Search Results (9,349)

This study investigates the relationship between community commercial center spaces and elderly behavior, focusing on governance mechanisms that shape these spaces and their impact on enhancing elderly life and the community environment. Field research was conducted in the ‘Guohe 1000’ community commercial project in Shanghai, targeting individuals aged 60 and above with independent mobility, including wheelchair users. Through behavioral observation and interviews, both individual and group activities were examined, emphasizing behavioral patterns, spatial domains, and social interactions. Findings reveal that factors such as gender, age, and social networks are positively correlated with the spatial development of community commercial centers. To foster elderly-friendly environments, improvements are needed in utilization balance, secondary activity spaces, age-sensitive design, and operational management. The paper’s novelty lies in two aspects: first, it broadens research into community commercial centers by tracing the construction process of spatial forms; second, it applies environmental behaviorism and environmental gerontology frameworks to integrate individual and collective elderly behaviors into systematic data collection and quantitative analysis. Together, these insights contribute to more inclusive strategies for designing and managing community commercial complexes that support active aging and enhance urban social sustainability. Full article

Federated learning (FL) enables distributed model training across sensor-equipped edge devices while preserving data privacy. However, its performance is often hindered by statistical heterogeneity among clients and system heterogeneity in dynamic wireless networks. To address these challenges, we propose TCS-FEEL, a topology-aware client selection framework that jointly considers user distribution, device-to-device (D2D) communication, and statistical similarity of client data. The proposed approach integrates randomized client sampling with an adaptive tree-based communication structure, where user devices not only participate in local model training but also serve as relays to exploit efficient D2D transmission. TCS-FEEL is particularly suited for sensor-driven edge intelligence scenarios such as autonomous driving, smart city monitoring, and the Industrial IoT, where real-time performance and efficient resource utilization are crucial. Extensive experiments on MNIST and CIFAR-10 under various non-IID data distributions and mobility settings demonstrated that TCS-FEEL consistently reduced the number of training rounds and shortened per-round wall-clock time compared with existing baselines while maintaining model accuracy. These results highlight that integrating topology control with client selection provides an effective solution for accelerating privacy-preserving and resource-efficient FL in dynamic, sensor-rich edge environments. Full article

Effective spam detection systems are essential in online social media networks (OSNs) and cybersecurity, and they directly influence the quality of decision-making pertaining to security. With today’s digital communications, unsolicited spam degrades user experiences and threatens platform security. Machine learning-based spam detection systems offer an automated defense. Despite their effectiveness, such methods are frequently hindered by the “black box” problem, an interpretability deficiency that constrains their deployment in security applications, which, in order to comprehend the rationale of classification processes, is crucial for efficient threat evaluation and response strategies. However, their effectiveness hinges on selecting an optimal feature subset. To address these issues, we propose a lightweight, explainable spam detection model that integrates a nature-inspired optimizer. The approach employs clean data with data preprocessing and feature selection using a swarm-based, nature-inspired meta-heuristic Rüppell’s Fox Optimization (RFO) algorithm. To the best of our knowledge, this is the first time the algorithm has been adapted to the field of cybersecurity. The resulting minimal feature set is used to train a supervised classifier that achieves high detection rates and accuracy with respect to spam accounts. For the interpretation of model predictions, Shapley values are computed and illustrated through swarm and summary charts. The proposed system was empirically assessed using two datasets, achieving accuracies of 99.10%, 98.77%, 96.57%, and 92.24% on Dataset 1 using RFO with DT, KNN, AdaBoost, and LR and 98.94%, 98.67%, 95.04%, and 94.52% on Dataset 2, respectively. The results validate the efficacy of the suggested approach, providing an accurate and understandable model for spam account identification. This study represents notable progress in the field, offering a thorough and dependable resolution for spam account detection issues. Full article

