Multi-modal knowledge graph completion(MMKGC)aims to complete missing entities or relations in multi-modal knowledge graphs,thereby discovering more previously unknown triples.Due to the continuous growth of data and ...Multi-modal knowledge graph completion(MMKGC)aims to complete missing entities or relations in multi-modal knowledge graphs,thereby discovering more previously unknown triples.Due to the continuous growth of data and knowledge and the limitations of data sources,the visual knowledge within the knowledge graphs is generally of low quality,and some entities suffer from the issue of missing visual modality.Nevertheless,previous studies of MMKGC have primarily focused on how to facilitate modality interaction and fusion while neglecting the problems of low modality quality and modality missing.In this case,mainstream MMKGC models only use pre-trained visual encoders to extract features and transfer the semantic information to the joint embeddings through modal fusion,which inevitably suffers from problems such as error propagation and increased uncertainty.To address these problems,we propose a Multi-modal knowledge graph Completion model based on Super-resolution and Detailed Description Generation(MMCSD).Specifically,we leverage a pre-trained residual network to enhance the resolution and improve the quality of the visual modality.Moreover,we design multi-level visual semantic extraction and entity description generation,thereby further extracting entity semantics from structural triples and visual images.Meanwhile,we train a variational multi-modal auto-encoder and utilize a pre-trained multi-modal language model to complement the missing visual features.We conducted experiments on FB15K-237 and DB13K,and the results showed that MMCSD can effectively perform MMKGC and achieve state-of-the-art performance.展开更多
Conversational recommender systems(CRSs)focus on refining preferences and providing personalized recommendations through natural language interactions and dialogue history.Large language models(LLMs)have shown outstan...Conversational recommender systems(CRSs)focus on refining preferences and providing personalized recommendations through natural language interactions and dialogue history.Large language models(LLMs)have shown outstanding performance across various domains,thereby prompting researchers to investigate their applicability in recommendation systems.However,due to the lack of task-specific knowledge and an inefficient feature extraction process,LLMs still have suboptimal performance in recommendation tasks.Therefore,external knowledge sources,such as knowledge graphs(KGs)and knowledge bases(KBs),are often introduced to address the issue of data sparsity.Compared to KGs,KBs possess higher retrieval efficiency,making them more suitable for scenarios where LLMs serve as recommenders.To this end,we introduce a novel framework integrating LLMs with KBs for enhanced retrieval generation,namely LLMKB.LLMKB initially leverages structured knowledge to create mapping dictionaries,extracting entity-relation information from heterogeneous knowledge to construct KBs.Then,LLMKB achieves the embedding calibration between user information representations and documents in KBs through retrieval model fine-tuning.Finally,LLMKB employs retrievalaugmented generation to produce recommendations based on fused text inputs,followed by post-processing.Experiment results on two public CRS datasets demonstrate the effectiveness of our framework.Our code is publicly available at the link:https://anonymous.4open.science/r/LLMKB-6FD0.展开更多
Event extraction stands as a significant endeavor within the realm of information extraction,aspiring to automatically extract structured event information from vast volumes of unstructured text.Extracting event eleme...Event extraction stands as a significant endeavor within the realm of information extraction,aspiring to automatically extract structured event information from vast volumes of unstructured text.Extracting event elements from multi-modal data remains a challenging task due to the presence of a large number of images and overlapping event elements in the data.Although researchers have proposed various methods to accomplish this task,most existing event extraction models cannot address these challenges because they are only applicable to text scenarios.To solve the above issues,this paper proposes a multi-modal event extraction method based on knowledge fusion.Specifically,for event-type recognition,we use a meticulous pipeline approach that integrates multiple pre-trained models.This approach enables a more comprehensive capture of the multidimensional event semantic features present in military texts,thereby enhancing the interconnectedness of information between trigger words and events.For event element extraction,we propose a method for constructing a priori templates that combine event types with corresponding trigger words.This approach facilitates the acquisition of fine-grained input samples containing event trigger words,thus enabling the model to understand the semantic relationships between elements in greater depth.Furthermore,a fusion method for spatial mapping of textual event elements and image elements is proposed to reduce the category number overload and effectively achieve multi-modal knowledge fusion.The experimental results based on the CCKS 2022 dataset show that our method has achieved competitive results,with a comprehensive evaluation value F1-score of 53.4%for the model.These results validate the effectiveness of our method in extracting event elements from multi-modal data.展开更多
