Search Results (56)

Log anomaly detection in modern distributed systems is challenging. Anomalous behaviors are rare. Manual labeling is expensive. Session boundaries are often set by fixed heuristics before model training. This fixed-boundary assumption is problematic because segmentation errors propagate into representation learning and cannot be corrected during optimization. To address this, this paper proposes BASN (Boundary-Aware Sessionization Network), a boundary-aware contrastive learning framework that jointly learns session boundaries and anomaly representations using a differentiable soft-reset mechanism. BASN does not treat sessionization as a separate step. Instead, it predicts boundary probabilities from event semantics and temporal gaps, then modulates end-to-end session-state updates. The session representations are optimized with self-supervised contrastive learning, enabling effective zero-shot anomaly detection and few-shot adaptation. Experiments on four benchmark datasets (BGL, HDFS, OpenStack, SSH) show strong zero-shot performance (area under the receiver operating characteristic curve, AUROC 0.935–0.975) and boundary alignment with expert-validated proxy segmentation (boundary F1 0.825–0.877). Comparative gains over baselines are reported in the article after bibliography correction, baseline verification, and expanded statistical analysis. BASN is also computationally efficient, requiring less than 10 ms per session on a Graphics Processing Unit (GPU) and less than 45 ms on a Central Processing Unit (CPU). This is compatible with real-time inference needs in the evaluated settings. However, cross-system transfer AUROC (0.735–0.812) remains below in-domain performance. Domain-specific adaptation is still needed for deployment in environments that differ greatly from the training domain. Full article

To address challenges in complex group decision-making (GDM)—specifically preference fuzziness, intricate subgroup segmentation, and non-cooperative behavior—this study proposes an adaptive consensus model based on probabilistic linguistic term sets (PLTSs). By integrating fuzzy C-means (FCM) clustering with a Gaussian mixture model (GMM), a fuzzy Gaussian mixture model (FGMM) is constructed to achieve soft segmentation of expert preference distributions. On this basis, an adaptive consensus feedback mechanism is developed, which dynamically integrates interactive and automated adjustment strategies via multi-level consensus thresholds, thereby balancing decision efficiency and quality. To identify and control non-cooperative behaviors, a cooperation index and a three-tier management strategy, which incorporates intra-group negotiation, weight penalties and an exit-delegation mechanism, were introduced. In the case of strategic decision-making of new energy vehicles (NEV), after four rounds of feedback iterations, the group consensus level increased from the initial 0.316 to 0.804, reaching the preset threshold and verifying the effectiveness of the consensus mechanism. Compared with the existing literature methods, the framework in this paper achieves more comprehensive integration and innovation in four aspects: preference expression, clustering mechanism, consensus feedback and behavior management. Full article

Deep learning has shown promise in tabular data modeling, yet challenges such as feature heterogeneity, sparse interactions, and expert prediction collapse remain unresolved. To address these issues, we propose DETTab (Diversity-Enhanced Tabular Experts), a framework that integrates feature gating, multi-expert fusion, and structure-aware regularization. DETTab first employs a Feature Gating Encoder to perform soft selection over input fields, enhanced by a Field Decorrelation Loss to promote embedding diversity. A Feature Interaction Encoder is then used to capture high-order dependencies among features via multi-head self-attention. Finally, a Multi-View Expert Fusion Module aggregates predictions from multiple experts through a soft routing mechanism, guided by an Expert Diversity Loss to mitigate prediction collapse and improve training stability. Extensive experiments on public tabular datasets demonstrate that DETTab achieves consistent improvements in predictive performance and training robustness across different settings, particularly in alleviating expert convergence collapse, thereby validating its effectiveness for tabular learning. Full article

Background/Objectives: The overuse and misuse of antibiotics contribute to antimicrobial resistance (AMR) globally. The appropriateness of antibiotic prescribing at the primary healthcare (PHC) level must be urgently addressed to reduce high levels of inappropriate antibiotic prescribing and associated AMR. This study aimed to develop quality indicators, based on the World Health Organization (WHO)’s Access, Watch, Reserve (AWaRe) guidance, to assess the appropriateness and quality regarding antibiotic prescribing in public PHC settings in South Africa. Methods: Potential indicators were identified from indicators developed by City St George’s, University of London (SGUL); a review of AWaRe-based indicators; and the results from point prevalence surveys at PHC clinics in South Africa. The indicators were developed using the RAND/UCLA Appropriateness Method. In Round 1, 12 experts individually rated 78 indicators for clarity and appropriateness. In Round 2, 10 experts rated 89 indicators for appropriateness and feasibility during an interactive online meeting. Results: The final set had 61/89 indicators (68.5%) that were rated both appropriate and feasible with agreement. Dental infections (9/9; 100%) alongside skin and soft tissue infections (11/13; 84.6%) had the highest percentage of indicators that were rated appropriate and feasible with agreement. Lower urinary tract infections (6/11; 54.5%) and general (4/8; 50%) categories had the lowest percentage of indicators rated appropriate and feasible with agreement. Conclusions: The process proved valuable in developing potential indicators for use in future antimicrobial stewardship programmes to improve antibiotic prescribing in public sector PHC facilities in South Africa and beyond. Full article