Background: The Fontan procedure is a palliative surgery for patients with single-ventricle congenital heart disease (CHD), but it is associated with postoperative and long-term mortality and morbidity. Accurate, individualized risk stratification remains a challenge with traditional models. This study aimed to develop and validate a deep learning (DL) model to predict postoperative mortality after the Fontan procedure and to identify key predictive factors. Methods: We retrospectively analysed data from 230 patients who underwent the Fontan procedure between 2010 and 2024. A Deep Neural Network (DNN) model was developed using comprehensive preoperative, intraoperative, and postoperative clinical, biochemical, and hemodynamic variables. The dataset was split using five-fold cross-validation, with 80% for training and 20% for testing in each fold. The Synthetic Minority Over-sampling Technique (SMOTE) was used to fix class imbalance. Model performance was evaluated using five-fold stratified cross-validation. We assessed accuracy, precision, recall, F1-score, and Area Under the Receiver Operating Characteristic Curve (AUC-ROC). SHapley Additive exPlanations (SHAP) analysis was employed to enhance model interpretability and identify the importance of features. A user-friendly clinical application interface was developed using Streamlit. This study was reported in accordance with the TRIPOD + AI reporting guidelines. Results: The DNN model demonstrated superior performance in predicting postoperative mortality, achieving an overall accuracy of 91.5% (95% CI: 87.2–94.8%), precision of 83.3% (95% CI: 76.5–89.1%), recall (sensitivity) of 90.9% (95% CI: 85.2–95.1%), specificity of 92.5% (95% CI: 88.3–95.7%), F1-score of 87.0% (95% CI: 82.1–91.3%), and an AUC-ROC of 0.94 (95% CI: 0.88–0.99). SHAP analysis identified key predictors of mortality, such as pulmonary artery pressure, ventricular end-diastolic pressure, preoperative BNP levels, and severity of AV valve regurgitation. The Streamlit application offered a user-friendly interface for personalized risk evaluation. Conclusions: A deep learning model that incorporates detailed clinical data can precisely forecast postoperative mortality in patients undergoing Fontan surgeries. This AI-based method, combined with interpretability techniques, provides a valuable tool for personalized risk assessment. It has the potential to improve preoperative counseling, optimize perioperative care, and enhance patient outcomes. However, additional external validation is needed to verify its broader applicability and clinical usefulness. Full article

In post-disaster scenarios, Unmanned Aerial Vehicles (UAVs) acting as Mobile Aerial Base Stations (MABSs) offer a flexible means of restoring communication for isolated user equipment (UE) when conventional infrastructure is unavailable. More broadly, clustering is a fundamental tool for organizing spatially distributed entities in wireless, IoT, and sensor networks. However, static algorithms such as Affinity Propagation Clustering (APC) often fail to generalize across diverse environments and user densities. This study introduces a hybrid clustering framework that dynamically selects between APC and density-based clustering (DBSCAN), guided by a neural classifier trained on spatial distribution features. The chosen centroids then seed a Genetic Algorithm (GA) that evolves UAV trajectories under multiple performance indicators, including coverage, capacity, and path efficiency. Simulation results demonstrate that the hybrid clustering approach improves the adaptability and effectiveness of UAV deployments by learning context-aware clustering strategies. Beyond UAV-assisted disaster recovery, the proposed framework illustrates how intelligent clustering selection can enhance performance in heterogeneous, real-time applications such as IoT connectivity, smart city monitoring, and large-scale sensor coordination. Full article

This paper presents a mobile application designed to support emergency operators during indoor missions, where Global Positioning System (GPS) often fails. The system combines a wearable waist-mounted Inertial Measurement Unit (IMU) with a network of pre-installed Radio Frequency Identification (RFID) tags, enabling robust, real-time geo-referenced tracking of both personnel and critical Points of Interest (PoIs), such as resources and threats. Development was guided by interviews and surveys with emergency professionals, ensuring the tool addresses real operational needs. Key features include dynamic updates of operator positions and nearby hazards, enabled by an Indoor Positioning System (IPS) that fuses IMU and RFID data to improve accuracy in position and heading estimation. The application also offers a user-friendly Human–Environment Interface (HEI) displaying information on a spatially referenced map. By merging advanced technology with expert feedback, this system enhances safety and coordination in critical scenarios, offering a promising solution for indoor navigation and Situational Awareness (SA) in emergency response. Full article

