Semantic Network Research Articles

TeKo: Text-Rich Graph Neural Networks With External Knowledge.

Graph neural networks (GNNs) have gained great prevalence in tackling various analytical tasks on graph-structured data (i.e., networks). Typical GNNs and their variants adopt a message-passing principle that obtains network representations by the attribute propagates along network topology, which however ignores the rich textual semantics (e.g., local word-sequence) that exist in numerous real-world networks. Existing methods for text-rich networks integrate textual semantics by mainly using internal information such as topics or phrases/words, which often suffer from an inability to comprehensively mine the textual semantics, limiting the reciprocal guidance between network structure and textual semantics. To address these problems, we present a novel text-rich GNN with external knowledge (TeKo), in order to make full use of both structural and textual information within text-rich networks. Specifically, we first present a flexible heterogeneous semantic network that integrates high-quality entities as well as interactions among documents and entities. We then introduce two types of external knowledge, that is, structured triplets and unstructured entity descriptions, to gain a deeper insight into textual semantics. Furthermore, we devise a reciprocal convolutional mechanism for the constructed heterogeneous semantic network, enabling network structure and textual semantics to collaboratively enhance each other and learn high-level network representations. Extensive experiments illustrate that TeKo achieves state-of-the-art performance on a variety of text-rich networks as well as a large-scale e-commerce searching dataset.

Impact of Disciplinary Variations on Second Language Mental Lexicon

This study investigates how academic disciplines impact second language (L2) lexical competencies. Prior L2 research has often overlooked the broader effects of disciplinary backgrounds on lexical development. To address this gap, this study utilized lexical decision, memory, and semantic fluency tasks to examine lexicon recognition, memory, and storage processes among L2 Chinese learners from various academic fields. The study participants comprised 16 students from the Humanities and Social Sciences (HSS) disciplines and 11 from the Science and Engineering (S&E) disciplines, all having passed HSK level 4. The tasks were conducted using E-Prime 2.0. The findings revealed distinct cognitive strategies and thinking patterns among the participants from different disciplines . These strategies and patterns affected lexical task performance in terms of accuracy and response times. Although lexical memory exhibited no significant variation among the groups, discipline-specific tendencies were observed in the formation of semantic networks. Learners from S&E disciplines preferred mastering field-relevant Chinese terminology, whereas HSS learners tended to acquire words outside their specific disciplines. This observation highlights how learners’ disciplinary backgrounds influence their vocabulary acquisition and the organization of semantic networks in L2.

Dynamics of online debates: insights from textual network analysis

AbstractTextual data analysis is critical for monitoring changing themes over time. To overcome challenges posed by data richness, graph theory emerges as a tool for investigating word-topic associations. We present an approach to clustering co-occurrence word networks that prioritises network similarity quantification over time. Addressing theoretical and network geometrical constraints, a statistical framework for manifold data analysis facilitates the grouping of semantic networks, partitioning the observed time frame into periods, and identifying dominant topics in each period via tensor decomposition. The analysis of Brexit-related tweets demonstrates the efficacy of modern methods for identifying social media patterns on public discourse.

3D reconstruction method of welding area by fusion of coding raster and semantic segmentation network

3D reconstruction method of welding area by fusion of coding raster and semantic segmentation network

Analysis of Semantic Network between Teacher’s Teaching Concepts and Student’s Acquisition Concepts on Photosynthesis Domain in the Middle School

Analysis of Semantic Network between Teacher’s Teaching Concepts and Student’s Acquisition Concepts on Photosynthesis Domain in the Middle School

Semantics Network Analysis of Achievement Standards in Physics of 2022 Revised Curriculum

Semantics Network Analysis of Achievement Standards in Physics of 2022 Revised Curriculum

Beyond virtual boundaries: the intersection of the metaverse technologies, tourism, and lifelong learning in China’s digital discourse

