Attention convolutional neural network and Residual Network information fusion and optimization. In the proposed framework, the augmentation process is performed on the training and testing samples, and then two developed deep models are trained. A self‐attention MobileNet‐V2 model is developed and trained using an augmented dataset. In parallel, a Self‐Attention DarkNet‐19 model is trained on the same dataset, whereas the hyperparameters have been initialized using the whale optimization algorithm (WOA). Features are extracted from the deeper layers of both models and fused using a canonical correlation analysis (CCA) approach. The CCA approach is further optimized using an improved WOA version named Quantum WOA that removes the irrelevant features and selects only important ones. The final selected features are classified using neural networks such as wide neural networks. The experimental process is performed on the augmented dataset that includes two sets: 100× and 400×. Using both sets, the proposed method obtained an accuracy of 98.7% and 96.3%. Comparison is conducted with a few state‐of‐the‐art (SOTA) techniques and shows a significant improvement in accuracy and precision rate.

Recent years have witnessed tremendous advancements in Al tools (e.g., ChatGPT, GPT-4, and Bard), driven by the growing power, reasoning, and efficiency of Large Language Models (LLMs). LLMs have been shown to excel in tasks ranging from poem writing and coding to essay generation and puzzle solving. Despite their proficiency in general queries, specialized tasks such as metaphor understanding and fake news detection often require finely tuned models, posing a comparison challenge with specialized Deep Learning (DL). We propose an assessment framework to compare task-specific intelligence with general-purpose LLMs on suicide and depression tendency identification. For this purpose, we trained two DL models on a suicide and depression detection dataset, followed by testing their performance on a test set. Afterward, the same test dataset is used to evaluate the performance of four LLMs (GPT-3.5, GPT-4, Google Bard, and MS Bing) using four classification metrics. The BERT-based DL model performed the best among all, with a testing accuracy of 94.61%, while GPT-4 was the runner-up with accuracy 92.5%. Results demonstrate that LLMs do not outperform the specialized DL models but are able to achieve comparable performance, making them a decent option for downstream tasks without specialized training. However, LLMs outperformed specialized models on the reduced dataset.

Individuals’ use of eHealth services has increased significantly. However, the recent pandemic of coronavirus disease 2019 (COVID-19) has resulted in a significant reallocation of health resources and support. This study investigated the impact of service quality dimensions on individuals’ continued trust in eHealth during COVID-19. A decision-making trial and evaluation laboratory (DEMATEL) approach was used to identify and analyse the causal relationships between service quality dimensions and individuals’ continued trust in eHealth services. A total of 134 eHealth users (78 males and 56 females; aged 29–61 years) responded to the DEMATEL questionnaire. The results showed a variation in the impact of service quality factors on individuals’ continued trust in eHealth services. This study found three core factors (responsiveness, assurance and tangibility) that influence individuals’ continued trust in eHealth services. Other secondary factors (e.g. content quality, reliability, efficiency and hedonic benefits) were found to be primarily influenced by the core factors. The identified relationships in this study can aid the decision-making process of healthcare providers and increase the efficiency of healthcare delivery.

Differences in individuals’ psychological and cognitive characteristics have been always found to play a significant role in influencing our behavior and preferences. While a number of studies have identified the impact of these characteristics on individuals’ visual design preferences, understanding how emotional intelligence (EI) would influence this process is yet to be explored. This study investigated the link between individuals’ EI dimensions (eg, emotionality, self-control, sociability, and well-being) and their eye movement behavior in an attempt to build a prediction model for visual design preferences. A total of 136 participants took part in this study. The feature selection and prediction of EI and eye movement data were performed using the genetic search method in conjunction with the bagging method. The results showed that participants high in self-control and emotionality exhibited different eye movement behaviors when performing five visual selection tasks. The prediction results (93.87% accuracy) revealed that specific eye parameters can predict the link between certain EI dimensions and preferences for visual design. This study adds new insights into human–computer interaction, EI, and rational choice theories. The findings also encourage researchers and designers to consider EI in the development of intelligent and adaptive systems.