We study control synthesis under Signal Temporal Logic (STL) specifications for driving scenarios where strict rule satisfaction is not always feasible and human experts exhibit context-dependent flexibility. We represent such behavior using robustness slackness—learned rule-wise lower bounds on STL robustness—and introduce sub-goals that encode intermediate intent in the state/output space (e.g., lane-level waypoints). Prior learning-based MPC–STL methods typically infer slackness with VAE priors and plug it into MPC, but these priors can underrepresent multimodal and rare yet valid expert behaviors and do not explicitly model intermediate intent. We propose a diffusion-guided MPC–STL framework that jointly learns slackness and sub-goals from demonstrations and integrates both into STL-constrained MPC. A conditional diffusion model generates pairs of (rule-wise slackness, sub-goal) conditioned on features from the ego vehicle, surrounding traffic, and road context. At run time, a few denoising steps produce samples for the current situation; slackness values define soft STL margins, while sub-goals shape the MPC objective via a terminal (optionally stage) cost, enabling context-dependent trade-offs between rule relaxation and task completion. In closed-loop simulations on held-out highD track-driving scenarios, our method improves task success and yields more realistic lane-changing behavior compared to imitation-learning baselines and MPC–STL variants using CVAE slackness or strict rule enforcement, while remaining computationally tractable for receding-horizon MPC in our experimental setting. Full article

The Citarum River Basin (DAS Citarum) in Indonesia faces significant challenges in waste management, necessitating a circular economy-based approach to reduce land-based pollution, which is critical for achieving the sustainability goals of the blue economy in the basin. This study addresses the complexity and inherent uncertainty in decision-making processes related to this challenge by developing a novel hybrid model, namely the Weighted Fuzzy N-Soft Set combined with the COmbinative Distance-based Assessment (CODAS) method. The model synergistically integrates the weighted 10R strategies in the circular economy, obtained via the Analytical Hierarchy Process (AHP), the capability of Fuzzy N-Soft Sets to represent uncertainty granularly, and the robust ranking mechanism of CODAS. Applied to a case study covering 16 types of waste in the Citarum River Basin, the model effectively processes expert assessments that are ambiguous regarding the 10R criteria. The results indicate that single-use plastics, particularly plastic bags (HDPE), styrofoam, transparent plastic sheets (PP), and plastic cups (PP), are the top priorities for intervention, in line with the high AHP weights for upstream strategies such as Refuse (0.2664) and Rethink (0.2361). Comparative analysis with alternative models, namely Fuzzy N-Soft Set-CODAS, Weighted Fuzzy N-Soft Set with row-column sum ranking, and Weighted Fuzzy N-Soft Set-TOPSIS, confirms the superiority of the proposed hybrid model in producing ecologically rational priorities, free from purely economic value biases. Further sensitivity analysis shows that the model remains highly robust across various weighting scenarios. This study concludes that the WFN-SS-CODAS framework provides a rigorous, data-driven, and reliable decision support tool for translating circular economy principles into actionable waste management priorities, directly supporting the restoration and sustainability goals of the blue economy in river basins. The findings suggest that targeting the high-priority waste types identified by the model addresses the dominant fraction of riverine pollution, indicating the potential for significant waste volume reduction. This research was conducted to directly contribute to achieving multiple targets under SDG 6 (Clean Water and Sanitation), SDG 12 (Responsible Consumption and Production), and SDG 14 (Life Below Water). Full article

Patients requiring breast reconstruction following radiation therapy or prior failed autologous breast reconstruction (ABR) or implant-based breast reconstruction (IBBR) represent a challenging cohort and often present with compromised vascularity, scarred anatomy, and subsequent increased rates of complications. In this review, we discuss microsurgical strategies designed to optimize donor tissue in these challenging clinical scenarios, including the use of stacked or bipedicled flaps, and the utility of intraoperative indocyanine green angiography. We also review approaches to alternate recipient vessel selection in the suboptimal chest, and we address specific strategies for the revision setting, like soft tissue support and hybrid reconstruction with ABR and IBBR. By synthesizing the current literature and expert experience, this narrative review provides a practical framework for microsurgeons managing complex breast reconstruction in higher-risk patients. Full article

