Search

Background: Timely and accurate prediction of disease progress is crucial for facilitating early intervention and treatment for various chronic diseases. However, due to the complicated and longitudinal nature of disease progression, the capacity and completeness of clinical data required for training deep learning models remains a significant challenge. This study aims to explore a new method that reduces data dependency and achieves predictive performance comparable to existing research. Methods: This study proposed DAPNet, a deep learning-based disease progression prediction model that solely utilizes the comorbidity duration (without relying on multi-modal data or comprehensive medical records) and disease associations from biomedical knowledge graphs to deliver high-performance prediction. DAPNet is the first to apply multi-view graph contrastive learning to disease progression prediction tasks. Compared with other studies on comorbidities, DAPNet innovatively integrates molecular-level disease association information, combines disease co-occurrence and ICD10, and fully explores the associations between diseases; Results: This study validated DAPNet using a de-identified clinical dataset derived from medical claims, which includes 2,714 patients and 10,856 visits. Meanwhile, a kidney dataset (606 patients) based on MIMIC-IV has also been constructed to fully validate its performance. The results showed that DAPNet achieved state-of-the-art performance on the severe pneumonia dataset (F1=0.84, with an improvement of 8.7%), and outperformed the six baseline models on the kidney disease dataset (F1=0.80, with an improvement of 21.3%). Through case analysis, we elucidated the clinical and molecular associations identified by the DAPNet model, which facilitated a better understanding and explanation of potential disease association, thereby providing interpretability for the model. Conclusions: The proposed DAPNet, for the first time, utilizes comorbidity duration and disease associations network, enabling more accurate disease progression prediction based on a multi-view graph contrastive learning, which provides valuable insights for early diagnosis and treatment of patients. Based on disease association networks, our research has enhanced the interpretability of disease progression predictions. [ABSTRACT FROM AUTHOR]

Background and aims Design and setting Participants and Measurements Findings Conclusions Amphetamine‐type stimulants are the second‐most used illicit drugs globally, yet there are no US Food and Drug Administration (FDA)‐approved treatments for amphetamine‐type stimulant use disorders (ATSUD). The aim of this study was to utilize a drug discovery framework that integrates artificial intelligence (AI)‐based drug prediction, clinical corroboration and mechanism of action analysis to identify FDA‐approved drugs that can be repurposed for treating ATSUD.An AI‐based knowledge graph model was first utilized to prioritize FDA‐approved drugs in their potential efficacy for treating ATSUD. Among the top 10 ranked candidate drugs, ketamine represented a novel candidate with few studies examining its effects on ATSUD. We therefore conducted a retrospective cohort study to assess the association between ketamine and ATSUD remission using US electronic health record (EHR) data. Finally, we analyzed the potential mechanisms of action of ketamine in the context of ATSUD.ATSUD patients who received anesthesia (n = 3663) or were diagnosed with depression (n = 4328) between January 2019 and June 2022. The outcome measure was the diagnosis of ATSUD remission within one year of the drug prescription.Ketamine for anesthesia in ATSUD patients was associated with greater ATSUD remission compared with other anesthetics: hazard ratio (HR) = 1.58, 95% confidence interval (CI) = 1.15–2.17. Similar results were found for ATSUD patients with depression when comparing ketamine with antidepressants and bupropion/mirtazapine with HRs of 1.51 (95% CI = 1.14–2.01) and 1.68 (95% CI = 1.18–2.38), respectively. Functional analyses demonstrated that ketamine targets several ATSUD‐associated pathways including neuroactive ligand‐receptor interaction and amphetamine addiction.There appears to be an association between clinician‐prescribed ketamine and higher remission rates in patients with amphetamine‐type stimulant use disorders. [ABSTRACT FROM AUTHOR]

ABSTRACT Current geographic knowledge graph research mainly focuses on static knowledge coupling through analysis of web text and maps, but its knowledge expression model fails to incorporate knowledge mechanisms such as process evolution and interactions. The geo‐scene is a synthesis of various geographic elements which prescribes spatial, temporal, semantic, attribute, and interrelationships. It has demonstrated its utility in facilitating geocomputation, spatiotemporal simulation and reasoning related to geo‐process. Therefore, developing a geo‐scene‐based knowledge graph is significant to improve the ability of spatiotemporal knowledge representation and service. In this study, we first analyze the basic characteristics of geo‐scenes and try to emphasize the evolution of geo‐scenes through redefining the ontologies ‘event’, ‘process’ and ‘state’. To describe the dynamic nested structure of geo‐scenes, we propose a geo‐scene knowledge representation framework, and illustrate the representation of a geo‐scene and its inner relationships. Moreover, to keep the integrity, interaction and dynamic of geo‐scenes, we develop a event‐process centered geo‐scene knowledge organization method. Furthermore, we have tested the utility of our proposed representation by the Beijing 731 heavy rainstorm. The case study represents heavy rainstorm to demonstrate the complicated structure of the disaster scene, and the support for precipitation process simulation, early warning information organization, and disaster situation information query. The proposed representation provides an idea for the construction of geo‐dynamic knowledge graphs, encodes both static and dynamic geographic knowledge, makes explicit the evolution and interaction knowledge in geo‐scenes, and expands the application service of geo‐scene knowledge graph in the era of GeoAI. [ABSTRACT FROM AUTHOR]

Background: Target identification and hit identification can be transformed through the application of biomedical knowledge analysis, AI-driven virtual screening and robotic cloud lab systems. However there are few prospective studies that evaluate the efficacy of such integrated approaches. Results: We synergistically integrate our in-house-developed target evaluation (SpectraView) and deep-learning-driven virtual screening (HydraScreen) tools with an automated robotic cloud lab designed explicitly for ultra-high-throughput screening, enabling us to validate these platforms experimentally. By employing our target evaluation tool to select IRAK1 as the focal point of our investigation, we prospectively validate our structure-based deep learning model. We can identify 23.8% of all IRAK1 hits within the top 1% of ranked compounds. The model outperforms traditional virtual screening techniques and offers advanced features such as ligand pose confidence scoring. Simultaneously, we identify three potent (nanomolar) scaffolds from our compound library, 2 of which represent novel candidates for IRAK1 and hold promise for future development. Conclusion: This study provides compelling evidence for SpectraView and HydraScreen to provide a significant acceleration in the processes of target identification and hit discovery. By leveraging Ro5's HydraScreen and Strateos' automated labs in hit identification for IRAK1, we show how AI-driven virtual screening with HydraScreen could offer high hit discovery rates and reduce experimental costs. Scientific contribution: We present an innovative platform that leverages Knowledge graph-based biomedical data analytics and AI-driven virtual screening integrated with robotic cloud labs. Through an unbiased, prospective evaluation we show the reliability and robustness of HydraScreen in virtual and high-throughput screening for hit identification in IRAK1. Our platforms and innovative tools can expedite the early stages of drug discovery. [ABSTRACT FROM AUTHOR]

Multimodal sentiment analysis is an important and active research field. Most methods construct fusion modules based on unimodal representations generated by pretrained models, which lack the deep interaction of multimodal information, especially the rich semantic-emotional information embedded in text. In addition, previous studies have focused on capturing modal coherence information and ignored differentiated information. We propose a text-enhanced multi-interactive attention and multitask learning network (TEMM). First, syntactic dependency graphs and sentiment graphs of the text are constructed, and additional graph embedding representations of the text are obtained using graph convolutional networks and graph attention networks. Then, self-attention and cross-modal attention are applied to explore intramodal and intermodal dynamic interactions, using text as the main cue. Finally, a multitask learning framework is constructed to exert control over the information flow by monitoring the mutual information between the unimodal and multimodal representations and exploiting the classification properties of the unimodal modality to achieve a more comprehensive focus on modal information. The experimental results on the CMU-MOSI, CMU-MOSEI, and CH-SIMS datasets show that the proposed model outperforms state-of-the-art models. [ABSTRACT FROM AUTHOR]

