Visual Knowledge Research Articles

Commonsense Knowledge in Foundation and Large Language Models

The development and continuous expansion of the transformer deep-learning architecture have produced enormous effects across various domains, including but not limited to natural language processing. The power of deep learning models has sparked a fresh interest in commonsense knowledge, which has been aided by transformer-based language models. Most of the recent research has concentrated on delving into the commonsense already built into these models' pre-trained parameters and finding ways to fill in any gaps in commonsense utilizing knowledge graphs and fine-tuning. In order to broaden a limited commonsense knowledge network that was originally generated solely from visual data, we are building on the demonstrated linguistic understanding of extremely large transformer-based language models. Compared to language models that are fine-tuned on a huge starting corpus, few-shotprompted pre-trained models are able to acquire the context of an initial knowledge graph with less bias. It has also been demonstrated that these models can contribute novel ideas to the visual knowledge networkIt is a new development in the field of commonsense knowledge generation that, as far as we can tell, can lead to a fivefold decrease in cost when compared to the current state of the art. Fuzzy language names assigned to the produced triples are another addition. Applying knowledge graphs as a framework, the procedure is comprehensive. It implies that the triples are expressed in natural language, analyzed, and then added to the commonsense knowledge network as triples again.

Research on Knowledge Graph Construction and Semantic Representation of Low Earth Orbit Satellite Spectrum Sensing Data

The growth of frequency-usage devices has made the electromagnetic spectrum posture complex, resulting in an urgent demand for frequency-usage posture cognition. However, the sensing of space-based platforms is limited by the transmission capacity of the satellite-to-ground link and the satellite processing capacity, which makes on-satellite data analysis and posture generation lack the efficient means. Facing the above issues, an idea of a knowledge graph construction and semantic representation for low Earth orbit (LEO) satellite spectrum sensing data is designed in this paper. In the designed construction process, technologies such as knowledge extraction, ontology construction, knowledge fusion and knowledge visualization are utilized to efficiently analyze on-satellite sensing data. Moreover, the constructed spectrum knowledge graph can be applied in the analysis and prediction of frequency-usage behavior and intelligent spectrum management, which exhibits the effectiveness of the spectrum knowledge graph. Finally, the further development of the spectrum knowledge graph is foreseen.

Global trends in research on MOG antibody-associated disease: bibliometrics and visualization analysis.

The purpose of this study was to investigate the current research status, focus areas, and developmental trends in the field of Myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD) through an analysis of scientific literature. The relevant research articles on MOGAD published from 1947 to 2022 were retrieved from the Web of Science database. The quantitative output of MOGAD related research articles, their distribution by country/region, data on collaborative publishing, influential authors, high-yield institutions, keywords, hotspots, and development trends were analyzed. Additionally, visual knowledge maps were generated using VOSviewer and Citespace. There has been a steady increase in the number of MOGAD related publications indicating that the subject has garnered increasing interest among researchers globally. The United States has been the leading contributor with 496 papers (19.25%), followed by China (244, 9.63%), Japan (183, 7.10%), the United Kingdom (154, 5.98%), and Germany (149, 5.78%). Among these countries, the United Kingdom boasts the highest citation frequency at the rate of 46.49 times per paper. Furthermore, active collaboration in MOGAD related research is observed primarily between the United States and countries such as Canada, Germany, Australia, Italy, the United Kingdom and Japan. Mayo Clinic ranks first in total articles published (109) and frequency of citations per article (77.79). Takahashi Toshiyuki from Tohoku University is the most prolific author, while Multiple Sclerosis and Related Disorders is the most widely read journal in this field. "Disease Phenotype", "Treatment", "Novel Coronavirus Infection and Vaccination", "Immunopathological Mechanisms", "Clinical characteristics of children" and "Prognosis" are the primary keywords clusters in this field. "Novel Coronavirus Infection and Vaccination" and "Immunopathological Mechanisms" are research hotspots and have great development potential. The past three decades have witnessed a significant expansion of research on MOGAD. The pathogenetic mechanism of MOGAD is poised to be the prominent research focus in this field in the foreseeable future.

Research hotpots and frontier trends of neuroprotective effects of magnesium from 1999 to 2023: A bibliometric analysis.