The knowledge graph with relational abundant information has been widely used as the basic data support for the retrieval platforms.Image and text descriptions added to the knowledge graph enrich the node information,...The knowledge graph with relational abundant information has been widely used as the basic data support for the retrieval platforms.Image and text descriptions added to the knowledge graph enrich the node information,which accounts for the advantage of the multi-modal knowledge graph.In the field of cross-modal retrieval platforms,multi-modal knowledge graphs can help to improve retrieval accuracy and efficiency because of the abundant relational infor-mation provided by knowledge graphs.The representation learning method is sig-nificant to the application of multi-modal knowledge graphs.This paper proposes a distributed collaborative vector retrieval platform(DCRL-KG)using the multi-modal knowledge graph VisualSem as the foundation to achieve efficient and high-precision multimodal data retrieval.Firstly,use distributed technology to classify and store the data in the knowledge graph to improve retrieval efficiency.Secondly,this paper uses BabelNet to expand the knowledge graph through multi-ple filtering processes and increase the diversification of information.Finally,this paper builds a variety of retrieval models to achieve the fusion of retrieval results through linear combination methods to achieve high-precision language retrieval and image retrieval.The paper uses sentence retrieval and image retrieval experi-ments to prove that the platform can optimize the storage structure of the multi-modal knowledge graph and have good performance in multi-modal space.展开更多
In the context of power generation companies, vast amounts of specialized data and expert knowledge have been accumulated. However, challenges such as data silos and fragmented knowledge hinder the effective utilizati...In the context of power generation companies, vast amounts of specialized data and expert knowledge have been accumulated. However, challenges such as data silos and fragmented knowledge hinder the effective utilization of this information. This study proposes a novel framework for intelligent Question-and-Answer (Q&A) systems based on Retrieval-Augmented Generation (RAG) to address these issues. The system efficiently acquires domain-specific knowledge by leveraging external databases, including Relational Databases (RDBs) and graph databases, without additional fine-tuning for Large Language Models (LLMs). Crucially, the framework integrates a Dynamic Knowledge Base Updating Mechanism (DKBUM) and a Weighted Context-Aware Similarity (WCAS) method to enhance retrieval accuracy and mitigate inherent limitations of LLMs, such as hallucinations and lack of specialization. Additionally, the proposed DKBUM dynamically adjusts knowledge weights within the database, ensuring that the most recent and relevant information is utilized, while WCAS refines the alignment between queries and knowledge items by enhanced context understanding. Experimental validation demonstrates that the system can generate timely, accurate, and context-sensitive responses, making it a robust solution for managing complex business logic in specialized industries.展开更多
To address the challenge of missing modal information in entity alignment and to mitigate information loss or bias arising frommodal heterogeneity during fusion,while also capturing shared information acrossmodalities...To address the challenge of missing modal information in entity alignment and to mitigate information loss or bias arising frommodal heterogeneity during fusion,while also capturing shared information acrossmodalities,this paper proposes a Multi-modal Pre-synergistic Entity Alignmentmodel based on Cross-modalMutual Information Strategy Optimization(MPSEA).The model first employs independent encoders to process multi-modal features,including text,images,and numerical values.Next,a multi-modal pre-synergistic fusion mechanism integrates graph structural and visual modal features into the textual modality as preparatory information.This pre-fusion strategy enables unified perception of heterogeneous modalities at the model’s initial stage,reducing discrepancies during the fusion process.Finally,using cross-modal deep perception reinforcement learning,the model achieves adaptive multilevel feature fusion between modalities,supporting learningmore effective alignment strategies.Extensive experiments on multiple public datasets show that the MPSEA method achieves gains of up to 7% in Hits@1 and 8.2% in MRR on the FBDB15K dataset,and up to 9.1% in Hits@1 and 7.7% in MRR on the FBYG15K dataset,compared to existing state-of-the-art methods.These results confirm the effectiveness of the proposed model.展开更多
The network security knowledge base standardizes and integrates network security data,providing a reliable foundation for real-time network security protection solutions.However,current research on network security kn...The network security knowledge base standardizes and integrates network security data,providing a reliable foundation for real-time network security protection solutions.However,current research on network security knowledge bases mainly focuses on their construction,while the potential to optimize intelligent security services for real-time network security protection requires further exploration.Therefore,how to effectively utilize the vast amount of historical knowledge in the field of network security and establish a feedback mechanism to update it in real time,thereby enhancing the detection capability of security services against malicious traffic,has become an important issue.Our contribution is fourfold.First,we design a feedback interface to update the knowledge base with information such as features of attack traffic,detection outcomes from network service functions(NSF),and system resource utilization.Second,we introduce a feature selection method that combines PageRank and RandomForest to identify influential features in the knowledge base and dynamically incorporate them into the NSFs.Third,we propose a path selection method that combines graph attention network(GAT)and deep reinforcement learning(DRL)to learn the local knowledge of the knowledge base and determine the optimal traffic path within the Service Function Chains(SFC).Finally,experimental results demonstrate that the knowledge base can be updated in real time according to feedback information,and the optimized service achieves an accuracy,recall,and F1 score exceeding 96%.Compared to preset paths and paths selected using the deep Q-network(DQN)method,our proposed method increases the malicious traffic detection rate by an average of 12.4%and 4.6%,respectively,enhances the total malicious traffic detection capability(TMTDC)of the path by 18.1%and 11.5%,and significantly reduces path detection delay.It has been verified that the proposed intelligent security optimization method can monitor malicious traffic in real time,update knowledge,and enhance the system’s detection capability against malicious traffic.展开更多