Decoding gene regulatory networks is essential for understanding the mechanisms underlying many complex diseases. GENET is developed, an automated system designed to extract and visualize extensive molecular relationships from published biomedical literature. Using natural language processing, entities and relations are identified from a randomly selected set of 1788 scientific articles, and visualized in a filterable knowledge graph. The performance of GENET is evaluated and compared with existing methods. The named entity recognition model has achieved an overall precision of 94.23% (4835/5131; 93.56–94.84%), recall of 97.72% (4835/4948; 97.27–98.10%), and an F1 score of 95.94%. The relation extraction model has demonstrated an overall precision of 91.63% (2593/2830; 90.55–92.59%), recall of 89.17% (2593/2908; 87.99–90.25%), and an F1 score of 90.38%. GENET significantly outperforms existing methods in extracting molecular relationships (P < 0.001). Additionally, GENET has successfully predicted WNT family member 4 regulates insulin‐like growth factor 2 via signal transducer and activator of transcription 3 in colon cancer. With RNA sequencing data and multiple immunofluorescence, the authenticity of this prediction is validated, supporting the promising feasibility of GENET. [ABSTRACT FROM AUTHOR]

The traditional model training approach based on negative sampling randomly samples a portion of negative samples for training, which can easily overlook important negative samples and adversely affect the training of knowledge graph embedding models. Some researchers have explored non-sampling model training frameworks that use all unobserved triples as negative samples to improve model training performance. However, both training methods inevitably introduce false negative samples and easy-to-separate negative samples that are far from the model's decision boundary, and they do not consider the adverse effects of long-tail entities and relations during training, thus limiting the improvement of model training performance. To address this issue, we propose a universal knowledge graph embedding framework based on high-quality negative sampling and weighting, called HNSW-KGE. First, we conduct pre-training based on the NS-KGE non-sampling training framework to quickly obtain an initial set of relatively high-quality embedding vector representations for all entities and relations. Second, we design a candidate negative sample set construction strategy that samples a certain number of negative samples that are neither false negatives nor easy-to-separate negatives for all positive triples, based on the embedding vectors obtained from pre-training. This ensures the provision of high-quality negative samples for model training. Finally, we apply weighting to the loss function based on the frequency of the entities and relations appearing in the triples to mitigate the adverse effects of long-tail entities and relations on model training. Experiments conducted on benchmark datasets FB15K237 and WN18RR using various knowledge graph embedding models demonstrate that our proposed framework HNSW-KGE, based on high-quality negative sampling and weighting, achieves better training performance and exhibits versatility, making it applicable to various types of knowledge embedding models. [ABSTRACT FROM AUTHOR]

The proliferation of the Internet and mobile devices has made it increasingly easy to propagate rumors on social media. Widespread rumors can incite public panic and have detrimental effects on individuals. In recent years, researchers have found that both the spatial structure of rumor diffusion and the temporal features of propagation can be effective in identifying rumors. However, existing methods tend to focus on either spatial structure or temporal information in isolation, and few models can effectively capture both types of information. Additionally, most existing methods treat continuously changing temporal information as static snapshots, neglecting the precise timing of propagation. Moreover, as users repost and comment, background knowledge associated with the posts also evolves dynamically, which is often ignored. To address these limitations, we propose CGNKP (Continuous-time Dynamic Graph Networks integrated with Knowledge Propagation), a model that jointly captures the spatial structure and continuous-time features of post propagation to fully understand the dynamics of background knowledge. Specifically, we introduce a novel method for encoding continuous-time dynamic graphs, modeling the propagation process through two dynamic graphs: a temporal propagation graph (for posts diffusion) and a temporal knowledge graph (for knowledge diffusion). Extensive experiments on real-world datasets demonstrate that CGNKP significantly outperforms multiple strong baselines, achieving accuracies of 0.861 on the Twitter15 dataset and 0.903 on the Twitter16 dataset. [ABSTRACT FROM AUTHOR]

Research suggests that individuals who experience prolonged exposure to stress may be at higher risk for developing psychological stress disorders. Currently, psychological stress is primarily evaluated by professional physicians using rating scales, which may be prone to subjective biases and limitations of the scales. Therefore, it is imperative to explore more objective, accurate, and efficient biomarkers for evaluating the level of psychological stress in an individual. In this study, we utilized 4D data-independent acquisition (4D-DIA) proteomics for quantitative protein analysis, and then employed support vector machine (SVM) combined with SHAP interpretation algorithm to identify potential biomarkers for psychological stress levels. Biomarkers validation was subsequently achieved through machine learning classification and a substantial amount of a priori knowledge derived from the knowledge graph. We performed cross-validation of the biomarkers using two batches of data, and the results showed that the combination of Glyceraldehyde-3-phosphate dehydrogenase and Fibronectin yielded an average area under the curve (AUC) of 92%, an average accuracy of 86%, an average F1 score of 79%, and an average sensitivity of 83%. Therefore, this combination may represent a potential approach for detecting stress levels to prevent psychological stress disorders. [ABSTRACT FROM AUTHOR]

Knowledge graph link prediction is crucial for constructing triples in knowledge graphs, which aim to infer whether there is a relation between the entities. Recently, graph neural networks and contrastive learning have demonstrated superior performance compared with traditional translation-based models; they successfully extracted common features through explicit linking between entities. However, the implicit associations between entities without a linking relationship are ignored, which impedes the model from capturing distant but semantically rich entities. In addition, directly applying contrastive learning based on random node dropout to link prediction tasks, or limiting it to triplet-level, leads to constrained model performance. To address these challenges, we design an implicit feature extraction module that utilizes the clustering characteristics of latent vector space to find entities with potential associations and enrich entity representations by mining similar semantic features from the conceptual level. Meanwhile, the subgraph mechanism is introduced to preserve the structural information of explicitly connected entities. Implicit semantic features and explicit structural features serve as complementary information to provide high-quality self-supervised signals. Experiments are conducted on three benchmark knowledge graph datasets. The results validate that our model outperforms the state-of-the-art baselines in link prediction tasks. [ABSTRACT FROM AUTHOR]

To address the issue of efficiently reusing the massive amount of unstructured knowledge generated during the handling of track circuit equipment faults and to automate the construction of knowledge graphs in the railway maintenance domain, it is crucial to leverage knowledge extraction techniques to efficiently extract relational triplets from fault maintenance text data. Given the current lag in joint extraction technology within the railway domain and the inefficiency in resource utilization, this paper proposes a joint extraction model for track circuit entities and relations, integrating Global Pointer and tensor learning. Taking into account the associative characteristics of semantic relations, the nesting of domain-specific terms in the railway sector, and semantic diversity, this research views the relation extraction task as a tensor learning process and the entity recognition task as a span-based Global Pointer search process. First, a multi-layer dilate gated convolutional neural network with residual connections is used to extract key features and fuse the weighted information from the 12 different semantic layers of the RoBERTa-wwm-ext model, fully exploiting the performance of each encoding layer. Next, the Tucker decomposition method is utilized to capture the semantic correlations between relations, and an Efficient Global Pointer is employed to globally predict the start and end positions of subject and object entities, incorporating relative position information through rotary position embedding (RoPE). Finally, comparative experiments with existing mainstream joint extraction models were conducted, and the proposed model's excellent performance was validated on the English public datasets NYT and WebNLG, the Chinese public dataset DuIE, and a private track circuit dataset. The F1 scores on the NYT, WebNLG, and DuIE public datasets reached 92.1%, 92.7%, and 78.2%, respectively. [ABSTRACT FROM AUTHOR]

