Purpose Due to its computational rigor, evidenced by strong mathematical foundations and demonstrably tractable results, the Weighted Influence Non-linear Gauge System (WINGS) method has attracted interest in applications where problems can be modeled as systems with components that have interdependent relationships. Given the ambiguities inherent in eliciting judgments within WINGS, integrating fuzzy set theory presents a straightforward agenda, leading to the development of the fuzzy WINGS method. Despite its advantages, methodological gaps remain, including the assumption of a neutral decision-maker and the potentially counterintuitive metric used for priority ranking. This work aims to bridge these gaps by developing a fuzzy WINGS variant called fWINGS-PR. This variant considers the degree of optimism among decision-makers while promoting an interpretable ranking mechanism using a compromise solution on an evaluation based on L2-norm. Design/methodology/approach In our proposed fWINGS-PR, the translation of triangular fuzzy numbers into real numbers associates an index of optimism, serving as a behavioral tuning knob that allows decision-makers to position themselves on either the safe or hopeful side. In addition, we use two metrics as inputs to the ranking process, generating positive and negative ideal solutions and evaluating the distance of each component from these points. A proposed metric derives a compromise priority ranking of the components. To demonstrate the efficacy of fWINGS-PR, a case study was implemented to evaluate the 10 adoption enablers of woodworking machinery that aid teaching-learning processes. Findings The results of the case study reveal that the innovativeness of the design, worker comfort, ease of use and portability are the top-priority adoption enablers, while aesthetic design and durability are given lower priority. These findings suggest that the design team must prioritize affordability, ease of use, safety and adaptability to institutional constraints over visual appeal and long-term durability. Both sensitivity and comparative analyses demonstrate the stability of the results in response to changes in the model parameter and consistency with other well-established methods, which highlight the practical value of the proposed approach. Originality/value This work introduces a novel variant of the fuzzy WINGS method that integrates decision-makers’ optimism levels and facilitates a more interpretable priority ranking mechanism based on the distances of system components to two ideal solutions.

Purpose Sensors are vital for autonomous driving, but they are susceptible to weather noise. To address the performance degradation of point cloud object detection in foggy conditions, this paper aims to propose an efficient dynamic pillar-based point cloud detection network tailored for foggy traffic scenarios. Design/methodology/approach The point cloud data are first pillarized, and then the model adaptively adjusts the feature extraction process through DynamicPillarVFE to significantly enhance the processing capability for sparse and irregular point cloud data. Second, a bird's eye view (BEV) backbone network based on residual connections is used to effectively mitigate the gradient vanishing problem in deep networks. In addition, this paper introduces Triplet Attention (TA) in the feature enhancement part, which can focus on important features more comprehensively and suppress noise interference. Findings In this paper, we mixed a small portion of the Multifog KITTI dataset on top of the KITTI dataset. Experimental results demonstrate that under moderate difficulty 3D evaluation metrics, our method achieves accuracy improvements of 5.03%, 6.66% and 7.78% for cars, pedestrians and cyclists, respectively, compared to the PointPillars baseline, significantly enhancing point cloud object detection performance in foggy conditions. Furthermore, the DPSF-Net architecture achieves an inference speed of 32.36 ms per frame, fully meeting the real-time processing requirements for autonomous driving applications. Originality/value This method is more economical to simulate the influence of sensors under foggy conditions, and through the training of mixed data sets, the network model can better cope with foggy interference.

Purpose In recent years, road trips have become a popular travel mode in China. However, existing attraction recommendation methods often overlook conflicting preferences among tourists and key factors specific to road trips, resulting in suboptimal recommendations. To address this issue, we propose a knowledge-enhanced non-compensatory group decision-making approach to improve the accuracy of road trip attraction recommendations. Design/methodology/approach First, our approach constructs a comprehensive tourism knowledge graph by integrating information about both tourists and attractions. Second, an 11-dimensional modeling framework is proposed to better portray tourist preferences and attraction characteristics in the road trip context. Finally, the non-compensatory group decision-making algorithm, Elimination Et Choix Traduisant la REalité III (ELECTRE-III), is applied to model each tourist's preferences within the group and rank attractions for the tourist group. Findings Experimental results demonstrate the effectiveness of the proposed method in road trip attraction recommendation. Compared to existing approaches, our method maintains a recommendation failure rate below 15% across varying levels of conflict rates and group sizes, consistently outperforming all baseline methods. Statistical analysis further confirms that the non-compensatory mechanism effectively identifies attractions that align with the collective preferences of tourist groups. Originality/value This paper proposes a novel method for recommending attractions to road trip tourist groups by integrating non-compensatory group decision-making with knowledge integration. The approach effectively incorporates individual preferences while maintaining a low recommendation failure rate, thereby enhancing both recommendation performance and overall tourist satisfaction.