The bidirectional power flows and time-varying characteristics generated by distributed photovoltaic integration into low-voltage distribution networks pose accuracy and fairness challenges to traditional line loss allocation methods. Existing methods, based on unidirectional power flow assumptions, are unable to quantify the true contributions of PV nodes and ignore the multi-dimensional value attributes of photovoltaics. Against this background, following the principle of “who caused the incremental part of loss, who is responsible for it“, this paper proposes a hierarchical line loss allocation model for low-voltage distribution networks with distributed photovoltaics. The first layer employs an enhanced marginal loss coefficient method to allocate the baseline line losses without PV integration to original distribution network users. The second layer utilizes spatiotemporal weighted Shapley values to quantify the marginal contributions of PV nodes to line loss variations, while establishing a multi-dimensional PV value correction system based on local consumption rate, spatiotemporal matching degree, and voltage support capability, and transforms the multi-dimensional PV values into economic incentive signals through an adaptive Softmax weighting algorithm. Finally, simulation analysis validates the effectiveness of the proposed line loss allocation method. Full article

Cyberbullying and hate-driven behavior on social media have become increasingly prevalent, posing serious psychological and social risks. This study proposes a machine learning-based approach to detect hate-driven content by integrating temporal and behavioral features—such as message frequency, interaction duration, and user activity patterns—alongside traditional text-based features. Furthermore, we extend our evaluation to include recent neural network architectures, namely ALBERT and BiLSTM, enabling a more robust representation of semantic and sequential patterns. Building on our previous research presented at JNIC-2024, we conduct a comparative evaluation of multiple classification algorithms using both existing and engineered datasets. The results show that incorporating non-textual features significantly improves detection accuracy and robustness. This work contributes to the development of intelligent cyberbullying detection systems and highlights the importance of behavioral context in online threat analysis. Full article

This paper analyzes a discrete-time single-server retrial queue with preemptive service, Bernoulli arrivals, and adaptive retrial times, tailored to telecommunications systems. In call centers, the model captures caller retries and priority interruptions, while in cellular networks, it represents user channel access attempts with preemption for emergency calls. Using a Markov chain framework, we derive the stationary distribution, establish a stability condition, and compute performance metrics, including the mean number of retrying callers or users and orbit size probabilities. The model incorporates a novel retrial time adaptation probability, reflecting dynamic retry behaviors in telecommunications. Numerical results demonstrate the impact of arrival rates, preemption probabilities, and retrial adaptations on system performance, offering insights for optimizing call center staffing and cellular network protocols. Applications to slotted ALOHA and TDMA systems highlight the model’s practical relevance. Full article

Detecting anomalies in network traffic is a central task in cybersecurity and digital infrastructure management. Traditional approaches rely on statistical models, rule-based systems, or machine learning techniques to identify deviations from expected patterns, but often face limitations in generalization across domains. This study proposes a cross-domain data enrichment framework that integrates behavioral embeddings with network traffic features through adversarial autoencoders. Each network traffic record is paired with the most similar behavioral profile embedding from user web activity data (Charles dataset) using cosine similarity, thereby providing contextual enrichment for anomaly detection. The proposed system comprises (i) behavioral profile clustering via autoencoder embeddings and (ii) cross-domain latent alignment through adversarial autoencoders, with a discriminator to enable feature fusion. A Deep Feedforward Neural Network trained on the enriched feature space achieves 97.17% accuracy, 96.95% precision, 97.34% recall, and 97.14% F1-score, with stable cross-validation performance (99.79% average accuracy across folds). Behavioral clustering quality is supported by a silhouette score of 0.86 and a Davies–Bouldin index of 0.57. To assess robustness and transferability, the framework was evaluated on the UNSW-NB15 and the CIC-IDS2017 datasets, where results confirmed consistent performance and reliability when compared to traffic-only baselines. This supports the feasibility of cross-domain alignment and shows that adversarial training enables stable feature integration without evidence of overfitting or memorization. Full article