The neuroprotective effect of magnesium has been widely discussed, and its effectiveness has been confirmed by animal and clinical trials. However, there are still difficulties in clinical translation in diseases such as cerebral ischemia and subarachnoid hemorrhage. Therefore, it is necessary to analyze the literatures about neuroprotection of magnesium to reveal a more comprehensive knowledge framework, research hotspots and trends in the future. Original articles and reviews related to neuroprotective effects of magnesium from 1999 to 2022 were retrieved from the Web of Science Core Collection (WoSCC). The bibliometrics CiteSpace 6.2.R4 software was used to conduct co-occurrence/co-citation network analysis and plot knowledge visualization maps. A total of 762 articles from 216 institutions in 64 countries were included in this study. The United States had the largest number of publications, followed by China and Canada. The University of California, UDICE-French Research Universities, and the University of Adelaide were the top three institutions in publication volume. Crowther Caroline A was the most published and cited author. Among the top 10 cited articles, there were seven articles on neuroprotection in preterm infants and three on acute stroke. Keyword cluster analysis obtained 11 clusters, showing that current research hotspots focused on magnesium therapy in neurovascular diseases such as cerebral ischemia, spinal cord injury, subarachnoid hemorrhage, and emerging treatment strategies. The neuroprotective effects of magnesium in preterm infants have been extensively studied and adequately confirmed. The therapeutic effects of magnesium on cerebral ischemia and subarachnoid hemorrhage have been demonstrated in animal models. However, the results of clinical studies were not satisfactory, and exploring new therapeutic strategies may be the solution. Recently, the combination of magnesium and hypothermia had great potential in neuroprotective therapy and may become a development trend and hotspot in the future.

Design and Development of Intelligent Learning System for University Innovation and Entrepreneurship Based on Knowledge Visualisation

To build a comprehensive learning resource system covering learning process, learning users, learning resources and other aspects, an intelligent learning system for innovation and entrepreneurship in universities was constructed and designed combined with knowledge visualisation theory. The system includes student information record module, system recommendation calculation module, learning material module, and knowledge visualisation module. The research focuses on the system recommendation calculation module, including the User-based Collaborative Filtering Algorithm (UCF) and the Joint Recommendation (JR) algorithm of the application content recommendation algorithm. The best neighbour number of JR is 20, and the best transmission path length is 2. The recommendation effect is optimal under this setting. The convergence iterations of the training and test set of the joint recommendation algorithm are 80 times and 100 times, respectively. For the training set and test set, the accuracy, precision, sensitivity, recall, running time, error, and other performance evaluation indicators of the joint recommendation algorithm are better than the corresponding values of other recommendation algorithms. The system has a good design effect and experience effect in the application of the online learning platform.

Pelatihan Pembuatan Digital Brosur Promosi UMKM di Kabupaten Banyumas

AbstractThe community service is aimed to provide knowledge about brochure visualization. The method used to achieve these objectives involves activities such as presentations, training, and mentoring, which include knowledge building, modeling, joint construction, and independent construction activities. To measure the success of the activity, there are three types of evaluation: (1) the initial stage through a Pre-test. Questions are posed through a Google form to assess the participants' initial knowledge, (2) the implementation stage/process of the activity through observation, discussion, practice, and assessment of the final products, and (3) the final stage through a post-test, where answers are assessed to determine the knowledge improvement after participating in this PkM. With these evaluation stages, the knowledge and skills of the participants are assessed to understand their improved understanding of brochure visualization knowledge and skills. As a result, there is an increase in knowledge about brochure visualization. Through this activity, issues related to optimizing knowledge and skills in brochure visualization can be addressed.

Bridging knowledge gaps: a bibliometric analysis of non-invasive ventilation in palliative care studies

BackgroundDespite being a useful strategy for providing respiratory support to patients with advanced or terminal illnesses, non-invasive ventilation (NIV) requires in-depth investigation in several key aspects.ObjectivesThis bibliometric analysis seeks to comprehensively examine the existing research on the subject. Its goal is to uncover valuable insights that can inform the prediction trajectory of studies, guide the implementation of corrective measures, and contribute to the improvement of research networks.MethodsA comprehensive review of literature on NIV in the context of palliative care was conducted using the Web of Science core collection online database. The search utilized the key terms “non-invasive ventilation” and “palliative care” to identify the most relevant articles. All data were gathered on November 7, 2023. Relevant information from documents meeting the specified criteria was extracted, and Journal Citation Reports™ 2022 (Clarivate Analytics) served as the data source. The analysis employed literature analysis and knowledge visualization tools, specifically CiteScope (version 6.2.R4) and VOSviewer (version 1.6.20).ResultsA dataset with bibliometric findings from 192 items was analyzed. We found a consistent upward of the scientific output trend over time. Guidelines on amyotrophic lateral sclerosis management received the highest number of citations. Most documents were published in top-ranked journals. Less than one-third of the documents pertain to clinical studies, especially retrospective analyses (25%). Key topics such as “decision making”, and “communication” were less addressed.ConclusionsGiven the substantial clinical implications, further high-quality studies on this subject are recommended. Encouraging international collaborations is needed. Despite the growing volume of documents in the field, this bibliometric analysis indicates a decline in collaborative networks.