Remote driving,an emergent technology enabling remote operations of vehicles,presents a significant challenge in transmitting large volumes of image data to a central server.This requirement outpaces the capacity of t...Remote driving,an emergent technology enabling remote operations of vehicles,presents a significant challenge in transmitting large volumes of image data to a central server.This requirement outpaces the capacity of traditional communication methods.To tackle this,we propose a novel framework using semantic communications,through a region of interest semantic segmentation method,to reduce the communication costs by transmitting meaningful semantic information rather than bit-wise data.To solve the knowledge base inconsistencies inherent in semantic communications,we introduce a blockchain-based edge-assisted system for managing diverse and geographically varied semantic segmentation knowledge bases.This system not only ensures the security of data through the tamper-resistant nature of blockchain but also leverages edge computing for efficient management.Additionally,the implementation of blockchain sharding handles differentiated knowledge bases for various tasks,thus boosting overall blockchain efficiency.Experimental results show a great reduction in latency by sharding and an increase in model accuracy,confirming our framework's effectiveness.展开更多
The“Opinions on Comprehensively Deepening Curriculum Reform to Fulfill the Fundamental Task of Strengthening Moral Education”,issued by China’s Ministry of Education in 2015,explicitly identified Project-Based Lear...The“Opinions on Comprehensively Deepening Curriculum Reform to Fulfill the Fundamental Task of Strengthening Moral Education”,issued by China’s Ministry of Education in 2015,explicitly identified Project-Based Learning(PBL)as a key strategy for cultivating students’core competencies.Since then,PBL has been widely implemented as a pilot initiative in primary and secondary schools,gaining increasing influence.Analyzing the intellectual foundations of PBL research in China can offer valuable insights into its theoretical and practical dimensions.This study uses CiteSpace to examine 156 PBL-related articles from the CSSCI database,revealing that the knowledge base of PBL research is primarily built on two major domains.The first is the theoretical foundation,characterized by frequently cited literature focusing on the conceptual framework,educational value,interdisciplinary approaches,core competency cultivation,and instructional objectives of PBL.The second is empirical research,where highly cited studies include case analyses across K–12 settings,general high schools,and higher education institutions.Moving forward,future research on PBL should explore its meaning and value from a dual-subject and integrated perspective,expand case studies to include vocational education,and further promote the interdisciplinary development of core competencies through PBL.展开更多
Objective:To explore the effectiveness of multi-modal teaching based on an online case library in the education of gene methylation combined with spiral computed tomography(CT)screening for pulmonary ground-glass opac...Objective:To explore the effectiveness of multi-modal teaching based on an online case library in the education of gene methylation combined with spiral computed tomography(CT)screening for pulmonary ground-glass opacity(GGO)nodules.Methods:From October 2023 to April 2024,66 medical imaging students were selected and randomly divided into a control group and an observation group,each with 33 students.The control group received traditional lecture-based teaching,while the observation group was taught using a multi-modal teaching approach based on an online case library.Performance on assessments and teaching quality were analyzed between the two groups.Results:The observation group achieved higher scores in theoretical and practical knowledge compared to the control group(P<0.05).Additionally,the teaching quality scores were significantly higher in the observation group(P<0.05).Conclusion:Implementing multi-modal teaching based on an online case library for pulmonary GGO nodule screening with gene methylation combined with spiral CT can enhance students’knowledge acquisition,improve teaching quality,and have significant clinical application value.展开更多
Objective: To develop a best-evidence-based optimal nutrition management plan for patients with chronic heart failure, apply it in clinical practice, and evaluate its effectiveness. Methods: Use the KTA knowledge tran...Objective: To develop a best-evidence-based optimal nutrition management plan for patients with chronic heart failure, apply it in clinical practice, and evaluate its effectiveness. Methods: Use the KTA knowledge translation model to guide evidence-based practice in nutrition management, and compare the nutritional status, cardiac function status, quality of life, and quality review indicators of chronic heart failure patients before and after the application of evidence. Results: After the application of evidence, the nutritional status indicators (MNA-SF score, albumin, hemoglobin) of two groups of heart failure patients significantly increased compared to before the application of evidence, with statistically significant differences (p Conclusion: The KTA knowledge translation model provides methodological guidance for the implementation of evidence-based practice for heart failure patients. This evidence-based practice project is beneficial for improving the outcomes of malnutrition in chronic heart failure patients and is conducive to standardizing nursing pathways, thereby promoting the improvement of nursing quality.展开更多