Knowledge Graph Embedding aims to encode both entities and relations into a continuous low-dimensional vector space, which is crucial for knowledge-driven application scenarios. As abstract entities in knowledge graphs, concepts inherently possess unique hierarchical structures and encompass rich semantic information. Although existing methods for jointly embedding concepts and instances achieve promising performance, they still face two issues: (1) They fail to explicitly reconstruct the hierarchical tree structure of concepts in the embedding space; (2) They ignore disjoint concept pairs and overlapping concept pairs derived from concepts. In this paper, we propose a novel concept representation approach, called Hyper Spherical Cone Concept Embedding (HCCE), to explicitly model the hierarchical tree structure of concepts in the embedding space. Specifically, HCCE represents each concept as a hyperspherical cone and each instance as a vector, maintaining the anisotropy of concept embeddings. We propose two variant methods to explore the impact of embedding concepts and instances in the same or different spaces. Moreover, we design score functions for disjoint concept pairs and overlapping concept pairs, using relative position relations to incorporate them seamlessly into our geometric models. Experimental results on three benchmark datasets show that HCCE outperforms most existing state-of-the-art methods on concept-related triples and achieves competitive results on instance-related triples. The visualization of embedding results intuitively shows the hierarchical tree structure of concepts in the embedding space. [ABSTRACT FROM AUTHOR]

The normal operation of automated equipment is essential for power grid regulation, making the accurate identification and diagnosis of defects in this equipment highly significant. Constructing a knowledge graph for automated equipment defects offers an effective solution to challenges such as delayed reporting, low efficiency, and data omissions in manually recorded defects. To address this, we developed a framework for constructing an automated equipment defect knowledge graph by designing appropriate patterns and data layers. For knowledge extraction, we introduced two models: RoBERTa-BiLSTM for named entity recognition (NER) and ALBERT-BiGRU for relation extraction (RE), both of which demonstrated improved performance in their respective tasks. Additionally, we applied the KBGAT model for knowledge graph completion. Finally, Neo4j was used for storing, visualizing, and analyzing the knowledge graph, highlighting its significance in the operation of power grids and the advancement of digital power systems. [ABSTRACT FROM AUTHOR]

Geospatial knowledge graphs provide critical technology for integrating geographic information and semantic knowledge, which are very useful for geographic data analysis. As the scale of geospatial knowledge graphs continues to grow, the distributed management of geospatial knowledge graphs is becoming an inevitable requirement. Geospatial knowledge graph partitioning is the core technology for the distributed management of geospatial knowledge graphs. To support geographic data analysis, spatial relationships between entities should be considered in the application of geospatial knowledge graphs. However, existing knowledge graph partitioning methods overlook the spatial relationships between entities, resulting in the low efficiency of spatial queries. To address this issue, this study proposes a geospatial knowledge graph partitioning method based on dual clustering which performs two different clustering methods step by step. First, the density peak clustering method (DPC) is used to cluster geographic nodes. The nodes within each cluster are merged into a super-node. Then, we use an efficient graph clustering method (i.e., Leiden) to identify the community structure of the graph. Nodes belonging to the same community are further merged to reduce the size of the graph. Finally, partitioning operations are performed on the compressed graph based on the idea of the Linear-Weighted Deterministic Greedy Policy (LDG). We construct a geospatial knowledge graph based on YAGO3 to evaluate the performance of the proposed graph partitioning method. The experimental results show that the proposed method outperforms ten comparison methods in terms of graph partitioning quality and spatial query efficiency. [ABSTRACT FROM AUTHOR]

Additive Manufacturing (AM) technology offers remarkable flexibility in fabricating products with intricate geometries, presenting unprecedented advantages in material efficiency and speed. The process planning of AM plays a pivotal role in ensuring overall quality and time-efficiency of printed products. This drives engineers and researchers to explore various approaches to achieve optimal AM process solutions. However, numerous challenges persist, particularly in logical relationship reasoning and information representation for complex manufacturing tasks and design requirements. In this study, a novel AM process reasoning method based on relation mining is proposed, leveraging knowledge graph representation and graph neural networks (GNN). An AM knowledge graph is constructed comprising essential process information, followed by implementing RED-GNN to accomplish graph reasoning tasks for parameter recommendation. We then focus on the process planning scenario of lattice structures, a common geometry used for designing products with weight-relief requirements and high sensitivity to process parameters. A series of lattice structure parts are designed and tested using our proposed method, demonstrating strong performance and unveiling new potentials and opportunities in advancing knowledge-based engineering and intelligent manufacturing. [ABSTRACT FROM AUTHOR]

The objective of knowledge embedding (KE) is to represent entities and relations in a knowledge graph (KG) in a continuous low-dimensional vector space, thereby facilitating the integration of the KG into various downstream applications. Existing KE methods can be classified into two categories: closed-world knowledge embedding methods (CWKEs) and open-world knowledge embedding methods (OWKEs). CWKEs necessitate the re-training of the entire model upon the addition of new entities to the KG. In contrast, OWKEs offer greater practical application value as they do not require re-training due to their capacity to generate vector representations for new entities. The current major OWKEs employ the description text of new entities to generate vector representations for new entities. However, these methods are highly dependent on the richness of entity relationships and attribute information present in the text, and they are ineffective when there are only a few textual resources, such as entity names. In reality, most KGs have only a limited number of textual resources. For this reason, we propose the first open-world Knowledge Embedding method (LTKE) in a low textual resource environment. LTKE is designed to efficiently generate vector representations for new entities when KGs have only a few textual resources. By efficiently serializing the local structure of entities, LTKE can obtain information-rich text when an entity has only the entity name. Additionally, we develop a new scoring function to enhance model efficiency. Extensive experiments on real-world KG datasets (Freebase, WordNet) have demonstrated the effectiveness of the model. [ABSTRACT FROM AUTHOR]

Entity alignment (EA) aims to discover the equivalent entities in different knowledge graphs (KGs), which plays an important role in knowledge engineering. Recently, EA with dangling entities has been proposed as a more realistic setting, which assumes that not all entities have corresponding equivalent entities. In this paper, we focus on this setting. Some work has explored this problem by leveraging translation API, pre-trained word embeddings, and other off-the-shelf tools. However, these approaches over-rely on the side information (e.g., entity names) and fail to work when the side information is absent. On the contrary, they still insufficiently exploit the most fundamental graph structure information in KG. To improve the exploitation of the structural information, we propose a novel entity alignment framework called Structure-aware Wasserstein Graph Contrastive Learning (SWGCL), which is refined on three dimensions: (i) Model. We propose a novel Gated Graph Attention Network to capture local and global graph structure attention. (ii) Training. Two learning objectives: contrastive learning and optimal transport learning, are designed to obtain distinguishable entity representations. (iii) Inference. In the inference phase, a PageRank-based method HOSS (Higher-Order Structural Similarity) is proposed to calculate higher-order graph structural similarity. Extensive experiments on two dangling benchmarks demonstrate that our SWGCL outperforms the current state-of-the-art methods with pure structural information in both traditional (relaxed) and dangling (consolidated) settings. [ABSTRACT FROM AUTHOR]

