Structure Of Knowledge Graph Research Articles

Unsafe behavior identification on construction sites by combining computer vision and knowledge graph–based reasoning

PurposeWorkers' unsafe behavior is the main cause of construction safety accidents, thereby highlighting the critical importance of behavior-based management. To compensate for the limitations of computer vision in tackling knowledge-intensive issues, semantic-based methods have gained increasing attention in the field of construction safety management. Knowledge graph provides an efficient and visualized method for the identification of various unsafe behaviors.Design/methodology/approachThis study proposes an unsafe behavior identification framework by integrating computer vision and knowledge graph–based reasoning. An enhanced ontology model anchors our framework, with image features from YOLOv5, COCO Panoptic Segmentation and DeepSORT integrated into the graph database, culminating in a structured knowledge graph. An inference module is also developed, enabling automated the extraction of unsafe behavior knowledge through rule-based reasoning.FindingsA case application is implemented to demonstrate the feasibility and effectiveness of the proposed method. Results show that the method can identify various unsafe behaviors from images of construction sites and provide mitigation recommendations for safety managers by automated reasoning, thus supporting on-site safety management and safety education.Originality/valueExisting studies focus on spatial relationships, often neglecting the diversified spatiotemporal information in images. Besides, previous research in construction safety only partially automated knowledge graph construction and reasoning processes. In contrast, this study constructs an enhanced knowledge graph integrating static and dynamic data, coupled with an inference module for fully automated knowledge-based unsafe behavior identification. It can help managers grasp the workers’ behavior dynamics and timely implement measures to correct violations.

Understanding the structure of knowledge graphs with ABSTAT profiles

While there has been a trend in the last decades for publishing large-scale and highly-interconnected Knowledge Graphs (KGs), their users often get overwhelmed by the task of understanding their content as a result of their size and complexity. Data profiling approaches have been proposed to summarize large KGs into concise and meaningful representations, so that they can be better explored, processed, and managed. Profiles based on schema patterns represent each triple in a KG with its schema-level counterpart, thus covering the entire KG with profiles of considerable size. In this paper, we provide empirical evidence that profiles based on schema patterns, if explored with suitable mechanisms, can be useful to help users understand the content of big and complex KGs. ABSTAT provides concise pattern-based profiles and comes with faceted interfaces for profile exploration. Using this tool we present a user study based on query completion tasks. We demonstrate that users who look at ABSTAT profiles formulate their queries better and faster than users browsing the ontology of the KGs. The latter is a pretty strong baseline considering that many KGs do not even come with a specific ontology to be explored by the users. To the best of our knowledge, this is the first attempt to investigate the impact of profiling techniques on tasks related to knowledge graph understanding with a user study.

Information Extraction of Aviation Accident Causation Knowledge Graph: An LLM-Based Approach

Summarizing the causation of aviation accidents is conducive to enhancing aviation safety. The knowledge graph of aviation accident causation, constructed based on aviation accident reports, can assist in analyzing the causes of aviation accidents. With the continuous development of artificial intelligence technology, leveraging large language models for information extraction and knowledge graph construction has demonstrated significant advantages. This paper proposes an information extraction method for aviation accident causation based on Claude-prompt, which relies on the large-scale pre-trained language model Claude 3.5. Through prompt engineering, combined with a few-shot learning strategy and a self-judgment mechanism, this method achieves automatic extraction of accident-cause entities and their relationships. Experimental results indicate that this approach effectively improves the accuracy of information extraction, overcoming the limitations of traditional methods in terms of accuracy and efficiency in processing complex texts. It provides strong support for subsequently constructing a structured knowledge graph of aviation accident causation and conducting causation analysis of aviation accidents.

Integrating AI with medical industry chain data: enhancing clinical nutrition research through semantic knowledge graphs.