Media convergence works by processing information from different modalities and applying them to different domains.It is difficult for the conventional knowledge graph to utilise multi-media features because the intro...Media convergence works by processing information from different modalities and applying them to different domains.It is difficult for the conventional knowledge graph to utilise multi-media features because the introduction of a large amount of information from other modalities reduces the effectiveness of representation learning and makes knowledge graph inference less effective.To address the issue,an inference method based on Media Convergence and Rule-guided Joint Inference model(MCRJI)has been pro-posed.The authors not only converge multi-media features of entities but also introduce logic rules to improve the accuracy and interpretability of link prediction.First,a multi-headed self-attention approach is used to obtain the attention of different media features of entities during semantic synthesis.Second,logic rules of different lengths are mined from knowledge graph to learn new entity representations.Finally,knowledge graph inference is performed based on representing entities that converge multi-media features.Numerous experimental results show that MCRJI outperforms other advanced baselines in using multi-media features and knowledge graph inference,demonstrating that MCRJI provides an excellent approach for knowledge graph inference with converged multi-media features.展开更多
In this paper, the knowledge based enterprise is considered as an organism, which possesses a set of capabilities. The organizational structure model of knowledge based enterprise organism is described in order to pos...In this paper, the knowledge based enterprise is considered as an organism, which possesses a set of capabilities. The organizational structure model of knowledge based enterprise organism is described in order to possess the essential capacity set. A dynamic capacity set is defined and analyzed based on the definition of the growth and development for knowledge based enterprise organism. The structure of the capacity base, a subset of the capacity set, is optimized for different periods of the organism ...展开更多
The knowledge representation mode and inference control strategy were analyzed according to the specialties of air-conditioning cooling/heating sources selection. The constructing idea and working procedure for knowle...The knowledge representation mode and inference control strategy were analyzed according to the specialties of air-conditioning cooling/heating sources selection. The constructing idea and working procedure for knowledge base and inference engine were proposed while the realization technique of the C language was discussed. An intelligent decision support system (IDSS) model based on such knowledge representation and inference mechanism was developed by domain engineers. The model was verified to have a small kernel and powerful capability in list processing and data driving, which was successfully used in the design of a cooling/heating sources system for a large-sized office building.展开更多
Aim To analyse the influence of knowledge base on the performance of the fuzzy controller of the electrohydraulic position control system,and to determine their selection cri- teria. Methods Experiments based on diffe...Aim To analyse the influence of knowledge base on the performance of the fuzzy controller of the electrohydraulic position control system,and to determine their selection cri- teria. Methods Experiments based on different membership functions,scaling factors and con-trol rules were done separately.The experiment results and the influence of different know- ledge base on the control performance were analysed in theory so that criteria of selcting knowledge base can be summarized correctly.Results Knowledge base,including membershipfunctions, scaling factors and control rules,has a crucial effect on the fuzzy control system.Suitably selected knowledge base can lead to good control performance of fuzzy control sys-tem. Conclusion Being symmetric,having an intersection ratio of 1 and satisfying width con- dition are three necessities for selecting membership functions.Selecting scaling factors dependson both the system requirement and a comprehensive analysis in the overshoot,oscillation, rising time and stability. Integrity and continuity must be guaranteed when determining control rules.展开更多
To semantically integrate heterogeneous resources and provide a unified intelligent access interface, semantic web technology is exploited to publish and interlink machineunderstandable resources so that intelligent s...To semantically integrate heterogeneous resources and provide a unified intelligent access interface, semantic web technology is exploited to publish and interlink machineunderstandable resources so that intelligent search can be supported. TCMSearch, a deployed intelligent search engine for traditional Chinese medicine （TCM）, is presented. The core of the system is an integrated knowledge base that uses a TCM domain ontology to represent the instances and relationships in TCM. Machine-learning techniques are used to generate semantic annotations for texts and semantic mappings for relational databases, and then a semantic index is constructed for these resources. The major benefit of representing the semantic index in RDF/OWL is to support some powerful reasoning functions, such as class hierarchies and relation inferences. By combining resource integration with reasoning, the knowledge base can support some intelligent search paradigms besides keyword search, such as correlated search, semantic graph navigation and concept recommendation.展开更多
In order to improve the learning ability of robots, we present a reinforcement learning approach with a knowledge base for mapping natural language instructions to executable action sequences. A simulated platform wit...In order to improve the learning ability of robots, we present a reinforcement learning approach with a knowledge base for mapping natural language instructions to executable action sequences. A simulated platform with physical engine is built as interactive environment. Based on the knowledge base, a reward function with immediate rewards and delayed rewards is designed to handle sparse reward problems. Also, a list of object states is produced by retrieving the knowledge base, as a standard to define the quality of action sequences. Experimental results demonstrate that our approach yields good performance on accuracy of action sequences production.展开更多