This article explores experts’ perspectives on the most important soft skills for entrepreneurial success in the Western Balkans (WB) and identifies effective educational and workplace practices to foster these skills. Using a qualitative Delphi study supported by a literature review, the research gathered and synthesized opinions from 20 experts representing Serbia, Albania, North Macedonia, and Bosnia and Herzegovina. Findings show that communication, adaptability, flexibility, teamwork, and critical thinking are essential for business success, while leadership, emotional intelligence, problem-solving, and teamwork are considered most vital for future entrepreneurs. Experts emphasized that group projects, specialized courses, and blended learning approaches are effective in educational settings, while workplace skill development benefits from training programs, mentoring, active communication, and openness to feedback. This study provides region-specific insights into skill-building strategies for young entrepreneurs, addressing a key research gap. By integrating expert consensus with evidence-based practices, the article offers a framework for educators, policymakers, institutions, and businesses to strengthen entrepreneurship education and workforce readiness across the WB region. Full article

University course scheduling is a kind of timetable problem and can be mathematically formulated as an integer linear programming problem. Essentially, a university course scheduling problem is an optimization problem that aims at most efficiently minimizing a cost function according to a set of constraints. The huge searching space for the course scheduling problem means a long time will be needed to find the optimal solution. Therefore, some studies have used soft computing approaches to solve course scheduling problems in order to reduce the searching space. However, in order to use soft computing approaches to solve university course scheduling problems, we may need to design algorithms and conduct numerous experiments to achieve maximum efficiency. Thus, in this study, instead of employing soft computing methods, we propose a SWI-PROLOG-based expert system to solve the course scheduling problem. An experiment was conducted using real-world data from a department at a national university in southern Taiwan. During the experiment, each teacher in the department chose five preferential time slots. The experimental results have shown that about 99% of courses were scheduled in teachers’ five preferential time slots with an acceptable computational time of executing SWI-PROLOG (127 milliseconds on a regular personal computer). This study has thus provided a framework for solving course scheduling problems using an expert system. This would be the main contribution of this study. Full article

The reliability of decision-making algorithms within soft set theory is fundamentally constrained by their underlying membership functions. Traditional binary approaches overlook the implicit connections between the attributes a candidate possesses and those it lacks—connections that can be inferred from the wider candidate pool. To address this core challenge, this paper puts forward the Relational Membership Value Calculation (RMVC), an algorithmic framework whose core is a fine-grained relational membership function. Our approach moves beyond binary logic to capture these nuanced interrelationships. We provide a rigorous theoretical analysis of the proposed algorithm, including its computational complexity and robustness, which is validated through a comprehensive sensitivity analysis. Crucially, a comparative analysis using the Gini Index quantitatively demonstrates that our method provides significantly higher granularity and discriminatory power on a representative case study. The RMVC is implemented as an open-source Python program, providing a foundational tool to enhance the reasoning capabilities of AI-driven decision support and expert systems. Full article

Most public seismic recordings, sampled at hundreds of Hz, tend to be unlabelled, i.e., not catalogued, mainly because of the sheer volume of samples and the amount of time needed by experts to confidently label detected events. This is especially challenging for very low signal-to-noise ratio microseismic events that characterise landslides during rock and soil mass displacement. Whilst numerous supervised machine learning models have been proposed to classify landslide events, they rely on a large amount of labelled datasets. Therefore, there is an urgent need to develop tools to effectively automate the data-labelling process from a small set of labelled samples. In this paper, we propose a semi-supervised method for labelling of signals recorded by seismometers that can reduce the time and expertise needed to create fully annotated datasets. The proposed Siamese network approach learns best class-exemplar anchors, leveraging learned similarity between these anchor embeddings and unlabelled signals. Classification is performed via soft-labelling and thresholding instead of hard class boundaries. Furthermore, network output explainability is used to explain misclassifications and we demonstrate the effect of anchors on performance, via ablation studies. The proposed approach classifies four landslide classes, namely earthquakes, micro-quakes, rockfall and anthropogenic noise, demonstrating good agreement with manually detected events while requiring few training data to be effective, hence reducing the time needed for labelling and updating models. Full article