One-shot knowledge graph completion (KGC) aims to infer unseen facts when only one support entity pair is available for a particular relationship. Prior studies learn reference representations from one support pair for matching query pairs. This strategy can be challenging, particularly when dealing with multiple relationships between identical support pairs, resulting in indistinguishable reference representations. To this end, we propose a disentangled representation learning framework for one-shot KGC. Specifically, to learn sufficient representations, we construct an entity encoder with a fine-grained attention mechanism to explicitly model the input and output neighbors. We adopt an orthogonal regularizer to promote the independence of learned factors in entity representation, enabling the matching processor with max pooling to adaptively identify the semantic roles associated with a particular relation. Subsequently, the one-shot KGC is accomplished by seamlessly integrating the aforementioned modules in an end-to-end learning manner. Extensive experiments on real-world datasets demonstrate the outperformance of the proposed framework. [ABSTRACT FROM AUTHOR]

The rapid advancement of artificial intelligence has significantly expanded the role of service robots in everyday life. This expansion necessitates the accurate recognition and prediction of human intentions to provide timely and appropriate services. However, existing methods often struggle to perform effectively in complex and unstructured environments. To address this challenge, we propose the Large language model and Knowledge graph based Intention Recognition Framework (LKIRF), which combines large language model (LLM) with knowledge graphs (KG) to enhance the intention recognition capabilities of service robots. Our approach constructs an offline KG from human motion and environmental data and builds an online reasoning graph through real-time interaction, utilizing LLM for interpretation. Experimental results indicate that compared to traditional methods, LKIRF not only improves prediction accuracy across various scenarios but also enhances the transparency and interpretability of the intention reasoning process. [ABSTRACT FROM AUTHOR]

Background: Migraine is a widespread, recurrent primary headache disorder primarily characterized by severe pulsatile headache, typically on one or both sides. It is often accompanied by nausea, vomiting, and hypersensitivity to sound and light. Despite the availability of multiple drugs for migraine management, the condition often becomes chronic due to untimely or irrational drug use, significantly distressing patients and increasing the burden on families and society. Over the past two decades, numerous clinical studies on migraine have been published. This study aimed to provide a comprehensive summary of the current status and trends of migraine clinical trials through bibliometric analysis. Methods: We used visual network tools such as CiteSpace and VOSviewer to perform a knowledge graph analysis of publications related to migraine clinical trials extracted from the WoSCC. Results: This study analyzed 1,129 articles published in 389 journals from 61 countries. The number of publications on migraine clinical trials has steadily increased from 2004 to 2023. The United States and Albert Einstein College of Medicine are the leading countries and institutions in this field, respectively. Richard B. Lipton is the most prolific author, making significant contributions to the research. The journal Headache has the highest number of publications and citations in this area. Keywords such as "efficacy," "RCT," "CGRP," "prophylaxis," "disability," "depression," "questionnaire," and "real-world effectiveness" received significant attention. Conclusion: This study identified reliable research hotspots and provided directions for clinicians. The treatment of migraine continues to be challenging. Future trends may include continued growth in migraine classification, risk factor analysis, and comorbidity studies. Research on CGRP and epigenetics will advance the progress of precision medicine in the migraine field. [ABSTRACT FROM AUTHOR]

Speech recognition based on large neural network models requires vast computing resources and storage, and this requirement severely limits their direct application in resource-constrained edge devices. To considerably reduce the size and complexity of models while retaining the core performance of the original models, this study proposed a novel graph representation augmented with knowledge distillation (GraphRAG) speech recognition method based on adaptive multilevel distillation pruning (AMDP). First, graph neural networks were used to build and optimize knowledge graph embeddings to enhance the model's semantic understanding of speech signals. Second, an adaptive multilevel distillation mechanism was introduced to build a teacher--student architecture composed of multiple models of different sizes. The large bottom-layer model served as the main teacher, and the student models at each layer sequentially learned from the teacher model of the previous layer. Third, after each round of distillation, the student model was refined and pruned to further compress the model size while maintaining high recognition accuracy. Last, detailed experimental tests were conducted on multiple benchmark speech recognition data sets to verify the effectiveness of AMDP-GraphRAG technology. Results demonstrate that, (1) AMDP-GraphRAG consumes minimal computing time and memory while ensuring a low word error rate (WER). WER with and without a graph is 8% and 10%, respectively. The memory usage of the baseline is about 500 MB, whereas the memory usage of AMDP-GraphRAG is about 350 MB, which is the lowest among all the values produce by all the compared models. (2) The AMDP-GraphRAG model can maintain or improve recognition accuracy while reducing the number of parameters. The baseline has 120 million parameters, and AMDP has only 10 million parameters. Moreover, the baseline's WER is 3.5%, and that of AMDP-GraphRAG is 2.8%. (3) Atlas information can enhance the generalization ability of the models. Models with graph information can adapt quickly and show improved recognition accuracy. This study can considerably reduce the scale of models, decrease computational complexity and resource consumption, and serve as a reference for improving the application of speech recognition technology in edge devices. [ABSTRACT FROM AUTHOR]

Under the global warming, the research on ecological effects caused by extreme climate events has received widespread attention. Based on a total of 913 literatures collected by the China National Knowledge Infrastructure (CNKI) and the Web of Science Core Collection database from 2000 to 2020, CiteSpace software was applied to visualize and analyze the current research status, progresses and trends of ecological effects caused by extreme climate events. The results showed that the number of publications in this field experienced three phases: slow growth, steady growth, and rapid growth. The number of publications in China was seriously insufficient and relatively scattered compared with that in foreign countries over the same period. The institutions and authors with the highest publications were the Chinese Academy of Sciences and Beierkuhnlein Carl, respectively. Keywords with high frequency of appearance included "climate change", "drought", "response", "temperature", and "extreme climate". Current research on the ecological effects of extreme climate events mainly tends to focus on the impacts of extreme drought, extreme precipitation and extreme temperature on terrestrial ecosystems. The carbon cycle, resistance and stability, sensitivity and dynamic evolution of ecosystems under extreme climate have been research hotspots in recent years. In the future, it is necessary to standardize the terminology related to extreme climate events and to carry out long-term ecological effects research on discrete and compound extreme climate eventsusing new techniques methods. [ABSTRACT FROM AUTHOR]

The application of wireless communication principles and scenarios in the field of intelligent logistics has triggered innovation and patent output in the wireless communication domain. This study analyses the patent network of wireless communication collaborative innovation in the logistics field, covering the period from 2000 to 2020, and identifies the main innovation entities and changes in the network structure. The number of patent holders increased from 14 individuals in 2000–2006 to 525 individuals in 2014–2020, with universities taking the lead followed by enterprises and research institutions. The node degree grew 150 times, from 6 to 1500. However, the network density decreased from 0.0659 to 0.0109, indicating that despite the increase in the number of innovation entities, the collaboration among them is not close. The overall network centrality is relatively low, but some central nodes emerged in the later stage of the study, including universities, research institutions, and state-owned enterprises. The study suggests the need to strengthen collaborative innovation activities and knowledge sharing as the network continues to expand. In terms of the distribution of underlying technologies, "wireless communication in logistics" primarily focuses on technologies such as NEW RADIO (NR), NEW CORE (NC), and LTE. [ABSTRACT FROM AUTHOR]