QNET-CFD is a thematic network on quality and trust for the industrial applications of Computational Fluid Dynamics （CFD）, developed under the European Union R＆D program. The main objectives of QNET-CFD were to col...QNET-CFD is a thematic network on quality and trust for the industrial applications of Computational Fluid Dynamics （CFD）, developed under the European Union R＆D program. The main objectives of QNET-CFD were to collect CFD and experimental data in a systematic and quality controlled way and to set the basis for a consistent Knowledge Base in support of CFD guidance and validation. The QNET-CFD activity was organized around six Thematic Areas （TAs） covering the following industry sectors： external aerodynamics; combustion ＆ heat transfer; chemical process, thermal hydraulics and nuclear safety; civil construction ＆ HVAC; environment; turbomachinery internal flows. The main outcome of the QNET-CFD actions is the Knowledge Base （KB） with contains in a user oriented interface, extensive experimental and CFD data for a large number of test cases subdivided into 53 Application Challenges （AC） and 43 Underlying Flow Regimes （UFR）. The KB contains, in addition to state-of-the-art reviews for each of the six thematic areas, Best Practice Advice （BPA） in the use of CFD for most of AC. This is considered as a significant contribution form the QNET-CFD activities and it is expected that the level of the thrust and quality in CFD will hereby be improved.展开更多
To improve the efficiency and accuracy of carbonate reservoir research,a unified reservoir knowledge base linking geological knowledge management with reservoir research is proposed.The reservoir knowledge base serves...To improve the efficiency and accuracy of carbonate reservoir research,a unified reservoir knowledge base linking geological knowledge management with reservoir research is proposed.The reservoir knowledge base serves high-quality analysis,evaluation,description and geological modeling of reservoirs.The knowledge framework is divided into three categories:technical service standard,technical research method and professional knowledge and cases related to geological objects.In order to build a knowledge base,first of all,it is necessary to form a knowledge classification system and knowledge description standards;secondly,to sort out theoretical understandings and various technical methods for different geologic objects and work out a technical service standard package according to the technical standard;thirdly,to collect typical outcrop and reservoir cases,constantly expand the content of the knowledge base through systematic extraction,sorting and saving,and construct professional knowledge about geological objects.Through the use of encyclopedia based collaborative editing architecture,knowledge construction and sharing can be realized.Geological objects and related attribute parameters can be automatically extracted by using natural language processing(NLP)technology,and outcrop data can be collected by using modern fine measurement technology,to enhance the efficiency of knowledge acquisition,extraction and sorting.In this paper,the geological modeling of fracture-cavity reservoir in the Tarim Basin is taken as an example to illustrate the construction of knowledge base of carbonate reservoir and its application in geological modeling of fracture-cavity carbonate reservoir.展开更多
基金funded by Research Project,grant number BHQ090003000X03。
文摘Multi-modal knowledge graph completion(MMKGC)aims to complete missing entities or relations in multi-modal knowledge graphs,thereby discovering more previously unknown triples.Due to the continuous growth of data and knowledge and the limitations of data sources,the visual knowledge within the knowledge graphs is generally of low quality,and some entities suffer from the issue of missing visual modality.Nevertheless,previous studies of MMKGC have primarily focused on how to facilitate modality interaction and fusion while neglecting the problems of low modality quality and modality missing.In this case,mainstream MMKGC models only use pre-trained visual encoders to extract features and transfer the semantic information to the joint embeddings through modal fusion,which inevitably suffers from problems such as error propagation and increased uncertainty.To address these problems,we propose a Multi-modal knowledge graph Completion model based on Super-resolution and Detailed Description Generation(MMCSD).Specifically,we leverage a pre-trained residual network to enhance the resolution and improve the quality of the visual modality.Moreover,we design multi-level visual semantic extraction and entity description generation,thereby further extracting entity semantics from structural triples and visual images.Meanwhile,we train a variational multi-modal auto-encoder and utilize a pre-trained multi-modal language model to complement the missing visual features.We conducted experiments on FB15K-237 and DB13K,and the results showed that MMCSD can effectively perform MMKGC and achieve state-of-the-art performance.
文摘Conversational recommender systems(CRSs)focus on refining preferences and providing personalized recommendations through natural language interactions and dialogue history.Large language models(LLMs)have shown outstanding performance across various domains,thereby prompting researchers to investigate their applicability in recommendation systems.However,due to the lack of task-specific knowledge and an inefficient feature extraction process,LLMs still have suboptimal performance in recommendation tasks.Therefore,external knowledge sources,such as knowledge graphs(KGs)and knowledge bases(KBs),are often introduced to address the issue of data sparsity.Compared to KGs,KBs possess higher retrieval efficiency,making them more suitable for scenarios where LLMs serve as recommenders.To this end,we introduce a novel framework integrating LLMs with KBs for enhanced retrieval generation,namely LLMKB.LLMKB initially leverages structured knowledge to create mapping dictionaries,extracting entity-relation information from heterogeneous knowledge to construct KBs.Then,LLMKB achieves the embedding calibration between user information representations and documents in KBs through retrieval model fine-tuning.Finally,LLMKB employs retrievalaugmented generation to produce recommendations based on fused text inputs,followed by post-processing.Experiment results on two public CRS datasets demonstrate the effectiveness of our framework.Our code is publicly available at the link:https://anonymous.4open.science/r/LLMKB-6FD0.