Visual Knowledge Domain of Artificial Intelligence in Computed Tomography: A Review Based on Bibliometric Analysis.

Artificial intelligence (AI)-assisted medical imaging technology is a new research area of great interest that has developed rapidly over the last decade. However, there has been no bibliometric analysis of published studies in this field. The present review focuses on AI-related studies on computed tomography imaging in the Web of Science database and uses CiteSpace and VOSviewer to generate a knowledge map and conduct the basic information analysis, co-word analysis, and co-citation analysis. A total of 7265 documents were included and the number of documents published had an overall upward trend. Scholars from the United States and China have made outstanding achievements, and there is a general lack of extensive cooperation in this field. In recent years, the research areas of great interest and difficulty have been the optimization and upgrading of algorithms, and the application of theoretical models to practical clinical applications. This review will help researchers understand the developments, research areas of great interest, and research frontiers in this field and provide reference and guidance for future studies.

Proposal of a Knowledge Capitalization Process to Construct Eco-Diars : A Knowledge-driven Platform Applied to Traditional Algerian Domestic Architecture

This article is part of doctoral research that proposes to capitalize on the environmental knowledge drawn from traditional Algerian domestic architecture, supported by a knowledge-based platform . This research aims to (1) build a capital of knowledge related to traditional environmental devices (EDs) allowing to suggest them as “references” to propose conceptual solutions during the upstream phases of the architectural design, (2) support the Algerian government's policy of preservation and digitization of architectural heritage, and (3) support the government's policy of reducing energy consumption in the building sector. From this perspective, this article proposes an interactive knowledge capitalization process involving Information and Communication Technologies (ICTs) for the modeling, exploitation, and visualization of ED-related knowledge. This article will present the proposed process for knowledge capitalization leading to the development of the knowledge-driven platform Eco-Diars intended to enable designers to efficiently perform their queries related to traditional environmental devices.

Knowledge mapping on the sun-induced chlorophyll fluorescence technology research: a scientometric and visualization analysis.

The sun-induced chlorophyll fluorescence (SIF) has received increasing attention over the past few years. This scientometric study analyzed the SIF research field, based on publications available in Web of Science Core Collection visualization, and cluster analysis enabled us to map the knowledge domain and intellectual landscape of this field and identify thematic trends, landmark articles, and emerging research themes. In this study, the software VOSviewer and CiteSpace were used to identify the intellectual base and research front using visualization and analysis. Then, we analyzed synthesized networks of co-authorship (author, institution, and country), co-citation (author, document, and journal), and co-occurring keywords. SIF has increased its publication output steadily since 2002. This study provided a visual knowledge map of the major research domains of SIF. Also, explanations and implications of the findings were explored, and the emerging trends were identified. This report provides a thorough overview of the current state and growth trends of SIF research, and it helps researchers and practitioners understand the major areas of attention in this area and provides insights into prospective areas for further study.

Knowledge Visualization in Outdoor Thermal Environment Studies: A Comprehensive Review and International Comparison

This review highlights the escalating challenges of outdoor thermal discomfort and urban heat islands due to urbanization and climate change, impacting sustainability and health. Analyzing global research trends, it reveals methodological differences between Chinese and Western scholars, with the former focusing more on quantitative analysis and the latter on diverse approaches including adaptation strategies. It underscores the importance of using emerging technologies like GIS, AI, and remote sensing for precise mapping, scenario modeling, and environmental monitoring. These tools offer innovative, data-driven solutions for urban planning, essential for addressing thermal impacts. The paper calls for enhanced Sino-Western cooperation and interdisciplinary research to develop context-specific, sustainable strategies for improving urban thermal environments.