In clinical nutrition research, the medical industry chain generates a wealth of multidimensional spatial data across various formats, including text, images, and semi-structured tables. This data's inherent heterogeneity and diversity present significant challenges for processing and mining, which are further compounded by the data's diverse features, which are difficult to extract. To address these challenges, we propose an innovative integration of artificial intelligence (AI) with the medical industry chain data, focusing on constructing semantic knowledge graphs and extracting core features. These knowledge graphs are pivotal for efficiently acquiring insights from the vast and granular big data within the medical industry chain. Our study introduces the Clinical Feature Extraction Knowledge Mapping ( ) model, designed to augment the attributes of medical industry chain knowledge graphs through an entity extraction method grounded in syntactic dependency rules. The model is applied to real and large-scale datasets within the medical industry chain, demonstrating robust performance in relation extraction, data complementation, and feature extraction. It achieves superior results to several competitive baseline methods, highlighting its effectiveness in handling medical industry chain data complexities. By representing compact semantic knowledge in a structured knowledge graph, our model identifies knowledge gaps and enhances the decision-making process in clinical nutrition research.

Knowledge Graph-Based Integration of Conversational Advisors and Faceted Filtering

Abstract Modern e-commerce websites often provide users with a variety of components, such as faceted filters and conversational recommender systems, that act as product advisors to help them find relevant products. However, these components are often treated separately and presented as independent components, leading to increased cognitive load and disruption in the search process. Also, the reasoning behind the resulting product recommendations is often not transparent. To address these limitations, we propose a novel approach that relies on a knowledge graph structure to seamlessly integrate faceted filtering and conversational advisors based on graphical user interfaces (GUI). Concretely, the knowledge graph is used to suggest filter values and products based on the user’s answers in the advisor, and, conversely, to determine follow-up questions based on the user’s selected filter values. The user interface also visualizes and explains the underlying relationships between answers given to the advisor and relevant product features in the filter component in order to increase the transparency of the search process. We conducted two user studies with a total of 448 participants to compare a system that integrates the different components according to our approach with a baseline system in which the mechanisms operate separately. Sequence analysis of the logged interaction data provided insights into participants’ behavior as they interacted with both systems. The results indicate that displaying recommended products and related explanations directly in the filter component increases acceptance and trust in the system. Also, the combination of a conversational advisor with values displayed in a filter interface, along with explanations of the underlying relationships, significantly contributes to the knowledge and understanding of those product features that are important in terms of the current search goal.

Which is better? Taxonomy induction with learning the optimal structure via contrastive learning

A taxonomy represents a hierarchically structured knowledge graph that forms the infrastructure for various downstream applications, including recommender systems, web search, and question answering. The exploration of automated induction from text corpora has yielded notable taxonomies such as CN-probase, CN-DBpedia, and Zhishi.schema. Despite these efforts, existing taxonomies still face two critical issues that result in sub-optimal hierarchical structures. On the one hand, commonly observed taxonomies exhibit a coarse-grained and “flat” structure, stemming from a noticeable lack of diversity in both nodes and edges. This limitation primarily originates from the biased and homogeneous data distribution. On the other hand, the semantic granularity among “siblings” within these taxonomies remains inconsistent, presenting a challenge in accurately and comprehensively identifying hierarchical relations. To address these issues, this study introduces a novel taxonomy induction framework composed of three meticulously designed components. Initially, we established a seed schema by leveraging statistical information from external data sources as distant supervision to append nodes and edges containing “generic semantics”, thereby rectifying biased data distributions. Subsequently, a clustering algorithm is employed to group the nodes based on their similarities, followed by a refinement operation of the hierarchical relations among these nodes. Building on this seed schema, we propose a fine-grained contrastive learning method in the expansion module to strengthen the utilization of taxonomic structures, consequently boosting the precision of query-anchor matching. Finally, we meticulously scrutinized the hierarchical relations between each query and its siblings to ensure the integrity of the constructed taxonomy. Extensive experiments on real-world datasets validated the efficacy of our proposed framework for constructing well-structured taxonomies.

Applications of knowledge graph in medical and financial fields: Data integration and intelligent decision-making from an interdisciplinary perspective

With the rapid advancement of AI and deep learning technologies, knowledge graphs have emerged as a key technology for improving the performance of intelligent decision-making systems and driving interdisciplinary innovation. This article outlines the core principles and structure of knowledge graphs, including how they construct knowledge networks to support complex queries and intelligent reasoning. It reviews their innovative applications in the healthcare and financial industries, emphasizing their significant roles in data integration, decision support, and risk assessment. In the healthcare domain, knowledge graphs contribute to improving the accuracy of medical diagnoses, accelerating drug discovery, and enabling intelligent semantic searches. In the financial sector, they optimize risk management and aid in fraud prevention. The article also looks ahead to the future potential of knowledge graphs, stressing the importance of interdisciplinary collaboration and technological innovation in their development. It aims to provide valuable references for further research and application of knowledge graphs.