基金supported by the National Natural Science Foundation of China(Grant No.81973695)Discipline with Strong Characteristics of Liaocheng University-Intelligent Science and Technology(Grant No.319462208).
文摘Event extraction stands as a significant endeavor within the realm of information extraction,aspiring to automatically extract structured event information from vast volumes of unstructured text.Extracting event elements from multi-modal data remains a challenging task due to the presence of a large number of images and overlapping event elements in the data.Although researchers have proposed various methods to accomplish this task,most existing event extraction models cannot address these challenges because they are only applicable to text scenarios.To solve the above issues,this paper proposes a multi-modal event extraction method based on knowledge fusion.Specifically,for event-type recognition,we use a meticulous pipeline approach that integrates multiple pre-trained models.This approach enables a more comprehensive capture of the multidimensional event semantic features present in military texts,thereby enhancing the interconnectedness of information between trigger words and events.For event element extraction,we propose a method for constructing a priori templates that combine event types with corresponding trigger words.This approach facilitates the acquisition of fine-grained input samples containing event trigger words,thus enabling the model to understand the semantic relationships between elements in greater depth.Furthermore,a fusion method for spatial mapping of textual event elements and image elements is proposed to reduce the category number overload and effectively achieve multi-modal knowledge fusion.The experimental results based on the CCKS 2022 dataset show that our method has achieved competitive results,with a comprehensive evaluation value F1-score of 53.4%for the model.These results validate the effectiveness of our method in extracting event elements from multi-modal data.
基金This work is supported by the Fundamental Research Funds for the Central Universities(Grant No.HIT.NSRIF.201714)Weihai Science and Technology Development Program(2016DX GJMS15)+1 种基金Weihai Scientific Research and Innovation Fund(2020)Key Research and Development Program in Shandong Provincial(2017GGX90103).
文摘The knowledge graph with relational abundant information has been widely used as the basic data support for the retrieval platforms.Image and text descriptions added to the knowledge graph enrich the node information,which accounts for the advantage of the multi-modal knowledge graph.In the field of cross-modal retrieval platforms,multi-modal knowledge graphs can help to improve retrieval accuracy and efficiency because of the abundant relational infor-mation provided by knowledge graphs.The representation learning method is sig-nificant to the application of multi-modal knowledge graphs.This paper proposes a distributed collaborative vector retrieval platform(DCRL-KG)using the multi-modal knowledge graph VisualSem as the foundation to achieve efficient and high-precision multimodal data retrieval.Firstly,use distributed technology to classify and store the data in the knowledge graph to improve retrieval efficiency.Secondly,this paper uses BabelNet to expand the knowledge graph through multi-ple filtering processes and increase the diversification of information.Finally,this paper builds a variety of retrieval models to achieve the fusion of retrieval results through linear combination methods to achieve high-precision language retrieval and image retrieval.The paper uses sentence retrieval and image retrieval experi-ments to prove that the platform can optimize the storage structure of the multi-modal knowledge graph and have good performance in multi-modal space.
文摘In the context of power generation companies, vast amounts of specialized data and expert knowledge have been accumulated. However, challenges such as data silos and fragmented knowledge hinder the effective utilization of this information. This study proposes a novel framework for intelligent Question-and-Answer (Q&A) systems based on Retrieval-Augmented Generation (RAG) to address these issues. The system efficiently acquires domain-specific knowledge by leveraging external databases, including Relational Databases (RDBs) and graph databases, without additional fine-tuning for Large Language Models (LLMs). Crucially, the framework integrates a Dynamic Knowledge Base Updating Mechanism (DKBUM) and a Weighted Context-Aware Similarity (WCAS) method to enhance retrieval accuracy and mitigate inherent limitations of LLMs, such as hallucinations and lack of specialization. Additionally, the proposed DKBUM dynamically adjusts knowledge weights within the database, ensuring that the most recent and relevant information is utilized, while WCAS refines the alignment between queries and knowledge items by enhanced context understanding. Experimental validation demonstrates that the system can generate timely, accurate, and context-sensitive responses, making it a robust solution for managing complex business logic in specialized industries.
基金partially supported by the National Natural Science Foundation of China under Grants 62471493 and 62402257(for conceptualization and investigation)partially supported by the Natural Science Foundation of Shandong Province,China under Grants ZR2023LZH017,ZR2024MF066,and 2023QF025(for formal analysis and validation)+1 种基金partially supported by the Open Foundation of Key Laboratory of Computing Power Network and Information Security,Ministry of Education,Qilu University of Technology(Shandong Academy of Sciences)under Grant 2023ZD010(for methodology and model design)partially supported by the Russian Science Foundation(RSF)Project under Grant 22-71-10095-P(for validation and results verification).