In the evolving landscape of digital education, the metaverse emerges as a revolutionary platform, particularly within the realm of tourism, viewed through the lens of lifelong learning. This study delves into the integration of the metaverse in the tourism and education sectors in China, aiming to uncover the intricate dynamics of public perceptions and the educational potential of this digital innovation. This research utilizes semantic network and sentiment analysis of social media discussions on China’s Weibo platform to provide a detailed understanding of sentiments toward the metaverse’s impact on tourism education. Semantic Network Analysis explores the complex dialog surrounding the metaverse in the context of lifelong learning. The findings reveal a balanced mix of enthusiasm for the immersive educational opportunities afforded by the metaverse and concerns regarding data privacy, digital addiction, and ethical considerations. It is evident from this duality that a balanced approach is necessary when the metaverse is integrated into tourism and learning while being sensitive to technological opportunities and ethical challenges. The metaverse could be an immersive, interactive learning environment endorsed by modern educational theories, supporting active learning through experience. The metaverse will provide valuable assistance to tourism education, which will help enhance this field through the simulation of real tourism for students to learn, which will foster cultural understanding and meet the demands of a globalized tourism industry. It further recommends that policies and practices should be formulated to ensure equal access, respect for the privacy of every participant, and the ethical use of the technology. This study contributes to the analysis of the potentials and challenges of the metaverse in the tourism and learning industry. It highlights the metaverse as an ‘emerging tool for lifelong learning’ and offers insights into how educators, policymakers, and industry leaders, all aiming at the integration of digital innovations in education, should exploit the metaverse.

Typification of Projects for the Transition to Cloud Services

Introduction. Digital solutions make the operation of the company clearer, reduce staff costs, and provide data security. Various aspects of automation, digitalization and cloud technologies are described in the literature. The question is raised about the quality of the methodological basis for such transformations. Global and narrow technical approaches are presented. As a rule, materials are presented from the perspective of experts in the implementation of digital technologies. In this paper, for the first time, the author's scheme is proposed that can allow not only providers, but also their customers to navigate the upcoming transition to the cloud. The latter will receive systematic information on how to select a contractor and the most economically feasible option of cooperation.Materials and Methods. The information model was based on the semantic network as a system of nodes, their characteristics and connections. Management of the cloud migration project and the migration itself were visualized. The contraction practice between providers and their customers was summarized. The specifics of the tasks of such projects were taken into account. The part of the subject area related to the implementation of a cloud service is algorithmic — a step-by-step transition to the cloud, a generalized scheme of the process taking into account the hierarchy of elements are presented.Results. For the first time, a method of self-preparation of a company for the implementation of cloud solutions is proposed. The algorithm systematizes the cloud migration processes. The activities related to goal setting, audit, selection of cloud environment and services, calculation of the economic efficiency of the project, planning and implementation of migration, technical support and scaling of processes are described. The possibilities of determining the economic feasibility of measures for the transition to the cloud are shown. The costs of equipment, data storage and processing, software licenses, salaries, information security, etc., are taken into account. The amount received is compared to the providers' offers. For the final decision, the costs of infrastructure support are taken into account — by the customer or the outsourcer. The best option is selected. As a result, the customer gets the opportunity to work with better profitability and scale the project. Feedback is provided, and processes are adjusted, starting with IT reaudit.Discussion and Conclusion. The proposed solution will give the customer's management a system view of the execution sequence when migrating to the cloud, the issues and tasks to discuss with a potential outsourcer. Providers can use the algorithm to typify and unify projects, which can eventually simplify the coordination of the list of services and the migration procedure with customers. In this way, the parties can free up significant resources in terms of time, labor and other costs. In addition, customers and providers can partially use the described semantic network to develop not only the organizational, but also the technical aspect of the project.