Fair-RGNN: Mitigating Relational Bias on Knowledge Graphs

Knowledge graph data are prevalent in real-world applications, and knowledge graph neural networks (KGNNs) are essential techniques for knowledge graph representation learning. Although KGNN effectively models the structural information from knowledge graphs, these frameworks amplify the underlying data bias that leads to discrimination towards certain groups or individuals in resulting applications. Additionally, as existing debiasing approaches mainly focus on entity-wise bias, eliminating the multi-hop relational bias that pervasively exists in knowledge graphs remains an open question. However, it is very challenging to eliminate relational bias due to the sparsity of the paths that generate the bias and the non-linear proximity structure of knowledge graphs. To tackle the challenges, we propose Fair-KGNN, a KGNN framework that simultaneously alleviates multi-hop bias and preserves the proximity information of entity-to-relation in knowledge graphs. The proposed framework is generalizable to mitigate relational bias for all types of KGNN. Fair-KGNN is applicable to incorporate two stateof- the-art KGNN models, RGCN and CompGCN, to mitigate gender-occupation and nationality-salary bias. The experiments carried out on three benchmark knowledge graph datasets demonstrate that Fair-KGNN can effectively mitigate unfair situations during representation learning while preserving the predictive performance of KGNN models. The source code of the proposed method is available at: https://github.com/ynchuang/Mitigating-Relational-Bias-on-Knowledge-Graphs .

Explainable AI Method for Tinnitus Diagnosis via Neighbor-Augmented Knowledge Graph and Traditional Chinese Medicine: Development and Validation Study.

Tinnitus diagnosis poses a challenge in otolaryngology owing to an extremely complex pathogenesis, lack of effective objectification methods, and factor-affected diagnosis. There is currently a lack of explainable auxiliary diagnostic tools for tinnitus in clinical practice. This study aims to develop a diagnostic model using an explainable artificial intelligence (AI) method to address the issue of low accuracy in tinnitus diagnosis. In this study, a knowledge graph-based tinnitus diagnostic method was developed by combining clinical medical knowledge with electronic medical records. Electronic medical record data from 1267 patients were integrated with traditional Chinese clinical medical knowledge to construct a tinnitus knowledge graph. Subsequently, weights were introduced, which measured patient similarity in the knowledge graph based on mutual information values. Finally, a collaborative neighbor algorithm was proposed, which scored patient similarity to obtain the recommended diagnosis. We conducted 2 group experiments and 1 case derivation to explore the effectiveness of our models and compared the models with state-of-the-art graph algorithms and other explainable machine learning models. The experimental results indicate that the method achieved 99.4% accuracy, 98.5% sensitivity, 99.6% specificity, 98.7% precision, 98.6% F1-score, and 99% area under the receiver operating characteristic curve for the inference of 5 tinnitus subtypes among 253 test patients. Additionally, it demonstrated good interpretability. The topological structure of knowledge graphs provides transparency that can explain the reasons for the similarity between patients. This method provides doctors with a reliable and explainable diagnostic tool that is expected to improve tinnitus diagnosis accuracy.

An embedding model for temporal knowledge graphs with long and irregular intervals

As the basis of various downstream applications such as automated question answering and knowledge reasoning, Temporal Knowledge Graph (TKG) embedding learning has attracted many interests. Recent works have explored several methods to learn dynamic embedding for TKGs and led to success in many applications, however, they are difficult to be optimized. In order to accurately catch semantic information propagation of entity-to-entity and entity-to-relation, as well as track dynamics of entities and relations in continuous time, we proposed a novel TKG embedding model ODETKGE. Instead of using recurrent neural network to model dynamics of semantic representations in continuous time, we leverage neural ordinary differential equations to model the dynamic evolution, which enable us to capture long-range dependencies between embeddings at different timestamps, track the dynamics of TKGs with irregular intervals and remain computationally efficient. Furthermore, to effectively model semantics of knowledge graph structure, we combine entity-relation collaboratively updated graph convolutional networks and neural ordinary differential equations. Additionally, to acquire important semantics such as creation and disappearance of relations, we utilize a graph transition layer. Comparing with RNN based models, the proposed ODETKGE can use fewer parameters to achieve deeper layers, and make the transformation of representations between two timestamps smoother. Extensive experiments have been executed on real-world datasets and the results verify that ODETKGE is superior than the state-of-the-art models in link forecasting, long-interval link forecasting, irregular-interval link forecasting, ablation study and training time.