文摘To address the challenge of missing modal information in entity alignment and to mitigate information loss or bias arising frommodal heterogeneity during fusion,while also capturing shared information acrossmodalities,this paper proposes a Multi-modal Pre-synergistic Entity Alignmentmodel based on Cross-modalMutual Information Strategy Optimization(MPSEA).The model first employs independent encoders to process multi-modal features,including text,images,and numerical values.Next,a multi-modal pre-synergistic fusion mechanism integrates graph structural and visual modal features into the textual modality as preparatory information.This pre-fusion strategy enables unified perception of heterogeneous modalities at the model’s initial stage,reducing discrepancies during the fusion process.Finally,using cross-modal deep perception reinforcement learning,the model achieves adaptive multilevel feature fusion between modalities,supporting learningmore effective alignment strategies.Extensive experiments on multiple public datasets show that the MPSEA method achieves gains of up to 7% in Hits@1 and 8.2% in MRR on the FBDB15K dataset,and up to 9.1% in Hits@1 and 7.7% in MRR on the FBYG15K dataset,compared to existing state-of-the-art methods.These results confirm the effectiveness of the proposed model.
基金supported by the National Key R&D Program of China under Grant No.2018YFA0701604NSFC under Grant No.62341102.
文摘The network security knowledge base standardizes and integrates network security data,providing a reliable foundation for real-time network security protection solutions.However,current research on network security knowledge bases mainly focuses on their construction,while the potential to optimize intelligent security services for real-time network security protection requires further exploration.Therefore,how to effectively utilize the vast amount of historical knowledge in the field of network security and establish a feedback mechanism to update it in real time,thereby enhancing the detection capability of security services against malicious traffic,has become an important issue.Our contribution is fourfold.First,we design a feedback interface to update the knowledge base with information such as features of attack traffic,detection outcomes from network service functions(NSF),and system resource utilization.Second,we introduce a feature selection method that combines PageRank and RandomForest to identify influential features in the knowledge base and dynamically incorporate them into the NSFs.Third,we propose a path selection method that combines graph attention network(GAT)and deep reinforcement learning(DRL)to learn the local knowledge of the knowledge base and determine the optimal traffic path within the Service Function Chains(SFC).Finally,experimental results demonstrate that the knowledge base can be updated in real time according to feedback information,and the optimized service achieves an accuracy,recall,and F1 score exceeding 96%.Compared to preset paths and paths selected using the deep Q-network(DQN)method,our proposed method increases the malicious traffic detection rate by an average of 12.4%and 4.6%,respectively,enhances the total malicious traffic detection capability(TMTDC)of the path by 18.1%and 11.5%,and significantly reduces path detection delay.It has been verified that the proposed intelligent security optimization method can monitor malicious traffic in real time,update knowledge,and enhance the system’s detection capability against malicious traffic.
基金supported in part by the National Natural Science Foundation of China under Grant No.62062031in part by the MIC/SCOPE#JP235006102+2 种基金in part by JST ASPIRE Grant Number JPMJAP2325in part by ROIS NII Open Collaborative Research under Grant 24S0601in part by collaborative research with Toyota Motor Corporation,Japan。
文摘Remote driving,an emergent technology enabling remote operations of vehicles,presents a significant challenge in transmitting large volumes of image data to a central server.This requirement outpaces the capacity of traditional communication methods.To tackle this,we propose a novel framework using semantic communications,through a region of interest semantic segmentation method,to reduce the communication costs by transmitting meaningful semantic information rather than bit-wise data.To solve the knowledge base inconsistencies inherent in semantic communications,we introduce a blockchain-based edge-assisted system for managing diverse and geographically varied semantic segmentation knowledge bases.This system not only ensures the security of data through the tamper-resistant nature of blockchain but also leverages edge computing for efficient management.Additionally,the implementation of blockchain sharding handles differentiated knowledge bases for various tasks,thus boosting overall blockchain efficiency.Experimental results show a great reduction in latency by sharding and an increase in model accuracy,confirming our framework's effectiveness.
基金Provincial-Level Quality Engineering Project,Preschool Education Teacher Training Base of Fuyang Normal University(Project No.:2023cyts023)University-Level Research Team Project,Collaborative Innovation Center for Basic Education in Northern Anhui(Project No.:kytd202418)。
文摘The“Opinions on Comprehensively Deepening Curriculum Reform to Fulfill the Fundamental Task of Strengthening Moral Education”,issued by China’s Ministry of Education in 2015,explicitly identified Project-Based Learning(PBL)as a key strategy for cultivating students’core competencies.Since then,PBL has been widely implemented as a pilot initiative in primary and secondary schools,gaining increasing influence.Analyzing the intellectual foundations of PBL research in China can offer valuable insights into its theoretical and practical dimensions.This study uses CiteSpace to examine 156 PBL-related articles from the CSSCI database,revealing that the knowledge base of PBL research is primarily built on two major domains.The first is the theoretical foundation,characterized by frequently cited literature focusing on the conceptual framework,educational value,interdisciplinary approaches,core competency cultivation,and instructional objectives of PBL.The second is empirical research,where highly cited studies include case analyses across K–12 settings,general high schools,and higher education institutions.Moving forward,future research on PBL should explore its meaning and value from a dual-subject and integrated perspective,expand case studies to include vocational education,and further promote the interdisciplinary development of core competencies through PBL.