The existing knowledge graph embedding (KGE) models based on graph neural networks (GNNs) typically aggregate unreliable neighboring node information, leading to a decrease in the accuracy of inference results. Additionally, incomplete triples commonly found in knowledge graphs(KGs) pose challenges for KGE models to conduct inductive reasoning. To address these issues, this paper proposes a KGE model named MAKE, which is based on GNN and multi-level analogical reasoning. MAKE model features include: (1) Utilization of multi-level semantic information: MAKE explicitly models semantic information at the levels of relations, entities, and triples using GNN. Through multi-level aggregation, it effectively combines semantic information to generate original aggregated representations, balancing the noise impact from neighbors and enhancing model robustness. (2) Multi-level analogical reasoning: MAKE introduces analogical reasoning methods at the aforementioned levels to generate analogical aggregated representations, thus compensating for the limitations of inductive reasoning. (3) Integration of inductive and analogical reasoning capabilities: By concatenating two representations using a transformation matrix, MAKE produces the final aggregated representation of semantic information, enabling the model to conduct inference more effectively. Extensive experiments demonstrate that MAKE outperforms all baseline models significantly on the FB15K-237 and WN18RR datasets. Particularly noteworthy is its performance on the complex dataset FB15K-237, where MAKE surpasses baseline values by 1 and 2.3% on the MRR and Hits@1 metrics, respectively. [ABSTRACT FROM AUTHOR]

User–item bipartite graphs and knowledge graphs are frequently employed in recommender systems due to their ability to provide rich information for user and item modeling. However, existing recommender systems predominantly focus on modeling either the user or item individually, with few studies simultaneously considering both aspects. In this paper, we propose a novel double-ended recommendation model, DEKGCI, which aims to fully leverage the advantages of these two information sources. Specifically, DEKGCI harnesses the high-order connectivity information in the user–item bipartite graph to extract user representations, while utilizing the semantic information in the knowledge graph to enrich item representations. By doing so, DEKGCI concurrently learns both user and item representations, effectively capturing the intricate interaction information between users and items. The DEKGCI model was evaluated on three real-world datasets. Computational results demonstrate the high effectiveness of the proposed DEKGCI model compared to seven state-of-the-art reference methods from the literature. In particular, compared to the best-performing KFGAN model, DEKGCI achieved AUC gains of 0.335% in movie recommendations, and F1 gains of 0.023%, 2.203%, and 0.530% in movie, book, and music recommendations, respectively. The code and data are available at https://github.com/miaomiao924/DEKGCI. [ABSTRACT FROM AUTHOR]

The efficient analysis and interpretation of biological sequence data remain major challenges in bioinformatics. Graphical representation, as an emerging and effective visualization technique, offers a more intuitive method for analyzing DNA sequences. However, many visualization approaches are dispersed across research databases, requiring urgent organization, integration, and analysis. Additionally, no single visualization method excels in all aspects. To advance these methods, knowledge graphs and advanced machine learning techniques have become key areas of exploration. This paper reviews the current 2D and 3D DNA sequence visualization methods and proposes a new research direction focused on constructing knowledge graphs for biological sequence visualization, explaining the relevant theories, techniques, and models involved. Additionally, we summarize machine learning techniques applicable to sequence visualization, such as graph embedding methods and the use of convolutional neural networks (CNNs) for processing graphical representations. These machine learning techniques and knowledge graphs aim to provide valuable insights into computational biology, bioinformatics, genomic computing, and evolutionary analysis. The study serves as an important reference for improving intelligent search systems, enriching knowledge bases, and enhancing query systems related to biological sequence visualization, offering a comprehensive framework for future research. [ABSTRACT FROM AUTHOR]

The demand for functionality, performance, safety, and reliability in complex product domains is continuously increasing, leading to growing product complexity. Traditional document-based systems engineering approaches face numerous challenges in terms of communication efficiency, traceability, and maintenance when dealing with complex product system design. Consequently, Model-Based Systems Engineering (MBSE) has been widely adopted. To improve the efficiency of system modelling, an effective method is needed to automate the process by reusing existing design resources. This study proposes a knowledge graph-based approach for the automatic generation of system model diagrams, consisting of three main steps. Firstly, a system model graph is constructed using an existing system model repository. Then, the required modelling elements are extracted from design documents, and reusable model elements are obtained based on the system model graph. Lastly, a mapping relationship is established between system model elements extracted from multi-source heterogeneous data and SysML metamodel, enabling the automatic generation of system model diagrams based on SysML views. This approach effectively leverages the information from the existing system model repository and achieves automated mapping conversion from multi-source heterogeneous data to the SysML system model. A communication satellite case study is presented to demonstrate the capability of this method. [ABSTRACT FROM AUTHOR]

Owing to the rapid development of Earth observation and Internet technology, researchers have acquired and shared a large amount of Earth observation data. However, traditional data sharing does not provide direct solutions to problems. The large amount of tacit knowledge contained in scientific data, scientific literature, analysis models, software/code, documentation, and other scientific resources on Earth observation applications has not been effectively organized and shared. To solve this problem, the Group on Earth Observations proposed an Earth Observation Knowledge Hub (EOKH); however, there is no unified and clear method for building an EOKH to date. This paper presents an automatic construction method for an EOKH on the basis of a knowledge graph, which describes scientific data, scientific literature, analysis models, software/code, documentation, and other scientific resources and their semantic relationships. An automatic discovery algorithm of scientific and technological resources was also constructed in this study on the basis of a knowledge graph from the Internet. This algorithm is capable of the automatic creation of knowledge packages and the construction of links between knowledge elements. Then, the knowledge discovery algorithm was evaluated through comparison with an existing method in relation to accuracy, and the results showed that our method outperforms the existing method. Lastly, the knowledge package was published on the Linked Open Data Cloud platform in the Resource Description Framework format, and an EOKH was created. Moreover, an application terminal based on SPARQL allowing users to search the EOKH was developed. A clear and operational method for the construction of an EOKH is proposed for the first time in this research, laying the foundation for the development of the EOKH. [ABSTRACT FROM AUTHOR]

This article has been retracted. A retraction notice can be found at . [ABSTRACT FROM AUTHOR]

Existing multi-hop knowledge graph question answering (KGQA) methods, which attempt to mitigate knowledge graph (KG) sparsity by introducing external text repositories instead of leveraging the question-answer information itself, ignore the semantic gap between the question modality and the knowledge graph modality as well as the role played by neighboring entities in the best answer selection. To address the above problems, we propose a Joint Reasoning-based Embedded Multi-hop KGQA (JREM-KGQA) method, which addresses these issues through three key innovations: 1) Early Joint Embedding. We construct a Question Answering-Knowledge Graph-Collaborative Work Diagram (QA-KG-CWD) and train the diagram using a knowledge graph embedding (KGE) model. This not only alleviates the knowledge graph sparsity but also effectively enhances the model's long-path reasoning ability. 2) Semantic Fusion Module. We narrowed the semantic gap between the question modality and the knowledge graph modality through the semantic fusion module to achieve more effective reasoning. 3) Node Relevance Scoring. We employ three node relevance scoring strategies to ensure that the best answer is selected from the huge knowledge graph. We evaluated our model on MetaQA as well as PQL datasets and compared it with other methods. The results demonstrate that our proposed model outperforms existing methods in terms of long-path reasoning ability, effective mitigation of knowledge graph sparsity, and overall performance. We have made our models source code available at github: [ABSTRACT FROM AUTHOR]