Hierarchical knowledge graph relationship prediction leverage of axiomatic fuzzy set graph structure

Knowledge graph embedding has found widespread application across various fields due to its inherent structured data. However, some non-graph-based models necessitate improvement in accurately describing complex hierarchical structures and connecting intricate relational paths. Graph neural networks embedding methods provide a promising direction for exploring complex hierarchical modeling, facilitating the flow of information from nodes along relational paths. Nevertheless, the deficient interpretable characterization by the saliency estimation in entities, which contributes to the efficient performance of the model, is difficult to comprehend. This paper proposes a novel model to mitigate the described phenomenon by intensifying the original structure of knowledge graphs. Specifically, it integrates the Axiomatic Fuzzy Set entity semantic extraction framework with a heterogeneous relationship self-attention mechanism to bolster node representation. Next, this framework incorporates a convolutional neural network model to increase the interaction capabilities of entities and relations. This interpretable entity-level attention mechanism complements the structural information of the knowledge graphs, enabling more comprehensive neighbor information capture and improving the ability to identify the correct entity location. Experimental validation on two relationally complex datasets demonstrates superior performance compared to current methods, obtaining a new Hit@1 value of 45.6% on WN18RR.

Knowledge graph question answering based on relational attention enhancement and feature cross-coding

Knowledge graph question answering aims to answer natural language questions using structured knowledge graph data. The key to achieving this is having a correct semantic understanding of the question phrases. Query graph generation is an important step for knowledge graph question answering systems to tackle complex questions. Unlike simple single-hop questions, complex questions often require reasoning between multiple triples to get the right answer due to multiple entities, relationships and constraints, making it difficult to generate correct query graphs. In previous studies, researchers have primarily focused on improving the extraction and representation of question features, neglecting the prior structural information implicated in the question itself. In this paper, we propose a question structure classifier to classify the question structure, and alleviate the noise interference in query graph through classification results. In the classifier, we strengthen the information about the question structure through the attention mechanism, while weakening the irrelevant information. Moreover, a query graph sorting module based on feature cross-coding is proposed to sort candidate paths in the query graph using fine-grained feature interaction between words. Extensive experiments are conducted on two public datasets (MetaQA and WebQuestionsSP) and the experimental results show that the proposed method outperforms other baselines.

Novel Human Activity Recognition by graph engineered ensemble deep learning model

This research delves into the domain of Human Activity Recognition (HAR) through sensor data analysis, offering a comprehensive exploration of three diverse datasets: UniMiB-SHAR, Motion Sense, and WISDM Actitracker. The UniMiB-SHAR dataset encompasses a diverse array of linear as well as non-linear and complex activities which involve the movement of more than one joint or muscle (for example Hitting Obstacles, jogging and falling with face down). This motion generates highly correlated sensor readings over a certain period of time. In this case, Convolution Neural Networks (CNNs) are effective in feature extraction as well as classification of HAR activities, but they may not fully grasp the combined features of spatial as well as temporal aspects in the HAR datasets and heavily rely on labelled data. Whereas, Graph convolution networks (GCN), with their capacity to model complex interactions through graph structure, complement CNN’s capabilities in classifying non-linear activities in the HAR dataset. By leveraging the Knowledge graph structure and acquiring the feature embeddings from the GCN model, in this study, a Noval ensemble CNN model is proposed for the classification of activities. The novel HAR pipeline is termed as Graph Engineered EnsemCNN HAR (GE-EnsemCNN-HAR) and its performance is evaluated on HAR datasets. Proposed model demonstrated a noteworthy accuracy of 93.5% on UniMiB-SHAR dataset, surpassing the Shallow CNN model with GNN with an improvement of 20.14%. The proposed model achieved a notable accuracy rate of 96.18% and 98% when evaluated on the Motion Sense and WISDM Actitracker dataset.