Purpose Graph neural networks (GNNs) have been widely applied to various recommendation scenarios due to their outstanding performance in capturing and modelling hidden relationships. However, as the performance of a GNN heavily relies on the graph structure, the structure bias can lead to significant unfairness problems in a GNN-based recommender system across different user groups. Furthermore, prior studies have shown that sensitive attributes of nodes can be revealed through the graph structure. In other words, even if a GNN-based recommendation system does not explicitly disclose a user’s sensitive information, it can still be inferred from the imbalance in users’ connections within the graph. This article therefore addresses the unfairness arising from such structural biases by proposing a novel model designed to mitigate these biases in GNN-based recommendation. Here, the fairness requirements refer to preventing the exposure of sensitive attributes during user modeling and avoiding biased recommendations across different demographic groups. Design/methodology/approach To address this issue, we propose a novel fairness-aware graph neural network model, EqStruct-GNN, which aims to mitigate inter-group unfairness at the structural level. The model is composed of three key modules: (1) the Adaptive Behavior Modeling Module, which computes fairness-aware node behavior probability distributions; (2) the Fairness-Aware Structure Construction Module, which builds graph structures based on these distributions and (3) the Fair Representation Learning Module, which generates unbiased node embeddings. Together, these modules enable EqStruct-GNN to construct a fairness-aware graph that reduces structural bias and improves fairness across sensitive groups. Findings Experimental results demonstrate that the proposed method achieves strong recommendation performance while ensuring fairness across multiple datasets. Furthermore, when applied to various backbone models, EqStruct-GNN not only significantly improves fairness metrics but also maintains or even enhances recommendation accuracy. These results highlight the generalization capability and practical effectiveness of the proposed framework. Originality/value Unlike existing methods that focus on representation-level fairness, our model tackles structural bias at its root by building a fairness-aware debiased graph through a novel graph construction mechanism. This structure enables the learning of fair and high-quality node representations by effectively mitigating structural bias while preserving critical interaction information. In addition, the proposed approach is universally adaptable and can be seamlessly applied to various GNN architectures, demonstrating strong generality and practical applicability.

Purpose Educational assignments are pivotal for student learning, yet educators face the dual burden of creating meaningful questions and providing timely, individualized feedback. This study introduces IntelliA, a generative AI-powered platform designed to address these challenges by holistically automating the entire assignment lifecycle, from pedagogically aware question generation to concept-centric evaluation, thereby alleviating instructor workload and enhancing student learning outcomes. Design/methodology/approach IntelliA’s core innovation is a synergistic architecture that operationalizes pedagogical principles directly into the AI-driven assignment workflow. For question generation, its pedagogical primitive-driven system allows educators to design assessments at a conceptual level – by selecting instructional goals – rather than manually crafting prompts. For evaluation, a novel three-stage, rubric-driven methodology is designed to produce pedagogically valuable feedback, grounding its supportive guidance in objective, evidence-based scoring. These subsystems are unified by a unified quality augmentation module, creating a self-reinforcing loop that continuously refines the system’s pedagogical alignment. This architecture’s ability to enhance both educational quality and operational efficiency was rigorously validated through a semester-long university deployment, assessed via a multi-faceted framework of instructor interviews, student surveys and cross-model benchmarking. Findings The results demonstrate that IntelliA significantly reduces assignment creation time for instructors while maintaining high pedagogical quality in generated questions. Student surveys revealed high satisfaction with the grading accuracy and the personalized, actionable feedback, which they reported as instrumental to their learning. The findings confirm that an architecturally sophisticated AI system can successfully move beyond simple automation to enhance both operational efficiency and educational quality. Research limitations/implications Our study’s validation was conducted within a single university STEM course, which limits the generalizability of our findings. Future research should expand the library of pedagogical primitives to diverse domains (e.g. humanities and K-12) and validate their effectiveness in these new contexts. Additionally, a more rigorous, controlled experiment is needed to quantitatively measure efficiency gains in terms of time-on-task and cognitive load. This work signals a shift in AI-in-education research, moving from single-task automation toward designing synergistic, architecturally aware systems. Practical implications IntelliA offers a practical, scalable solution for educators to significantly reduce the workload associated with assignment creation and feedback personalization. By automating routine tasks, the platform empowers instructors to shift their focus from content generation to higher-value pedagogical activities, such as refining learning objectives and providing nuanced student support. This model demonstrates how AI can be integrated into educational workflows as a collaborative partner, enhancing both operational efficiency and the quality of pedagogical practice. Social implications The widespread adoption of synergistic AI systems like IntelliA could democratize access to high-quality, personalized learning support, potentially narrowing educational equity gaps. However, it also raises critical social questions about the future of the teaching profession and the risk of over-reliance on automation. The key social implication is the need for a public discourse on responsible AI integration in education, ensuring that technology serves to amplify educator agency and enhance human-centric learning, rather than deskilling teachers or creating uniform, passive learning pathways. Originality/value This study presents a significant leap forward by proposing a novel, synergistic system architecture for AI in education. Unlike prior systems that address isolated tasks, IntelliA offers a holistically integrated and self-improving ecosystem. By demonstrating how to embed pedagogical principles directly into an AI’s core logic, IntelliA provides a scalable and flexible blueprint for a new generation of educational tools. This approach moves beyond simple automation, fostering a new dynamic where AI functions not as a mere instrument but as a genuine partner in the pedagogical process.