Existing risk assessment approaches for soft rock tunnels in fault-fractured zones typically employ single weighting schemes, inadequately integrate subjective and objective weights, and fail to define clear risk. This study proposes a risk-grading methodology that integrates an enhanced game theoretic weight-balancing algorithm with an optimized cloud entropy extension cloud model. Initially, a comprehensive indicator system encompassing geological (surrounding rock grade, groundwater conditions, fault thickness, dip, and strike), design (excavation cross-section shape, excavation span, and tunnel cross-sectional area), and support (support stiffness, support installation timing, and construction step length) parameters is established. Subjective weights obtained via the analytic hierarchy process (AHP) are combined with objective weights calculated using the entropy, coefficient of variation, and CRITIC methods and subsequently balanced through a game theoretic approach to mitigate bias and reconcile expert judgment with data objectivity. Subsequently, the optimized cloud entropy extension cloud algorithm quantifies the fuzzy relationships between indicators and risk levels, yielding a cloud association evaluation matrix for precise classification. A case study of a representative soft rock tunnel in a fault-fractured zone validates this method’s enhanced accuracy, stability, and rationality, offering a robust tool for risk management and design decision making in complex geological settings. Full article

Accurate prediction of the Normalized Difference Vegetation Index (NDVI) is crucial for regional ecological management and precision decision-making. Existing methodologies often rely on smoothed NDVI data as ground truth, overlooking uncertainties inherent in data acquisition and processing. Fuzzy time series (FTS) prediction models based on the Fuzzy C-Means (FCM) clustering algorithm address some of these uncertainties by enabling soft partitioning through membership functions. However, the method remains limited by its reliance on expert experience in setting fuzzy parameters, which introduces uncertainty in the definition of fuzzy intervals and negatively affects prediction performance. To overcome these limitations, this study enhances the interval type-2 fuzzy clustering time series (IT2-FCM-FTS) model by developing a pixel-level time series forecasting framework, optimizing fuzzy interval divisions, and extending the model from unidimensional to spatial time series forecasting. Experimental results from 2021 to 2023 demonstrate that the proposed model outperforms both the Autoregressive Integrated Moving Average (ARIMA) and conventional FCM-FTS models, achieving the lowest RMSE (0.0624), MAE (0.0437), and SEM (0.000209) in 2021. Predictive analysis indicates a general ecological improvement in the Aksu region (Xinjiang, China), with persistent growth areas comprising 61.12% of the total and persistent decline areas accounting for 2.6%. In conclusion, this study presents an improved fuzzy model for NDVI time series prediction, providing valuable insights into regional desertification prevention and ecological strategy formulation. Full article

Liver metastases from soft tissue sarcoma (STS) (excluding gastrointestinal stromal tumors) are rare and more commonly arise from retroperitoneal and intra-abdominal primary sites. Chemotherapy remains the mainstay of treatment for disseminated disease, but its effectiveness is limited and patients typically have a dismal prognosis with short survival. However, when metastases are confined to the liver (without pulmonary involvement), some patients may benefit from local techniques, either surgical or nonsurgical, that can provide long periods of disease-free survival. Due to the rarity of STS, especially with liver metastases, and the heterogeneity of histologies and biological behavior, there is a lack of standardized treatment guidelines and universally accepted criteria for this specific setting. To fill this gap, a multidisciplinary working group of experts in sarcoma and liver surgery reviewed the literature and available evidence and developed a set of clinical recommendations to be voted and discussed in the I Ibero-American Consensus on the Management of Metastatic Sarcoma, held at the III Spanish-Portuguese Update Meeting on the Treatment of Sarcomas in May 2024. Herein, the voting results of this meeting and the resulting consensus recommendations are presented, and their applicability, strengths, and limitations are discussed. Full article

This paper presents N-bipolar soft expert (N-BSE) sets, a novel framework designed to enhance multi-attribute group decision-making (MAGDM) by incorporating expert input, bipolarity, and non-binary evaluations. Existing MAGDM approaches often lack the ability to simultaneously integrate positive and negative assessments, especially in nuanced, multi-valued evaluation spaces. The proposed N-BSE model addresses this limitation by offering a comprehensive, mathematically rigorous structure for decision-making (DM). Fundamental operations of the N-BSE model are defined and analyzed, ensuring its theoretical consistency and applicability. To demonstrate its practical utility, the N-BSE model is applied to a general case study on sustainable energy solutions, illustrating its effectiveness in handling complex DM scenarios. An algorithm is proposed to streamline the DM process, enabling systematic and transparent identification of optimal alternatives. Additionally, a comparative analysis emphasizes the advantages of the N-BSE model over existing MAGDM frameworks, highlighting its capacity to integrate diverse expert opinions, evaluate both positive and negative attributes, and support multi-valued assessments. By bridging the gap between theoretical development and practical application, this paper contributes to advancing DM methodologies. Full article