Knowledge graph‐based multimodal neural networks for smart‐grid defect detection

AbstractThe application of artificial intelligence in smart grid operations, particularly for defect detection, has seen remarkable progress. However, effectively leveraging multi‐modal data sources to enhance defect detection performance remains challenging. This paper presents a novel Knowledge Graph‐based Multimodal Neural Network (KGMNN) approach for robust and efficient defect detection in smart grids. Our proposed method first leverages multiple heterogeneous data sources, such as structural data, temporal sensor readings, and maintenance records, creating a comprehensive representation of the grid system. The system is then modeled as a knowledge graph, with nodes representing various entities and edges signifying their interconnections and interactions. This paper introduces a novel neural network architecture that effectively leverages the knowledge graph structure. This architecture incorporates multi‐modal fusion techniques, enabling the model to process and integrate diverse data sources. It can recognize intricate patterns indicative of potential defects that are often undetectable in individual data modes. Experimental results on several real‐world smart grid datasets demonstrate that the KGMNN significantly outperforms state‐of‐the‐art techniques in terms of precision, recall, and computational efficiency. Furthermore, the KGMNN's ability to provide explainable predictions, aided by the inherent interpretability of the knowledge graph structure, is of significant practical value in the field. This research paves the way for a new generation of AI‐powered defect detection in smart grids, facilitating enhanced grid reliability and operational efficiency.

Cybersecurity threat perception technology based on knowledge graph

<p class="Abstract">The issue of complex sources, difficult to understand and share security threat intelligence, this paper realizes deep learning of threat intelligence features based on Restricted Boltzmann Machine, which graphs the original threat intelligence features from high dimensional space to low dimensional space layer by layer, and constructs the cyberspace security threat knowledge graphs. The deep learning used to build a multi-level and structured knowledge graph of cyberspace security threats can reflect the structural characteristics of the knowledge graph, making the graph have a lower dimension and a higher level of abstraction. The experiment verifies the feasibility of constructing the cyberspace security threat knowledge graph, and verifies the security threat perception method based on the knowledge graph is more suitable for the perception of high-intensity security threats by comparing with traditional threat detection methods.<strong></strong></p>

Integrating users’ long-term and short-term interests with knowledge graph to improve restaurant recommendation

User preferences are typically analyzed by recommendation systems for recommending items. However, in most existing restaurant recommendation methods, recommendation similarity (e.g., reviews, sentiment) is considered only from the perspective of users, whereas restaurant attributes and potential connections between restaurants are ignored. Furthermore, the dining interests of users are diverse and not fixed. For improved recommendations, the changes of user interests should be considered. To address these problems, this study proposed a recommendation approach that combines knowledge graphs and long-term and short-term interests of users. In the approach, user preferences were analyzed from the attributes of restaurant using the semantic structure of knowledge graphs to obtain long-term user representations. Next, we filtered recent historical restaurants based on time series to obtain short-term user representations through the gated recurrent unit. To integrate long- and short-term interests, a novel gated structure was designed to obtain dynamic interest weights. Extensive experiments on a real-world dataset revealed that the proposed approach outperformed baseline methods.

Learning from Hierarchical Structure of Knowledge Graph for Recommendation

Knowledge graphs (KGs) can help enhance recommendations, especially for the data-sparsity scenarios with limited user-item interaction data. Due to the strong power of representation learning of graph neural networks (GNNs), recent works of KG-based recommendation deploy GNN models to learn from both knowledge graph and user-item bipartite interaction graph. However, these works have not well considered the hierarchical structure of knowledge graph, leading to sub-optimal results. Despite the benefit of hierarchical structure, leveraging it is challenging since the structure is always partly-observed. In this work, we first propose to reveal unknown hierarchical structures with a supervised signal detection method and then exploit the hierarchical structure with disentangling representation learning. We conduct experiments on two large-scale datasets, of which the results well verify the superiority and rationality of the proposed method. Further experiments of ablation study with respect to key model designs have demonstrated the effectiveness and rationality of our proposed model. The code is available at https://github.com/tsinghua-fib-lab/HIKE .