Semantic Segmentation Network for Unstructured Rural Roads Based on Improved SPPM and Fused Multiscale Features

Semantic Segmentation Network for Unstructured Rural Roads Based on Improved SPPM and Fused Multiscale Features

3D Point Cloud Semantic Segmentation based on Multi-scale Dense Nested Networks

Aiming at the problem that the relationship between geometric features and semantic features is ignored in the point cloud data downsampling process, which leads to inaccurate segmentation of object boundaries and structural details, this paper proposes a 3D point cloud semantic segmentation network based on multi-scale dense nested type. Firstly, a dense nested network architecture is constructed by nesting multiple multi-scale feature fusion modules to fuse multi-scale features of different directions between encoder-decoder paths, so as to effectively propagate local ge-ometric context information and enhance the ability of cross-scale information interaction. Secondly, a local feature aggregation unit is constructed in the multi-scale feature fusion module, which strengthens the structural awareness within the local point set based on graph convolution and attention mechanism, and promotes the complementarity of local geometric features and abstract semantic information. Then, the cross-layer multi-loss supervision module is combined to further optimize the multi-scale feature propagation, which makes the network training more stable and improves the point cloud segmentation accuracy. Finally, this paper verified the proposed network on the S3DIS dataset. The experimental results show that the proposed network has the mean intersection over Union of 71.2% and the overall accuracy of 88.7%, which is 1.2 and 0.7 percentage points higher than RandLA-Net, respectively, which proves that the proposed network can effectively improve the accuracy of 3D point cloud semantic segmentation.

Samba: Semantic segmentation of remotely sensed images with state space model

High-resolution remotely sensed images pose challenges to traditional semantic segmentation networks, such as Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs). CNN-based methods struggle to handle high-resolution images due to their limited receptive field, while ViT-based methods, despite having a global receptive field, face challenges when processing long sequences. Inspired by the Mamba network, which is based on a state space model (SSM) to efficiently capture global semantic information, we propose a semantic segmentation framework for high-resolution remotely sensed imagery, named Samba. Samba utilizes an encoder-decoder architecture, with multiple Samba blocks serving as the encoder to efficiently extract multi-level semantic information, and UperNet functioning as the decoder. We evaluate Samba on the LoveDA, ISPRS Vaihingen, and ISPRS Potsdam datasets using the mIoU and mF1 metrics, and compare it with top-performing CNN-based and ViT-based methods. The results demonstrate that Samba achieves unparalleled performance on commonly used remotely sensed datasets for semantic segmentation. Samba is the first to demonstrate the effectiveness of SSM in segmenting remotely sensed imagery, setting a new performance benchmark for Mamba-based techniques in this domain of semantic segmentation. The source code and baseline implementations are available at https://github.com/zhuqinfeng1999/Samba.

Employee satisfaction and retention:social marketing and happiness

this research analyses the critical factors influencing employee satisfaction and retention in the context of the Great Resignation by exploring the key organizational elements that significantly impact employee contentment and loyalty. The justification for this study lies in the need fororganizations to understand and address the issues leading to high turnover rates and employee dissatisfaction, which have been exacerbated byrecent global workforce changes. The main objectives are to identify the key factors that contribute to employee satisfaction and to provide insights into how organizations can enhance their work environments, career growth opportunities, and management practices to retain talent. Themethodology involves a semantic network analysis of 138,764 employee reviews from 136 companies, collected via Glassdoor between 2021 and 2023. These reviews were analyzed using the Semantic Networks Tool to identify recurring themes related to work environment, career growth, workload, and management practices. The main results indicate that both current and former employees highly value a supportive and positive work environment and career growth opportunities. However, significant disadvantages include overwork, long hours, and poor managementpractices. The study highlights that the critical factors influencing employee satisfaction and retention include a positive work environment,opportunities for career growth, and effective management practices

