Knowledge graphs (KGs) are published across diverse repositories using heterogeneous metadata models that vary in their descriptive elements, level of detail, vocabularies, and support for semantic interoperability, in part due to the absence of a standardized specification. This lack of consistency in metadata practices hinders effective discovery and reuse of KGs. To address this gap, we present a KG metadata specification and validation framework designed to foster structured, FAIR-aligned KG descriptions. Developed through a community-driven process, the specification defines 33 metadata elements formalized using the Shapes Constraint Language (SHACL) to enable automated validation. We applied the framework to metadata from 1,573 datasets in the Linked Open Data Cloud by mapping its 17 metadata fields to those in our specification. This mapping demonstrated the specification’s compatibility with the largest Linked Data repository and validation process revealed common issues and areas for improvement in metadata quality. Furthermore, we demonstrated how metadata can be published in a way that enhances its visibility for end users and its discoverability by search engines, achieving full compliance with FAIR-Checker evaluations through the publication of specification-conformant metadata for the Food Health Claims KG. The framework provides a foundation for KG publishers, repository managers, and researchers, promoting uniform, high-quality metadata practices and advancing FAIR principles within the KG community.
Probabilities or confidence values produced by artificial intelligence (AI) and machine learning (ML) models often do not reflect their true accuracy, with some models being under or overconfident in their predictions. For example, if a model is 80% sure of an outcome, is it correct 80% of the time? Probability calibration metrics measure the discrepancy between confidence and accuracy, providing an independent assessment of model calibration performance that complements traditional accuracy metrics. Understanding calibration is important when the outputs of multiple systems are combined, to avoid overconfident subsystems dominating the output. Such awareness also underpins assurance in safety or business-critical contexts and builds user trust in models. This article provides a comprehensive review of probability calibration metrics for classifier models, organizing them according to multiple groupings to highlight their relationships. We identify 94 metrics, and group them into four main families: point-based, bin-based, kernel or curve-based, and cumulative. For each metric, we catalog properties of interest and provide equations in a unified notation, facilitating implementation and comparison by future researchers. Finally, we provide recommendations for which metrics should be used in different situations.
This research introduces an enhanced stratified sampling-density-based spatial clustering of applications with noise (SS-DBSCAN), a scalable and robust density-based clustering algorithm designed to tackle challenges in high-dimensional and complex data analysis. The algorithm integrates advanced parameter optimization techniques to improve clustering accuracy and interpretability. Key innovations include a fast grid search method for optimizing the search of optimal minimum points (MinPts) by keeping the ϵ parameter obtained constant. Notably, this study emphasizes the often-overlooked MinPts parameter, introducing a dynamic approach that initiates by calculating density metrics within a specified ϵ distance and adjusting the MinPts range based on the standard deviation of these metrics. This approach identifies optimal MinPts values based on the maximum allowed range. Comprehensive experiments on five real-world datasets demonstrate SS-DBSCAN’s superior performance compared to density-based spatial clustering of applications with noise (DBSCAN), hierarchical DBSCAN, and ordering points to identify the clustering structure (OPTICS), evidenced by higher silhouette and Davies–Bouldin index scores. The results highlight SS-DBSCAN’s ability to capture intrinsic clustering structures accurately, providing deeper insights across various research domains. SS-DBSCAN’s scalability and adaptability to diverse data densities make it a valuable tool for analyzing large, complex datasets.
This study employs bibliometric methods to conduct an empirical analysis of 53 vocational education projects approved under the National Education Science Planning (NESP) Program between 2020 and 2024 in higher vocational institutions (HVI) across China. The aim is to reveal the current state, developmental trajectory, key characteristics, and future directions of research in higher vocational education. The findings indicate that the number and categories of approved projects are significantly influenced by national policy orientations, exhibiting notable fluctuations and overall imbalances in distribution. The geographical distribution of project-leading institutions is uneven, and the distinctive characteristics of higher vocational colleges have yet to be fully demonstrated. While the research themes increasingly align with national vocational education policies, the construction of research teams remains weak. Based on the structural analysis of the approved projects, it is recommended that HVI focus on type-based positioning, coordinate research capacity with regional development needs, optimize the allocation of research resources, concentrate on key tasks such as quality improvement and excellence cultivation, and enhance the research motivation of young faculty members. These efforts are essential for advancing vocational education research toward a new stage of high-quality development.