Pelvic fractures healing in osteosynthesis by external device

Aim: To analyze radiological data of pelvic fractures healing after osteosynthesis by external device, to determine the possibility of correction during the union and terms of fixation. Background. Visualization of fracture healing and knowledge of X-ray manifestations in dynamics makes it possible to control and manage the reparative process of bone tissue during union of the pelvic bones. Material and methods. The clinical and radiological picture of pelvic fractures healing and the results of treatment in 207 polytrauma patients were studied. All observations are divided by the type of fracture according to M.Tile classification. Results: The analysis of long-term treatment outcomes of pelvic trauma was carried out in 154(74,39%) patients, in the period from 1,5 to 16 years after surgery. All patients showed complete healing of the pelvic fractures and long bones injuries and their functional recovery. Conclusions: Correction of the pelvic ring injuries with an external fixation device is possible in almost all cases. X-ray dynamics provides control of the union process of the pelvic bones and allows to manage it.

PointMCD: Boosting Deep Point Cloud Encoders Via Multi-View Cross-Modal Distillation for 3D Shape Recognition

As two fundamental representation modalities of 3D objects, 3D point clouds and multi-view 2D images record shape information from different domains of geometric structures and visual appearances. In the current deep learning era, remarkable progress in processing such two data modalities has been achieved through respectively customizing compatible 3D and 2D network architectures. However, unlike multi-view image-based 2D visual modeling paradigms, which have shown leading performance in several common 3D shape recognition benchmarks, point cloud-based 3D geometric modeling paradigms are still highly limited by insufficient learning capacity due to the difficulty of extracting discriminative features from irregular geometric signals. In this article, we explore the possibility of boosting deep 3D point cloud encoders by transferring visual knowledge extracted from deep 2D image encoders under a standard teacher-student distillation workflow. Generally, we propose PointMCD, a unified multi-view cross-modal distillation architecture, including a pretrained deep image encoder as the teacher and a deep point encoder as the student. To perform heterogeneous feature alignment between 2D visual and 3D geometric domains, we further investigate visibility-aware feature projection (VAFP), by which point-wise embeddings are reasonably aggregated into view-specific geometric descriptors. By pair-wisely aligning multi-view visual and geometric descriptors, we can obtain more powerful deep point encoders without exhausting and complicated network modification. Experiments on 3D shape classification, part segmentation, and unsupervised learning strongly validate the effectiveness of our method. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">The code and data will be publicly available at</i> <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/keeganhk/PointMCD</uri> .

Towards Visual-Prompt Temporal Answer Grounding in Instructional Video.

Temporal answer grounding in instructional video (TAGV) is a new task naturally derived from temporal sentence grounding in general video (TSGV). Given an untrimmed instructional video and a text question, this task aims at locating the frame span from the video that can semantically answer the question, i.e., visual answer. Existing methods tend to solve the TAGV problem with a visual span-based predictor, taking visual information to predict the start and end frames in the video. However, due to the weak correlations between the semantic features of the textual question and visual answer, current methods using the visual span-based predictor do not work well in the TAGV task. In this paper, we propose a visual-prompt text span localization (VPTSL) method, which introduces the timestamped subtitles for a text span-based predictor. Specifically, the visual prompt is a learnable feature embedding, which brings visual knowledge to the pre-trained language model. Meanwhile, the text span-based predictor learns joint semantic representations from the input text question, video subtitles, and visual prompt feature with the pre-trained language model. Thus, the TAGV is reformulated as the task of the visual-prompt subtitle span localization for the visual answer. Extensive experiments on five instructional video datasets, namely MedVidQA, TutorialVQA, VehicleVQA, CrossTalk and Coin, show that the proposed method outperforms several state-of-the-art (SOTA) methods by a large margin in terms of mIoU score, which demonstrates the effectiveness of the proposed visual prompt and text span-based predictor. Besides, all the experimental codes and datasets are open-sourced on the website https://github.com/wengsyx/VPTSL.