In this age of big data and natural language processing, to what extent can we leverage new technologies and new tools to make progress in organizing disparate biomedical data sources? Imagine a system in which one could bring together sequencing data with phenotypes, gene expression data, and clinical information all under the same conceptual heading where applicable. Bio-ontologies seek to carry this out by organizing the relations between concepts and attaching the data to their corresponding concept. However, to accomplish this, we need considerable time and human input. Instead of resorting to human input alone, we describe a novel approach to obtaining the foundation for bio-ontologies: obtaining propositions (links between concepts) from biomedical text so as to fill the ontology. The heart of our approach is applying logic rules from Aristotelian logic and natural logic to biomedical information to derive propositions so that we can have material to organize knowledge bases (ontologies) for biomedical research. We demonstrate this approach by constructing a proof-of-principle bio-ontology for COVID-19 and related diseases. [ABSTRACT FROM AUTHOR]

Legal knowledge involves multidimensional heterogeneous knowledge such as legal provisions, judicial interpretations, judicial cases, and defenses, which requires extremely high relevance and accuracy of knowledge. Meanwhile, the construction of a legal knowledge reasoning system also faces challenges in obtaining, processing, and sharing multisource heterogeneous knowledge. The knowledge graph technology, which is a knowledge organization form with triples as the basic unit, is able to efficiently transform multisource heterogeneous information into a knowledge representation form close to human cognition. Taking the automated construction of the Chinese legal knowledge graph (CLKG) as a case scenario, this paper presents a joint knowledge enhancement model (JKEM), where prior knowledge is embedded into a large language model (LLM), and the LLM is fine-tuned through the prefix of the prior knowledge data. Under the condition of freezing most parameters of the LLM, this fine-tuning scheme adds continuous deep prompts as prefix tokens to the input sequences of different layers, which can significantly improve the accuracy of knowledge extraction. The results show that the knowledge extraction accuracy of the JKEM in this paper reaches 90.92 % . Based on the superior performance of this model, the CLKG is further constructed, which contains 3480 knowledge triples composed of 9 entities and 2 relationships, providing strong support for an in-depth understanding of the complex relationships in the legal field. [ABSTRACT FROM AUTHOR]

Over the years, the field of landslide disaster research has amassed a wealth of data and specialized knowledge. However, these resources originate from a wide array of sources and often feature complex data structures, highlighting a persistent lack of methods to integrate multi-source, heterogeneous data. Traditional landslide monitoring methods typically focus on singular monitoring targets and data sources, which limits a comprehensive understanding of the complex processes involved in landslides. This paper introduces a landslide monitoring model based on a knowledge graph. This model employs P-Tuning to fine-tune ChatGLM2 for the extraction of triples. Differential InSAR (D-InSAR) is utilized to extract ground deformation data, which is then integrated with the knowledge graph for landslide monitoring and analysis. This study focuses on the co-seismic landslide in Jishishan, Gansu, China. By analyzing the landslide knowledge graph and the spatiotemporal deformation map, the results are as follows: (1) For this event, 106 entities and attributes were constructed, along with two recommended calculation routes. (2) The deformation at the earthquake's central region reached up to 8.784 cm, with a slightly smaller deformation zone to the northwest peaking at 9.662 cm. Significant unilateral subsidence was observed in the mountain range to the southwest. (3) The area affected by the co-seismic landslide primarily includes farmland and villages, covering an area of 0.3408 square kilometers. (4) Analysis based on the knowledge graph indicates that this landslide was primarily caused by the rapid liquefaction of water-saturated soil layers due to the earthquake, resulting in instability. This study contributes to the analysis of post-disaster losses, attribution, and impacts. [ABSTRACT FROM AUTHOR]

The advent of the Internet has led to the emergence of multi-source heterogeneous knowledge graphs, which have become a crucial means of storing and disseminating knowledge. Node importance estimation is a technique that is employed extensively in a number of fields, including recommender systems, intelligent search and resource allocation. This study introduces a Multi-perspective attention mechanism Fusion Algorithm for the mapping of multi-perspective features of knowledge graphs to Node Importance Estimation (MFA-NIE). First, structural embedding features, relational predicate features, and attribute features (both textual and quantitative) are established for the nodes. Subsequently, an enhanced attention mechanism is employed to extract, compress, and fuse these features. The fused hidden layer vector is employed in the design of a key-based attention mechanism, which enables the propagation of messages from neighboring nodes to the source node. This process results in the iterative updating of the source node's hidden features. Finally, TOPSIS centrality, based on the topology of the source node, is employed to dynamically adjust the mapping between the fused features and node importance. Experiments were conducted on real-world, large-scale, multi-source, heterogeneous knowledge graphs, comparing the MFA-NIE algorithm with traditional and advanced baseline algorithms, including PR, PPR, GENI, RGTN, CLINE, MCRL, and others. The results demonstrate that the MFA-NIE algorithm significantly improves effectiveness and accuracy, showcasing its superiority, versatility, and practical application value. [ABSTRACT FROM AUTHOR]

The knowledge graph is a structured representation of knowledge, which typically employs a triplet to convey information. However, the long-tail distribution that characterizes most large-scale knowledge bases poses serious challenges regarding entity and relationship information in the long-tail segment, such as data sparsity and incomplete knowledge. To address these challenges and enhance the completeness of the knowledge graph, this paper proposes a novel model for knowledge representation learning in the context of knowledge graph completion. Specifically, the model maps entities to a two-dimensional Minkowski space and defines relationships as rotations. Additionally, the Adam optimizer is employed to optimize the training process. Finally, the model's efficacy is evaluated via link prediction tasks using public datasets FB15k, WN18, FB15k-237, and WB18RR, with results being compared to those derived from classical knowledge representation models. The empirical analysis reveals that the proposed model achieves superior performance across most evaluation metrics. [ABSTRACT FROM AUTHOR]

A novel framework called TravelRAG is introduced in this paper, which is built upon a large language model (LLM) and integrates Retrieval-Augmented Generation (RAG) with knowledge graphs to create a retrieval system framework designed for the tourism domain. This framework seeks to address the challenges LLMs face in providing precise and contextually appropriate responses to domain-specific queries in the tourism field. TravelRAG extracts information related to tourist attractions from User-Generated Content (UGC) on social media platforms and organizes it into a multi-layer knowledge graph. The travel knowledge graph serves as the core retrieval source for the LLM, enhancing the accuracy of information retrieval and significantly reducing the generation of erroneous or fabricated responses, often termed as "hallucinations". As a result, the accuracy of the LLM's output is enhanced. Comparative analyses with traditional RAG pipelines indicate that TravelRAG significantly boosts both the retrieval efficiency and accuracy, while also greatly reducing the computational cost of model fine-tuning. The experimental results show that TravelRAG not only outperforms traditional methods in terms of retrieval accuracy but also better meets user needs for content generation. [ABSTRACT FROM AUTHOR]

Online learning has been greatly widespread since the explosion of COVID-19. However, due to the lack of interaction between teachers and students in online courses, it is very difficult for students to focus on course content and complete the learning. Therefore, we developed a novel intelligent multilevel knowledge graph to help students quickly and systematically grasp the framework and key content of video lectures. Specifically, our method introduced cues into the interactive knowledge mapping. This approach not only allows students to interact with the graphs but also displays different levels of knowledge based on its importance. In addition, a quasi-experiment was implemented to verify the effectiveness of our method. Specifically, we compare the effects of knowledge graphs in three weighting strategies for cues with traditional manual concept maps on students' learning effectiveness and learning perception. Our study shows that although static artificial concept maps do enhance student learning to a certain extent compared to traditional learning, its development efficiency and enhancement of student learning outcomes are not ideal. In contrast, interactive graphs presenting overlay cues can effectively help students to restructure and systematize their knowledge, direct their attention and thus improve their learning outcomes. [ABSTRACT FROM AUTHOR]