Abstract Although several studies have looked at the relationship between emotional characteristics and viewing behaviour, understanding how emotional intelligence (EI) contributes to individuals’ viewing behaviour is not clearly understood. This study examined the viewing behaviour of people (74 male and 80 female) with specific EI profiles while viewing five facial expressions. An eye-tracking methodology was employed to examine individuals’ viewing behaviour in relation to their EI. We compared the performance of different machine learning algorithms on the eye-movement parameters of participants to predict their EI profiles. The results revealed that EI profiles of individuals high in self-control, emotionality, and sociability responded differently to the visual stimuli. The prediction results of these EI profiles achieved 94.97% accuracy. The findings are unique in that they provide a new understanding of how eye-movements can be used in the prediction of EI. The findings also contribute to the current understanding of the relationship between EI and emotional expressions, thereby adding to an emerging stream of research that is of interest to researchers and psychologists in human–computer interaction, individual emotion, and information processing.

Social media sites, such as Twitter, provide the means for users to share their stories, feelings, and health conditions during the disease course. Anemia, the most common type of blood disorder, is recognized as a major public health problem all over the world. Yet very few studies have explored the potential of recognizing anemia from online posts. This study proposed a novel mechanism for recognizing anemia based on the associations between disease symptoms and patients' emotions posted on the Twitter platform. We used k-means and Latent Dirichlet Allocation (LDA) algorithms to group similar tweets and to identify hidden disease topics. Both disease emotions and symptoms were mapped using the Apriori algorithm. The proposed approach was evaluated using a number of classifiers. A higher prediction accuracy of 98.96 % was achieved using Sequential Minimal Optimization (SMO). The results revealed that fear and sadness emotions are dominant among anemic patients. The proposed mechanism is the first of its kind to diagnose anemia using textual information posted on social media sites. It can advance the development of intelligent health monitoring systems and clinical decision-support systems.

Abstract Anxiety and self-regulation are the most common problems among the college student population. There are few attempts found in the literature to promote the development of students’ cognitive and metacognitive abilities in online learning environments. In addition, mechanisms for overcoming or reducing individuals’ anxiety in a computer-mediated environment is yet to be fully characterized. This study was conducted to investigate the potential of integrating the concept of flow into the design of a Metalearner (MTL) to help reduce anxiety and increase self-regulation among students. The design of MTL was based on the development of adaptive strategies to balance between the challenge of the task and user skills. A total of 260 participants were asked to use the system and respond to an online questionnaire that asked about flow antecedents, experience, and consequences. The structural model results showed that incorporating flow into the design of MTL can help reduce anxiety and improve self-regulation among students. Our findings can be used to enrich students’ online learning experience and inform designers and developers of learning systems about the importance of regulating task complexity according to the challenge/skills balance. This would help learners to process the presented information meaningfully and to make the inferences necessary for understanding the learning content.

Advances in mobile phone technology and the growth of associated networks have been phenomenal over the last decade. Therefore, they have been the focus of much academic research, driven by commercial and end-user demands for increasingly faster technology. The most recent generation of mobile network technology is the fifth generation (5G). 5G networks are expected to launch across the world by 2020 and to work with existing 3G and 4G technologies to provide extreme speed despite being limited to wireless technologies. An alternative network, Y-Communication (Y-Comm), proposes to integrate the current wired and wireless networks, attempting to achieve the main service requirements of 5G by converging the existing networks and providing an improved service anywhere at any time. Quality of service (QoS), vertical handover, and security are some of the technical concerns resulting from this heterogeneity. In addition, it is believed that the Y-Comm convergence will have a greater influence on security than was the case with the previous long-term evolution (LTE) 4G networks and with future 5G networks. The purpose of this research is to satisfy the security recommendations for 5G mobile networks. This research provides a policy-based security management system, ensuring that end-user devices cannot be used as weapons or tools of attack, for example, IP spoofing and man-in-the-middle (MITM) attacks. The results are promising, with a low disconnection rate of less than 4% and 7%. This shows the system to be robust and reliable.

With modern smart phones and powerful mobile devices, Mobile apps provide many advantages to the community but it has also grown the demand for online availability and accessibility. Cloud computing is provided to be widely adopted for several applications in mobile devices. However, there are many advantages and disadvantages of using mobile applications and cloud computing. This paper focuses in providing an overview of mobile cloud computing advantages, disadvantages. The paper discusses the importance of mobile cloud applications and highlights the mobile cloud computing open challenges