Masked Generative Light Field Prompting for Pixel-Level Structure Segmentations

Pixel-level structure segmentations have attracted considerable attention, playing a crucial role in autonomous driving within the metaverse and enhancing comprehension in light field-based machine vision. However, current light field modeling methods fail to integrate appearance and geometric structural information into a coherent semantic space, thereby limiting the capability of light field transmission for visual knowledge. In this paper, we propose a general light field modeling method for pixel-level structure segmentation, comprising a generative light field prompting encoder (LF-GPE) and a prompt-based masked light field pretraining (LF-PMP) network. Our LF-GPE, serving as a light field backbone, can extract both appearance and geometric structural cues simultaneously. It aligns these features into a unified visual space, facilitating semantic interaction. Meanwhile, our LF-PMP, during the pretraining phase, integrates a mixed light field and a multi-view light field reconstruction. It prioritizes considering the geometric structural properties of the light field, enabling the light field backbone to accumulate a wealth of prior knowledge. We evaluate our pretrained LF-GPE on two downstream tasks: light field salient object detection and semantic segmentation. Experimental results demonstrate that LF-GPE can effectively learn high-quality light field features and achieve highly competitive performance in pixel-level segmentation tasks.

Construction and Application of User Check-in Spatiotemporal Knowledge Graph Based on Neo4j

With the rapid acceleration of urbanization and the widespread adoption of social networks, A vast amount of spatiotemporal check-in data from social media users has been generated. Traditional relational databases have limitations in handling massive spatiotemporal data, whereas graph databases, with their efficiency in data querying and processing, offer a novel solution for the analysis of spatiotemporal data. The paper introduces a method for constructing a user check-in spatiotemporal knowledge graph based on the Neo4j graph database, utilizing a designed graph model to create a visual knowledge graph. By incorporating temporal and spatial characteristics into entities, we realize efficient management and complex query of spatiotemporal data of social media user check-ins. By adding sentiment attributes to relationships through sentiment analysis of check-in text, the application of check-in text is realized. Ultimately, employing Cypher to achieve knowledge querying, we analyze the hot spots of urban activities and categories.

ChartGPT: Leveraging LLMs to Generate Charts from Abstract Natural Language.

The use of natural language interfaces (NLIs) to create charts is becoming increasingly popular due to the intuitiveness of natural language interactions. One key challenge in this approach is to accurately capture user intents and transform them to proper chart specifications. This obstructs the wide use of NLI in chart generation, as users' natural language inputs are generally abstract (i.e., ambiguous or under-specified), without a clear specification of visual encodings. Recently, pre-trained large language models (LLMs) have exhibited superior performance in understanding and generating natural language, demonstrating great potential for downstream tasks. Inspired by this major trend, we propose ChartGPT, generating charts from abstract natural language inputs. However, LLMs are struggling to address complex logic problems. To enable the model to accurately specify the complex parameters and perform operations in chart generation, we decompose the generation process into a step-by-step reasoning pipeline, so that the model only needs to reason a single and specific sub-task during each run. Moreover, LLMs are pre-trained on general datasets, which might be biased for the task of chart generation. To provide adequate visualization knowledge, we create a dataset consisting of abstract utterances and charts and improve model performance through fine-tuning. We further design an interactive interface for ChartGPT that allows users to check and modify the intermediate outputs of each step. The effectiveness of the proposed system is evaluated through quantitative evaluations and a user study.

A Leap From Theory to Reality: Knowledge Visualization of Quantum Computing

A Leap From Theory to Reality: Knowledge Visualization of Quantum Computing

Retraction Notice: Vr Technology in English Teaching from the Perspective of Knowledge Visualization

Retraction Notice: Vr Technology in English Teaching from the Perspective of Knowledge Visualization

Regimes of visibility and algorithmic surveillance on digital platforms: operationalizing the thought through images in a collaborative perspective

In this article, I present and analyze the results obtained in the first interview for the ongoing thesis project, which investigates how visibility regimes interfere with the algorithmic surveillance regime of bodies in contemporary times. In an initiative stimulated by the exchanges carried out in the Thematic Group Digital and Popular Communication, promoted by the IX ALAIC Summer School, I propose to connect in a single interview the two methodological fronts addressed: the construction of a collaborative visual atlas in the theoretical wake of archeology of visual knowledge, inspired by Aby Warburg, and conducting semi-structured interviews with collective leaders who question aspects of the capitalist culture of surveillance. To this end, the interviewee was presented with a set of images of artistic works related to the themes of body visibility and algorithmic surveillance. The data obtained brought up three thematic nodes: 1) Historical struggles, 2) Countercolonial cosmovisions, and 3) Thinking of images in opposition, which will be analyzed in resonance with the reflections of the authors Antônio Bispo do Rosário, Tarcízio Silva, and Kate Crawford.