Knowledge graphs have recently been introduced to the verification strategy field successfully representing the complexity of verification in real‐life applications. This format provides a scale‐free analysis of verification strategies compared to the more traditional verification artifacts such as requirement traceability matrices and verification matrices. Complexities can be observed visually and numerically both in terms of the problem scope and the entity interdependencies. In this paper, we retrieve verification strategy information patterns representing different aspects of verification. This is achieved by tapping into the network properties of knowledge graphs. They are dissected to detect knowledge patterns emerging from different parts of the verification artifacts. Similarities and differences between the two verification strategies are explained numerically and semantically. Seemingly unrelated requirements and verification activities are connected through indirect connections, and orthogonalities between independent requirements are analyzed. These findings validate the scalability of verification planning and assessment based on knowledge graphs. [ABSTRACT FROM AUTHOR]

The rapid growth of biomedical literature makes it challenging for researchers to stay current. Integrating knowledge from various sources is crucial for studying complex biological systems. Traditional text-mining methods often have limited accuracy because they don't capture semantic and contextual nuances. Deep-learning models can be computationally expensive and typically have low interpretability, though efforts in explainable AI aim to mitigate this. Furthermore, transformer-based models have a tendency to produce false or made-up information—a problem known as hallucination—which is especially prevalent in large language models (LLMs). This study proposes a hybrid approach combining text-mining techniques with graph neural networks (GNNs) and fine-tuned large language models (LLMs) to extend biomedical knowledge graphs and interpret predicted edges based on published literature. An LLM is used to validate predictions and provide explanations. Evaluated on a corpus of experimentally confirmed protein interactions, the approach achieved a Matthews correlation coefficient (MCC) of 0.772. Applied to insomnia, the approach identified 25 interactions between 32 human proteins absent in known knowledge bases, including regulatory interactions between MAOA and 5-HT2C, binding between ADAM22 and 14-3-3 proteins, which is implicated in neurological diseases, and a circadian regulatory loop involving RORB and NR1D1. The hybrid GNN-LLM method analyzes biomedical literature efficiency to uncover potential molecular interactions for complex disorders. It can accelerate therapeutic target discovery by focusing expert verification on the most relevant automatically extracted information. [ABSTRACT FROM AUTHOR]

Graph databases are becoming increasingly popular across scientific disciplines, being highly suitable for storing and connecting complex heterogeneous data. In systems biology, they are used as a backend solution for biological data repositories, ontologies, networks, pathways, and knowledge graph databases. In this review, we analyse all publications using or mentioning graph databases retrieved from PubMed and PubMed Central full-text search, focusing on the top 16 available graph databases, Publications are categorized according to their domain and application, focusing on pathway and network biology and relevant ontologies and tools. We detail different approaches and highlight the advantages of outstanding resources, such as UniProtKB, Disease Ontology, and Reactome, which provide graph-based solutions. We discuss ongoing efforts of the systems biology community to standardize and harmonize knowledge graph creation and the maintenance of integrated resources. Outlining prospects, including the use of graph databases as a way of communication between biological data repositories, we conclude that efficient design, querying, and maintenance of graph databases will be key for knowledge generation in systems biology and other research fields with heterogeneous data. [ABSTRACT FROM AUTHOR]

Drug repurposing has emerged as a effective and efficient strategy to identify new treatments for a variety of diseases. One of the most effective approaches for discovering potential new drug candidates involves the utilization of Knowledge Graphs (KGs). This review comprehensively explores some of the most prominent KGs, detailing their structure, data sources, and how they facilitate the repurposing of drugs. In addition to KGs, this paper delves into various artificial intelligence techniques that enhance the process of drug repurposing. These methods not only accelerate the identification of viable drug candidates but also improve the precision of predictions by leveraging complex datasets and advanced algorithms. Furthermore, the importance of explainability in drug repurposing is emphasized. Explainability methods are crucial as they provide insights into the reasoning behind AI-generated predictions, thereby increasing the trustworthiness and transparency of the repurposing process. We will discuss several techniques that can be employed to validate these predictions, ensuring that they are both reliable and understandable. [ABSTRACT FROM AUTHOR]

Predicting drug-target interactions (DTI) is a crucial stage in drug discovery and development. Understanding the interaction between drugs and targets is essential for pinpointing the specific relationship between drug molecules and targets, akin to solving a link prediction problem using information technology. While knowledge graph (KG) and knowledge graph embedding (KGE) methods have been rapid advancements and demonstrated impressive performance in drug discovery, they often lack authenticity and accuracy in identifying DTI. This leads to increased misjudgment rates and reduced efficiency in drug development. To address these challenges, our focus lies in refining the accuracy of DTI prediction models through KGE, with a specific emphasis on causal intervention confidence measures (CI). These measures aim to assess triplet scores, enhancing the precision of the predictions. Comparative experiments conducted on three datasets and utilizing 9 KGE models reveal that our proposed confidence measure approach via causal intervention, significantly improves the accuracy of DTI link prediction compared to traditional approaches. Furthermore, our experimental analysis delves deeper into the embedding of intervention values, offering valuable insights for guiding the design and development of subsequent drug development experiments. As a result, our predicted outcomes serve as valuable guidance in the pursuit of more efficient drug development processes. • The confidence method based on causal intervention is applied to drug target link prediction. • Three data sets and nine KGE models were used to verify the robustness of the proposed method. • The robustness of the results was evaluated by the optimal score and the dimension of the intervention input vector. • A comprehensive introduction to the algorithm and extensive experiments are conducted. [ABSTRACT FROM AUTHOR]

Integrating Large Language Models (LLMs) with Knowledge Graphs (KGs) enhances the interpretability and performance of AI systems. This research comprehensively analyzes this integration, classifying approaches into three fundamental paradigms: KG-augmented LLMs, LLM-augmented KGs, and synergized frameworks. The evaluation examines each paradigm's methodology, strengths, drawbacks, and practical applications in real-life scenarios. The findings highlight the substantial impact of these integrations in fundamentally improving real-time data analysis, efficient decision-making, and promoting innovation across various domains. In this paper, we also describe essential evaluation metrics and benchmarks for assessing the performance of these integrations, addressing challenges like scalability and computational overhead, and providing potential solutions. This comprehensive analysis underscores the profound impact of these integrations on improving real-time data analysis, enhancing decision-making efficiency, and fostering innovation across various domains. [ABSTRACT FROM AUTHOR]