Multilayer spatial-spectral segmentation network for detecting AFB1

Aflatoxin, a toxin produced by Aspergillus flavus, is commonly found in peanuts and poses a significant threat to human health. In this paper, we propose an adjustable Mul-U-Net model, capable of learning both spectral and spatial information in parallel, facilitating rapid detection of aflatoxins by adjusting the number of central spatial U units. First, AFB1 was semantically segmented by inputting hyperspectral image cubes into six semantic segmentation models: Mul-U-Net, U-Net, FCN, SegNet, CNN, and RefineNet. Experimental results demonstrate that Mul-U-Net outperformed the other models, achieving an overall accuracy of 95.64 % and an RMSE of 0.2079. We further examined the influence of different spatial U units on model performance and found that the optimal spatial U count matched the number of labels. In the aflatoxin dataset, the Mul-2 U-Net model achieved the highest accuracy of 96.12 %. Further validation on the Indian Pines and Pavia University datasets showed that Mul-16 U-Net and Mul-9 U-Net achieved the highest accuracies of 98.12 % and 98.01 %, respectively. This study presents a robust, high-performance, and tunable Mul-U-Net semantic segmentation network, offering a valuable new approach for the identification of aflatoxins in peanuts.

CiC-NET: a real-time semantic segmentation network for dam surface crack detection

CiC-NET: a real-time semantic segmentation network for dam surface crack detection

Vibration displacement measurement of bridge structural models using image super-resolution reconstruction and visual object detection network

Abstract Visual vibration measurement has emerged in the field of structural health monitoring in recent years, but it still has some shortcomings in terms of resolution, recognition rate and real-time performance. Considering the three aspects of recovering high-frequency image details, improving the compactness of the target bounding box, and reducing the computational time, we use the constructed image super-resolution reconstruction model and target detection model to measure the vibration displacement of the bridge structural model. First, we integrate the Transformer module into the Unet network with a simple structure. The Swin and Global Transformer Unet (SGTU) module constructed in this form can reduce the computational cost while reconstructing the large-resolution feature map target, and it can sharply edge information of the vibration target. We use the framework of the YOLOv5 algorithm as the backbone, and use the GhostBottleneck (GB) module to reduce the time for convolution operations to generate similar features. In addition, the proposed DWCBottleneck (DWCB) fusion module is also able to achieve high-level semantic fusion and network depth expansion with minimal computational cost. Finally, the center point offset of the bounding box predicted by the model can be used to obtain the displacement offset of the object in the image sequence. The position information of the target in the first frame image is used as the reference frame for calculating the offset, and the vibration displacement of the flexible structure in the image coordinate system is obtained by calculating the deviation of the displacement between the remaining frames and the first frame. We perform qualitative and quantitative comparisons in three aspects: video super-resolution reconstruction, visual detection robustness, and sensor vibration measurement displacement using a homemade vibration image dataset. The time-frequency domain displacement curves regressed by the visual vibration measurement algorithm are compared with the curves acquired after accelerometer acquisition, indicating the necessity of super-resolution reconstruction in visual vibration measurement.

Information technology and business strategic alignment in small and medium organizations: an exploratory study in Mexico

Purpose This qualitative study aims to fill gaps in a widely studied and relevant organizational feature: the alignment between information technologies and business strategies. Design/methodology/approach This research is a qualitative study. The authors used focus groups, content analysis and semantic networks as research approaches to identify the main factors that prevent or foster such alignment. Findings Results reveal a leading role of innovation, organizational culture, access to information and financial factors that could promote or inhibit alignment and competitiveness. Originality/value This research was conducted only in small and medium organizations in Mexico, which represents about 52% of the Mexican Gross Domestic Product (for Mexico as one of the leading trade partners of the USA).

Efficient and accurate semi-supervised semantic segmentation for industrial surface defects

Deep learning-based defect detection methods have gained widespread application in industrial quality inspection. However, limitations such as insufficient sample sizes, low data utilization, and issues with accuracy and speed persist. This paper proposes a semi-supervised semantic segmentation framework that addresses these challenges through perturbation invariance at both the image and feature space. The framework employs diverse perturbation cross-pseudo-supervision to reduce dependency on extensive labeled datasets. Our lightweight method incorporates edge pixel-level semantic information and shallow feature fusion to enhance real-time performance and improve the accuracy of defect edge detection and small target segmentation in industrial inspection. Experimental results demonstrate that the proposed method outperforms the current state-of-the-art (SOTA) semi-supervised semantic segmentation methods across various industrial scenarios. Specifically, our method achieves a mean Intersection over Union (mIoU) 3.11% higher than the SOTA method on our dataset and 4.39% higher on the public KolektorSDD dataset. Additionally, our semantic segmentation network matches the speed of the fastest network, U-net, while achieving a mIoU 2.99% higher than DeepLabv3Plus.