Purpose This study aims to enhance cryptocurrency price and trend prediction by applying advanced machine learning (ML) techniques. Given the market’s high volatility and complexity, the research identifies effective models for different conditions, providing insights for investors and risk management. Design/methodology/approach This study proposes a six-stage framework for cryptocurrency price prediction, integrating advanced ML techniques. Data from ten cryptocurrencies are processed, extracting 37 key features, including return, the Fear and Greed Index and various technical indicators. The model employs Deep Q-Networks (DQN), Long Short-Term Memory (LSTM) and multiple regression methods such as linear regression, support vector regression, ridge, LASSO, decision tree, Random Forest, multi-layer perceptron, stochastic gradient descent, elastic net and Bayesian regression. Model performance is evaluated using trading strategies and metrics like accuracy, sensitivity, recall, MSE, MAE and F1-score. Findings The results indicate that complex models like DQN and LSTM excel in volatile markets due to their ability to capture intricate price patterns, whereas simpler models such as linear regression and ridge regression perform better in stable conditions. The multi-layered parallel design enhances computational efficiency, enabling independent asset evaluation. These findings highlight the potential of artificial intelligence in improving prediction accuracy and supporting informed investment decisions. Originality/value This research introduces a novel six-stage ML framework incorporating diverse predictive models and key features for cryptocurrency forecasting. The multi-layered parallel approach enhances computational efficiency, setting this study apart from existing research. The comparative analysis of models offers valuable guidance for investors, traders and financial analysts navigating volatile cryptocurrency markets.

PurposeThe livestock industry is undergoing a critical transition to intensive, large-scale farming. Intelligent monitoring technologies are essential for improving epidemic early warning systems, reducing breeding costs, and promoting sustainable production. This study aimed to develop a novel pig behavior recognition method using advanced computer vision technology to support intelligent livestock farming.Design/methodology/approachThe YOLOv5 model was utilized to achieve contactless and efficient monitoring of daily pig activities. The study enhanced the YOLOv5 model by improving its input mechanism, backbone network and by incorporating the shuffle attention module. These modifications significantly improved the ability of the model to capture and interpret the spatiotemporal features of pig behavior.FindingsThe experimental results demonstrate that compared with the original YOLOv5 model, the Precision, Recall, mAP@0.5 and mAp@0.5:0.95 of the proposed model has improved by 3.0%, 2.3%, 2.6% and 10.5%, respectively. These findings showcase the model’s effectiveness and potential for real-world applications in intelligent livestock farming, and highlight the feasibility of employing advanced computer vision models to enhance monitoring and management in animal farming environments.Originality/valueThis study presents a novel approach to pig behavior recognition by integrating cutting-edge computer vision techniques with YOLOv5 enhancements. This study contributes to the field by addressing the challenges of spatiotemporal feature extraction and demonstrating the practical application of these methods in intelligent livestock farming. Future research directions include generalization to other animal species, integration with other sensor data, teal-time monitoring and decision support and application in wildlife and laboratory animal research, thus further advancing the intelligent breeding industry.