Knowledge representation learning is to project entities and relations into low-dimensional vector spaces, and efficiently calculate entities, relations, and complex semantic associations between them. However, the translation models that fuse multi-source information fail to specially pay attention to the semantic information of important sentences in the textual descriptions of an entity, which lead to the low quality of embedded representation of textual descriptions, and fail to utilize the primary neighbor of the entity to further constrain the entity representation, resulting in the weak knowledge representation learning ability of the models. To address these issues, we propose a joint knowledge representation learning model of sentence vectors weighting and primary neighbor constraints called SW &NC-Joint. First, the weight distribution of the sentences is calculated according to the frequency of the neighbors in each sentence, and then the textual descriptions are represented by the sentence vectors weighting. Secondly, the concept of neighbor association degree is put forward, and the primary neighbor is filtered out for each entity according to the association degree. Finally, the weighted textual description information and triple structure information are jointly trained, the most crucial thing that the model is constrained by using the distance between the entity and its primary neighbor in the training process, so as to improve the representation learning ability of the model. The link prediction and triple classification experiments are carried out on FB15K and WN18 datasets. The experimental results demonstrate that our model significantly outperforms existing state-of-the-art text-enhanced knowledge representation learning model for link prediction and triple classification tasks. [ABSTRACT FROM AUTHOR]

Knowledge-based visual question answering (KB-VQA) requires to answer questions according to the given image with the assistance of external knowledge. Recently, researchers generally tend to design different multimodal networks to extract visual and text semantic features for KB-VQA. Despite the significant progress, 'caption' information, a textual form of image semantics, which can also provide visually non-obvious cues for the reasoning process, is often ignored. In this paper, we introduce a novel framework, the Knowledge Based Caption Enhanced Net (KBCEN), designed to integrate caption information into the KB-VQA process. Specifically, for better knowledge reasoning, we make utilization of caption information comprehensively from both explicit and implicit perspectives. For the former, we explicitly link caption entities to knowledge graph together with object tags and question entities. While for the latter, a pre-trained multimodal BERT with natural implicit knowledge is leveraged to co-represent caption tokens, object regions as well as question tokens. Moreover, we develop a mutual correlation module to discern intricate correlations between explicit and implicit representations, thereby facilitating knowledge integration and final prediction. We conduct extensive experiments on three publicly available datasets (i.e., OK-VQA v1.0, OK-VQA v1.1 and A-OKVQA). Both quantitative and qualitative results demonstrate the superiority and rationality of our proposed KBCEN. [ABSTRACT FROM AUTHOR]

Knowledge graph-based recommender systems integrate user–item interactions with knowledge graph information through Graph Neural Networks (GNNs), demonstrating their effectiveness in addressing data sparsity and cold start issues. However, existing knowledge graph has more invalid connections; these noise connections are amplified during message aggregation in GNNs. To alleviate this problem, this paper proposes a contrastive learning filtering method based on multi-view generation. Traditional denoising methods mainly use edge perturbation or graph diffusion to randomly add or remove data, which are highly uncertain and often destroy the semantic information of the original data. We proposed the multi-view generation-based contrastive learning method can perform multiple different samples on the original data to create contrasting views. This ensures the semantic completeness of the original data while also providing rich semantic samples for contrastive learning. Specifically, the method first performs multiple rounds of data sampling from the user–item interaction graph and the knowledge graph to generate multiple views of items. Subsequently, knowledge embedding techniques are used to vectorize entities and relationships within these views. Finally, a contrastive task uses a designed loss to share semantic info among items, controlling node connections with item similarity. Through this method, the information of neighboring nodes (entities or relationships) can be propagated, the importance of neighboring nodes can be distinguished, and the recommendation result can be more precise. Extensive experiments on three benchmark datasets demonstrate that our proposed multi-view contrastive learning filtering approach significantly enhances performance in knowledge graph-based recommendation tasks. [ABSTRACT FROM AUTHOR]

In this paper, we present a knowledge-aware recommendation model based on neuro-symbolic graph embeddings that encode first-order logic rules. Our approach is based on the intuition that is the basis of neuro-symbolic AI systems: to combine deep learning and symbolic reasoning in one single model, in order to take the best out of both the paradigms. To this end, we start from a knowledge graph (KG) encoding information about users, ratings, and descriptive properties of the items and we design a model that combines background knowledge encoded in logical rules mined from the KG with explicit knowledge encoded in the triples of the KG itself to obtain a more precise representation of users and items. Specifically, our model is based on the combination of: (i) a rule learner that extracts first-order logic rules based on the information encoded in the knowledge graph; (ii) a graph embedding module, that jointly learns a vector space representation of users and items based on the triples encoded in the knowledge graph and the rules previously extracted; (iii) a recommendation module that uses the embeddings to feed a deep learning architecture that provides users with top-k recommendations. In the experimental section, we evaluate the effectiveness of our strategy on three datasets, and the results show that the combination of knowledge graph embeddings and first-order logic rules led to an improvement in the predictive accuracy and in the novelty of the recommendations. Moreover, our approach overcomes several competitive baselines, thus confirming the validity of our intuitions. [ABSTRACT FROM AUTHOR]

Research on named entity recognition (NER) and command-line generation for network security evaluation tools is relatively scarce, and no mature models for recognition or generation have been developed thus far. Therefore, in this study, the aim is to build a specialized corpus for network security evaluation tools by combining knowledge graphs and information entropy for automatic entity annotation. Additionally, a novel NER approach based on the KG-BERT-BiLSTM-CRF model is proposed. Compared to the traditional BERT-BiLSTM model, the KG-BERT-BiLSTM-CRF model demonstrates superior performance when applied to the specialized corpus of network security evaluation tools. The graph attention network (GAT) component effectively extracts relevant sequential content from datasets in the network security evaluation domain. The fusion layer then concatenates the feature sequences from the GAT and BiLSTM layers, enhancing the training process. Upon successful NER execution, in this study, the identified entities are mapped to pre-established command-line data for network security evaluation tools, achieving automatic conversion from textual content to evaluation commands. This process not only improves the efficiency and accuracy of command generation but also provides practical value for the development and optimization of network security evaluation tools. This approach enables the more precise automatic generation of evaluation commands tailored to specific security threats, thereby enhancing the timeliness and effectiveness of cybersecurity defenses. [ABSTRACT FROM AUTHOR]

In specific private cloud scenarios, how to defend against malicious software and ensure data security is one of the current research hotspots, and sandbox is an important detection method. This paper proposes a dynamic behavior detection technique based on sandboxing, which real-time monitors and analyzes malicious software behavior. By improving the sandbox behavior weight, integrating virtual resources, and designing fine-grained access control, the detection accuracy and efficiency are enhanced based on zero trust access control system. The simulated attacks are identified on the testing platform, drawing knowledge graphs, achieving effective discovery and tracing. Meanwhile, this paper verified through experiments that the system consumption of the detection method is within an acceptable range, expanding the detection range and reducing the missed detection rate. [ABSTRACT FROM AUTHOR]

Matching tabular data to a knowledge graph (KG) is critical for understanding the semantic column types, column relationships, and entities of a table. Existing matching approaches rely heavily on core columns that represent primary subject entities on which other columns in the table depend. However, discovering these core columns before understanding the table's semantics is challenging. Most prior works use heuristic rules, such as the leftmost column, to discover a single core column, while an insightful discovery of the core column set that accurately captures the dependencies between columns is often overlooked. To address these challenges, we introduce Dependency-aware Core Column Set Discovery (DaCo), an iterative method that uses a novel rough matching strategy to identify both inter-column dependencies and the core column set. Additionally, DaCo can be seamlessly integrated with pre-trained language models, as proposed in the optimization module. Unlike other methods, DaCo does not require labeled data or contextual information, making it suitable for real-world scenarios. In addition, it can identify multiple core columns within a table, which is common in real-world tables. We conduct experiments on six datasets, including five datasets with single core columns and one dataset with multiple core columns. Our experimental results show that DaCo outperforms existing core column set detection methods, further improving the effectiveness of table understanding tasks. [ABSTRACT FROM AUTHOR]