Attributed Abnormality Graph Embedding for Clinically Accurate X-Ray Report Generation.

Despite the recent success of deep learning models for text generation, generating clinically accurate reports remains challenging. More precisely modeling the relationships of the abnormalities revealed in an X-ray image has been found promising to enhance the clinical accuracy. In this paper, we first introduce a novel knowledge graph structure called an attributed abnormality graph (ATAG). It consists of interconnected abnormality nodes and attribute nodes for better capturing more fine-grained abnormality details. In contrast to the existing methods where the abnormality graph are constructed manually, we propose a methodology to automatically construct the fine-grained graph structure based on annotated X-ray reports and the RadLex radiology lexicon. We then learn the ATAG embeddings as part of a deep model with an encoder-decoder architecture for the report generation. In particular, graph attention networks are explored to encode the relationships among the abnormalities and their attributes. A hierarchical attention attention and a gating mechanism are specifically designed to further enhance the generation quality. We carry out extensive experiments based on the benchmark datasets, and show that the proposed ATAG-based deep model outperforms the SOTA methods by a large margin in ensuring the clinical accuracy of the generated reports.

Leveraging Knowledge Graphs and Natural Language Processing for Automated Web Resource Labeling and Knowledge Mobilization in Neurodevelopmental Disorders: Development and Usability Study.

Patients and families need to be provided with trusted information more than ever with the abundance of online information. Several organizations aim to build databases that can be searched based on the needs of target groups. One such group is individuals with neurodevelopmental disorders (NDDs) and their families. NDDs affect up to 18% of the population and have major social and economic impacts. The current limitations in communicating information for individuals with NDDs include the absence of shared terminology and the lack of efficient labeling processes for web resources. Because of these limitations, health professionals, support groups, and families are unable to share, combine, and access resources. We aimed to develop a natural language-based pipeline to label resources by leveraging standard and free-text vocabularies obtained through text analysis, and then represent those resources as a weighted knowledge graph. Using a combination of experts and service/organization databases, we created a data set of web resources for NDDs. Text from these websites was scraped and collected into a corpus of textual data on NDDs. This corpus was used to construct a knowledge graph suitable for use by both experts and nonexperts. Named entity recognition, topic modeling, document classification, and location detection were used to extract knowledge from the corpus. We developed a resource annotation pipeline using diverse natural language processing algorithms to annotate web resources and stored them in a structured knowledge graph. The graph contained 78,181 annotations obtained from the combination of standard terminologies and a free-text vocabulary obtained using topic modeling. An application of the constructed knowledge graph is a resource search interface using the ordered weighted averaging operator to rank resources based on a user query. We developed an automated labeling pipeline for web resources on NDDs. This work showcases how artificial intelligence-based methods, such as natural language processing and knowledge graphs for information representation, can enhance knowledge extraction and mobilization, and could be used in other fields of medicine.

Multi-task feature and structure learning for user-preference based knowledge-aware recommendation

Data sparsity and long-tail items recommendation, two typical problems of recommender systems, can be effectively alleviated by leveraging user-side information and item-side information. Knowledge graph (KG) is a source of side information. Due to the specificity of data distribution, the current KG-enhanced recommendation algorithms have the problem of weak generalization and robustness (the performance varies greatly in multiple data sets). Moreover, the previous knowledge-aware recommendation models are insufficient to distill the collaborative signal and personal features from the collective behaviors of users while simultaneously distilling the connectivity and exclusive entity features from the knowledge graph on the item-side. In this paper, we propose a multi-task framework Multi-Rec with high flexibility and adaptability based on alternating learning, which consists of a recommender system (RS) task and a knowledge graph embedding (KGE) task. The RS task consists of user feature learning and user structure learning modules. Correspondingly, the KGE task consists of knowledge graph feature learning and knowledge graph structure learning modules. The correlation between each sub-learning task is carried out by designed cross units and exchange units. Through the end-to-end framework, our model generates high-efficiency node embeddings for downstream recommender tasks and outperforms several state-of-the-art baselines on four real-world datasets through extensive experiments.