With the advancement of scientific research and the rapid growth of the internet, academic users increasingly face challenges in obtaining accurate information about peer research. As a key component of big data analytics, user profiling has emerged as a critical focus in the scientific research community. While graph neural networks (GNNs) perform well in various graph learning tasks, their scalability to large graphs becomes problematic as the number of nodes increases due to computational complexity. To address this issue, this study proposes a novel academic user profiling model based on graph neural networks tailored to the unique characteristics of scientific research networks. The main contributions of this work are as follows: (1) We propose a simplified transformer architecture that reduces model complexity to a linear relationship with the number of nodes. (2) By integrating the simplified transformer with GNNs, neighborhood information is aggregated while maintaining global attention. The experimental results demonstrate that the proposed model delivers exceptional performance in terms of both accuracy and efficiency.
Few-shot image semantic segmentation aims to achieve pixel-level classification for novel classes using only a few labeled examples. The method first trains the segmentation model on base classes, and then adapts it to novel classes. Although existing methods have achieved remarkable performance in few-shot image semantic segmentation, they still face the following challenges. Traditional methods typically rely on mask average pooling to generate single-category prototype vectors and perform feature matching via metric learning, but they exhibit significant limitations in modeling inter-category relationships and addressing complex background interference. Inspired by the analogy-based transfer mechanisms in cognitive psychology, we propose a Generalized Prototype Network (GPNet) to enhance the model's generalization ability for unseen categories and improve robustness in feature matching. GPNet consists of two key modules. The first is a generalized prototype enhancement module, which explores potential inter-category relationships to construct more discriminative category prototype representations. The second is a multi-scale feature alignment module, which dynamically aligns support and query features across multiple scales using an attention mechanism, thus mitigating background interference in complex scenarios. Experimental results demonstrate that the proposed method significantly outperforms existing state-of-the-art approaches on several few-shot semantic segmentation benchmarks, validating its effectiveness and generalization capabilities.
Modern compression and acceleration methods for exploring efficient deep neural networks render real-world applications more feasible. Existing approaches uniformly apply the same procedure to every input image, overlooking instancewise complexity variations. Moreover, owing to pruning or decomposition techniques, the upper bound of network representation capabilities might be permanently diminished. In this work, an input-dependent multiscale dynamic inference method (MSDI) is developed to strike a better balance between model performance and inference acceleration. Specifically, we modify the main body of a convolutional network to obtain a series of parameter-sharing subnetworks with varying levels of complexity. A side branch structure is then introduced to assign an input instance to a suitable subnetwork as its inference route, and we expect to accelerate the inference by assigning the easy input to the subnetwork with low capacity. We further propose multiscale distillation training to optimize the training of the modified subnetworks. Additionally, we compare the entropy-based and learning-based grading approaches, aiming to obtain a more suitable route assignment method. Experiments show that MSDI can accelerate most existing convolutional models, achieving up to 74.7
With the increasing severity of global water pollution, underwater debris has become a serious threat to marine ecosystems. Autonomous underwater robots equipped with efficient debris detection systems can significantly increase underwater cleanup efforts. This paper proposes a high-precision and lightweight underwater debris detection method. To address the challenges posed by the diverse shapes and varying scales of underwater debris, we introduce an improved detection approach. Using YOLOv8 as the backbone, we replace conventional convolution with Alterable Kernel Convolution (AKConv), which dynamically adjusts the convolutional kernel shape on the basis of input features, enabling more accurate capture of object details. In the neck network, an improved high-level screening-feature fusion pyramid network (HSFPN) replaces the original YOLOv8 neck, incorporating coordinate attention instead of channel attention to enhance multiscale feature fusion. The experimental results demonstrate that the proposed AH-YOLO method achieves significant improvements in mAP50 and mAP50:95 on the TrashCan dataset compared with the original YOLOv8.
Traditional supply chain finance has problems in four aspects: business flow, logistics, capital flow, and information flow. Blockchain technologies such as smart contracts and consensus mechanisms can help break through the above problems. They can improve the implementation effect of supply chain finance, and reduce risks. Based on blockchain technology, this paper constructs a multi-dimensional dynamic supply chain financial risk control system. The system includes multi-dimensional dynamic supply chain financial risk control technology, a credit evaluation system, and a capital monitoring and early warning system. By constructing a multi-dimensional dynamic supply chain financial risk control system, effective risk complete monitoring and penetrating supervision can be achieved. At the same time, this paper introduces technologies such as machine learning and deep learning to build an effective risk control system, which is of great significance for reducing supply chain financial risks.