In the field of natural language processing, depression forecasting from social media has gained extensive attention, as platforms like X (formerly Twitter) offer real-time user-generated content that can reflect psychological states. Common approaches typically rely on static text analysis, which overlooks how users’ emotions change over time. To address this limitation, we propose a temporal modeling approach that applies deep learning models to capture both textual and temporal patterns in users’ tweet histories. Our experiments evaluated LSTM networks and Transformer architectures with pretrained embeddings on a dataset of over 3 million tweets. We demonstrate that incorporating temporal features significantly improved performance in depression forecasting. The best setting, which combines Llama 2 embeddings with personalized time-difference features, achieved 99.4% accuracy and 0.996 AUC. These results highlight the importance of modeling temporal dynamics for improving depression forecasting and suggest that personalized temporal signals provide capabilities beyond static content analysis. Full article

Compared with other fossil energy sources, natural gas is characterized by compressibility, low energy density, high storage costs, and imbalanced usage. Natural gas pipeline supply systems possess unique attributes such as closed transportation and a highly integrated upstream, midstream, and downstream structure. Moreover, pipelines are almost the only economical means of onshore natural gas transportation. Given that the upstream of the pipeline features multi-entity and multi-channel supply including natural gas, coal-to-gas, and LNG vaporized gas, while the downstream presents a competitive landscape with multi-market and multi-user segments (e.g., urban residents, factories, power plants, and vehicles), there is an urgent social demand for non-discriminatory and fair opening of natural gas pipeline network infrastructure to third-party entities. However, after the fair opening of natural gas pipeline networks, the original “point-to-point” transaction model will be replaced by market-driven behaviors, making the verification and allocation of gas transmission capacity a key operational issue. Currently, neither pipeline operators nor government regulatory authorities have issued corresponding rules, regulations, or evaluation plans. To address this, this paper proposes a multi-dimensional quantitative evaluation model based on the Analytic Hierarchy Process (AHP), integrating both commercial and technical indicators. The model comprehensively considers six indicators: pipeline transportation fees, pipeline gas line pack, maximum gas storage capacity, pipeline pressure drop, energy consumption, and user satisfaction and constructs a quantitative evaluation system. Through the consistency check of the judgment matrix (CR = 0.06213 < 0.1), the weights of the respective indicators are determined as follows: 0.2584, 0.2054, 0.1419, 0.1166, 0.1419, and 0.1357. The specific score of each indicator is determined based on the deviation between each evaluation indicator and the theoretical optimal value under different gas volume allocation schemes. Combined with the weight proportion, the total score of each gas volume allocation scheme is finally calculated, thereby obtaining the recommended gas volume allocation scheme. The evaluation model was applied to a practical pipeline project. The evaluation results show that the AHP-based evaluation model can effectively quantify the advantages and disadvantages of different gas volume allocation schemes. Notably, the gas volume allocation scheme under normal operating conditions is not the optimal one; instead, it ranks last according to the scores, with a score 0.7 points lower than that of the optimal scheme. In addition, to facilitate rapid decision-making for gas volume allocation schemes, this paper designs a program using HTML and develops a gas volume allocation evaluation program with JavaScript based on the established model. This self-developed program has the function of automatically generating scheme scores once the proposed gas volume allocation for each station is input, providing a decision support tool for pipeline operators, shippers, and regulatory authorities. The evaluation model provides a theoretical and methodological basis for the dynamic optimization of natural gas pipeline gas volume allocation schemes under the fair opening model. It is expected to, on the one hand, provide a reference for transactions between pipeline network companies and shippers, and on the other hand, offer insights for regulatory authorities to further formulate detailed and fair gas transmission capacity transaction methods. Full article