Decoupled semantic graph neural network for knowledge graph embedding

Knowledge graph embedding (KGE) learns an embedding space for a more accurate representation of entities and relations. Although the KGE model has proliferated, it often fails to fully capture the rich semantics in knowledge graphs. Some studies attempt to apply decoupling methods to decompose node representations, yet they neglect the semantic noise generated during the decoupled process. To address these issues, we introduce the decoupled semantic graph network for KGE (named DSGNet), which employs a novel approach that combines semantic decoupling with structural awareness. DSGNet begins by projecting the knowledge graph into distinct semantic spaces, ensuring minimal correlation between them to achieve effective semantic decoupling. To mitigate semantic noise, a top-k sampling technique is applied to the decoupled graphs. DSGNet then aggregates these different semantic graphs using a relation-aware aggregation module, followed by a multi-layer aggregation process for enhanced node representation. The extensive experiments demonstrate that DSGNet performs competitively on two popular datasets. We make our code publicly available at https://github.com/HubuKG/DSGNet.

Minimal invasive extracorporeal circulation: A bibliometric network analysis of the global scientific output.

Minimal Invasive Extracorporeal Circulation (MiECC) has recently emerged as a more 'physiologic' alternative to conventional extracorporeal circulation. However, its adoption is still limited due to lack of robust scientific evidence and ongoing debate about its potential benefits. This bibliometric analysis aims to analyze the scientific articles on MiECC and identify current research domains and existing gaps to be addressed in future studies. Pertinent articles were retrieved from the Web of Science (WOS) database. The search string included 'minimal invasive extracorporeal circulation' and its synonyms. The VOSviewer (version 1.6.17) software was used to conduct comprehensive analyses. Semantic and research networks, bibliographic coupling and journal analysis were performed. Of the 1777 articles identified in WOS, 292 were retrieved. The trend in publications increased from 1991 to date. Most articles focused on transfusion requirements, acute kidney injury, inflammatory markers and cytokines, inflammation and delirium, though the impact of intraoperative optimal fluid and hemodynamic management as far as the occurrence of postoperative complications were poorly addressed. The semantic network analysis found inter-connections between the terms "cardiopulmonary bypass", "inflammatory response", and "cardiac surgery". Perfusion contributed the highest number of published documents. The most extensive research partnerships were between Germany, Greece, Italy, and England. Notwithstanding the scientific community's growing interest in MiECC, crucial topics (i.e., the best anesthetic management and intraoperative need for inotropes, vasopressors and fluids) still require more comprehensive exploration. This investigation may prove to be a useful tool for clinicians, scientists, and students concerning global publication output and for the use of MiECC in cardiac surgery.

A new methodological quest to evaluate South Korean digital diplomacy in US government web domains

This study employs a novel methodological approach to evaluate the effectiveness of South Korea’s digital diplomacy within U.S. government web spaces. Given the United States’ experience with modern digital public diplomacy, its foreign policy has a significant impact on both Korea and Japan. We examine how key issues in the Korea-Japan territorial dispute are perceived and preserved on U.S. government web spaces, as well as their effects on the public. The study used web archives for data collection, as well as a composite analysis combining web impact studies and semantic network analysis. The analysis revealed that Japan’s digital public diplomacy is more prominent, primarily through online media operated by the U.S. Embassy in Japan. This suggests that Korea needs strategies similar to Japan’s, leveraging digital technology and media.