PurposeAccurately identifying the risk of poverty-returning is a complex and critical challenge in current poverty alleviation efforts. However, there is currently no study on evaluation methods for the risk of poverty-returning. This study aims to establish a robust and systematic approach for an evaluation framework for the risk of poverty-returning.Design/methodology/approachBased on relevant assessment criteria, a maximum deviation method was established to identify the weights of the indicators. A complex evaluation methodology using prospect theory (PT), a q-rung orthopair fuzzy set (QrOFS) and evaluation relying on distance from average solution [EDAS] (QrOFS-PT-EDAS) was developed to evaluate the poverty-returning risks. Some policy recommendations to reduce the risk of poverty-returning have also been put forward.FindingsHis study identifies the risk factors of poverty relapse from nine aspects, including natural disasters, accidents and policy-driven poverty relapse. In addressing the evaluation challenge arising from uncertain decision-making, the QrOFS aligns more with people’s thinking habits and expression methods in complex environments. The proposed hybrid evaluation framework accurately measures the poverty-returning risk, which is beneficial for the formulation of policy recommendations.Originality/valueA scientific and comprehensive assessment system index for poverty-returning is constructed. A hybrid QrOFS-PT-EDAS framework is presented to make the evaluation results more scientific and objective. Several strategic recommendations for reducing the poverty-returning risk are presented. This study offers a novel framework for assessing poverty-returning issues that can be extended to many other areas.

PurposeGathering, analyzing and securing electronic data from various digital devices for use in legal or investigative procedures is the key process of computer forensics. Information retrieved from servers, hard drives, cellphones, tablets and other devices is all included in this. This article tackles the challenging problem of how to prioritize different kinds of computer forensics and figure out which kind is most useful in cases of cybercrime, fraud, theft of intellectual property, harassment and espionage.Design/methodology/approachTherefore, we first introduce enhanced versions of Hamacher power aggregation operators (AOs) within the framework of bipolar complex fuzzy (BCF) sets. These include BCF Hamacher power averaging (BCFHPA), BCF Hamacher power-weighted averaging (BCFHPWA), BCF Hamacher power-ordered weighted averaging (BCFHPOWA), BCF Hamacher power geometric (BCFHPG), BCF Hamacher power-weighted geometric (BCFHPWG) and BCF Hamacher power-ordered-weighted geometric (BCFHPOWG) operators. Employing the devised AOs, we devise a technique of decision-making (DM) for dealing with DM dilemmas with the BCF set (BCFS).FindingsWe prioritize different types of computer forensic by taking artificial data in a numerical example and getting the finest computer forensic. Further, by this example, we reveal the applicability of the proposed theory. This work provides a more elaborate and versatile procedure for classifying computer forensics with dual aspects of criteria and extra fuzzy information. It allows for better and less biased DM in the more intricate digital investigations, which may lead to better DM and time-saving in real-life forensic scenarios. To demonstrate the significance and impression of the devised operators and techniques of DM, they are compared with existing ones.Originality/valueThis research is the first to combine Hamacher and power AOs in BCFS for computer forensics DM. It presents new operators and a DM approach that is not encountered in the existing literature and is specifically designed to deal with the challenges and risks associated with the classification of computer forensics. The framework’s capacity to accommodate bipolar criteria and extra fuzzy information is a major development in the field of digital forensics and decision science.

Purpose The anomaly detection task for oil and gas pipelines based on acoustic signals faces issues such as background noise coverage, lack of effective features, and small sample sizes, resulting in low fault identification accuracy and slow efficiency. The purpose of this paper is to study an accurate and efficient method of pipeline anomaly detection. Design/methodology/approach First, to address the impact of background noise on the accuracy of anomaly signals, the adaptive multi-threshold center frequency variational mode decomposition method(AMTCF-VMD) method is used to eliminate strong noise in pipeline signals. Secondly, to address the strong data dependency and loss of local features in the Swin Transformer network, a Hybrid Pyramid ConvNet network with an Agent Attention mechanism is proposed. This compensates for the limitations of CNN’s receptive field and enhances the Swin Transformer’s global contextual feature representation capabilities. Thirdly, to address the sparsity and imbalance of anomaly samples, the SpecAugment and Scaper methods are integrated to enhance the model’s generalization ability. Findings In the pipeline anomaly audio and environmental datasets such as ESC-50, the AMTCF-VMD method shows more significant denoising effects compared to wavelet packet decomposition and EMD methods. Additionally, the model achieved 98.7% accuracy on the preprocessed anomaly audio dataset and 99.0% on the ESC-50 dataset. Originality/value This paper innovatively proposes and combines the AMTCF-VMD preprocessing method with the Agent-SwinPyramidNet model, addressing noise interference and low accuracy issues in pipeline anomaly detection, and providing strong support for oil and gas pipeline anomaly recognition tasks in high-noise environments.