This paper proposes a scheme for enhancing covert communication in cognitive radio networks (CRNs) using a reconfigurable intelligent surface (RIS), which ensures that transmissions by secondary users (SUs) remains statistically undetectable by adversaries (e.g., wardens like Willie). However, there exist stringent challenges in CRNs due to the dual constraints of avoiding detection and preventing harmful interference to primary users (PUs). Leveraging the RIS’s ability to dynamically reconfigure the wireless propagation environment, our scheme jointly optimizes the SU’s transmit power, communication block length, and RIS’s passive beamforming (phase shifts) to maximize the effective covert throughput (ECT) under rigorous covertness constraints quantified by detection error probability or relative entropy while strictly adhering to PU interference limits. Crucially, the RIS configuration is explicitly designed to simultaneously enhance signal quality at the legitimate SU receiver and degrade signal quality at the warden, thereby relaxing the inherent trade-off between covertness and throughput imposed by the fundamental square root law. Furthermore, we analyze the impact of unequal transmit prior probabilities (UTPPs), demonstrating their superiority over equal priors (ETPPs) in flexibly balancing throughput and covertness, and extend the framework to practical scenarios with Poisson packet arrivals typical of IoT networks. Extensive results confirm that RIS assistance significantly boosts ECT compared to non-RIS baselines and establishes the RIS as a key enabler for secure and spectrally efficient next-generation cognitive networks. Full article

The increasing availability of User-Generated Content during large-scale events is transforming spectators into active co-creators of live narratives while simultaneously introducing challenges in managing heterogeneous sources, ensuring content quality, and orchestrating distributed infrastructures. A trial was conducted to evaluate automated orchestration, media enrichment, and real-time quality assessment in a live sporting scenario. A key innovation of this work is the use of a cloud-native architecture based on Kubernetes, enabling dynamic and scalable integration of smartphone streams and remote production tools into a unified workflow. The system also included advanced cognitive services, such as a Video Quality Probe for estimating perceived visual quality and an AI Engine based on YOLO models for detection and recognition of runners and bib numbers. Together, these components enable a fully automated workflow for live production, combining real-time analysis and quality monitoring, capabilities that previously required manual or offline processing. The results demonstrated consistently high Mean Opinion Score (MOS) values above 3 72.92% of the time, confirming acceptable perceived quality under real network conditions, while the AI Engine achieved strong performance with a Precision of 93.6% and Recall of 80.4%. Full article

Collaborative filtering (CF) continues to be a fundamental approach in recommendation systems for providing users with personalized suggestions. However, such kind of recommender systems are prone to performance issues when faced with noisy, inconsistent, or deliberately manipulated user ratings. Although Generative Adversarial Networks (GANs) offer promising solutions to capture complex user-item interactions in these CF situations, many existing GAN-based methods assume uniform reliability across all ratings, reducing their effectiveness under uncertain conditions. To overcome this challenge, this paper presents DST-AttentiveGAN to introduce a confidence-aware adversarial framework specifically designed to denoise inconsistent ratings in collaborative filtering scenarios. The proposed approach employs Dempster-Shafer Theory (DST) to compute confidence scores by aggregating diverse behavioral indicators, such as item popularity, user activity, and rating variance. These scores guide both components of the GAN architecture in which the generator incorporates a cross-attention mechanism to highlight trustworthy features, while the discriminator uses DST-based confidence to evaluate the credibility of input ratings. Training is carried out using a stabilized Wasserstein GAN objective that promotes both robustness and convergence efficiency. Experimental results in three benchmark data sets show that DST-AttentiveGAN consistently surpasses conventional GAN-based models, delivering more accurate and reliable recommendations under conditions of uncertainty. Full article