基金supported by the Autonomous Region Industry-Education Integration Project“Application of DNA Methylation Combined with Spiral CT in the Screening of Pulmonary Ground-Glass Nodules and AI Recognition Systems in Teaching Practice”(Project No.2023210016)the“Open Project of the State Key Laboratory of High Incidence Diseases in Central Asia”(Project No.SKL-HIDCA-2021-28).
文摘Objective:To explore the effectiveness of multi-modal teaching based on an online case library in the education of gene methylation combined with spiral computed tomography(CT)screening for pulmonary ground-glass opacity(GGO)nodules.Methods:From October 2023 to April 2024,66 medical imaging students were selected and randomly divided into a control group and an observation group,each with 33 students.The control group received traditional lecture-based teaching,while the observation group was taught using a multi-modal teaching approach based on an online case library.Performance on assessments and teaching quality were analyzed between the two groups.Results:The observation group achieved higher scores in theoretical and practical knowledge compared to the control group(P<0.05).Additionally,the teaching quality scores were significantly higher in the observation group(P<0.05).Conclusion:Implementing multi-modal teaching based on an online case library for pulmonary GGO nodule screening with gene methylation combined with spiral CT can enhance students’knowledge acquisition,improve teaching quality,and have significant clinical application value.
文摘Objective: To develop a best-evidence-based optimal nutrition management plan for patients with chronic heart failure, apply it in clinical practice, and evaluate its effectiveness. Methods: Use the KTA knowledge translation model to guide evidence-based practice in nutrition management, and compare the nutritional status, cardiac function status, quality of life, and quality review indicators of chronic heart failure patients before and after the application of evidence. Results: After the application of evidence, the nutritional status indicators (MNA-SF score, albumin, hemoglobin) of two groups of heart failure patients significantly increased compared to before the application of evidence, with statistically significant differences (p Conclusion: The KTA knowledge translation model provides methodological guidance for the implementation of evidence-based practice for heart failure patients. This evidence-based practice project is beneficial for improving the outcomes of malnutrition in chronic heart failure patients and is conducive to standardizing nursing pathways, thereby promoting the improvement of nursing quality.
基金National College Students’Training Programs of Innovation and Entrepreneurship,Grant/Award Number:S202210022060the CACMS Innovation Fund,Grant/Award Number:CI2021A00512the National Nature Science Foundation of China under Grant,Grant/Award Number:62206021。
文摘Media convergence works by processing information from different modalities and applying them to different domains.It is difficult for the conventional knowledge graph to utilise multi-media features because the introduction of a large amount of information from other modalities reduces the effectiveness of representation learning and makes knowledge graph inference less effective.To address the issue,an inference method based on Media Convergence and Rule-guided Joint Inference model(MCRJI)has been pro-posed.The authors not only converge multi-media features of entities but also introduce logic rules to improve the accuracy and interpretability of link prediction.First,a multi-headed self-attention approach is used to obtain the attention of different media features of entities during semantic synthesis.Second,logic rules of different lengths are mined from knowledge graph to learn new entity representations.Finally,knowledge graph inference is performed based on representing entities that converge multi-media features.Numerous experimental results show that MCRJI outperforms other advanced baselines in using multi-media features and knowledge graph inference,demonstrating that MCRJI provides an excellent approach for knowledge graph inference with converged multi-media features.
文摘In this paper, the knowledge based enterprise is considered as an organism, which possesses a set of capabilities. The organizational structure model of knowledge based enterprise organism is described in order to possess the essential capacity set. A dynamic capacity set is defined and analyzed based on the definition of the growth and development for knowledge based enterprise organism. The structure of the capacity base, a subset of the capacity set, is optimized for different periods of the organism ...
文摘The knowledge representation mode and inference control strategy were analyzed according to the specialties of air-conditioning cooling/heating sources selection. The constructing idea and working procedure for knowledge base and inference engine were proposed while the realization technique of the C language was discussed. An intelligent decision support system (IDSS) model based on such knowledge representation and inference mechanism was developed by domain engineers. The model was verified to have a small kernel and powerful capability in list processing and data driving, which was successfully used in the design of a cooling/heating sources system for a large-sized office building.
文摘Aim To analyse the influence of knowledge base on the performance of the fuzzy controller of the electrohydraulic position control system,and to determine their selection cri- teria. Methods Experiments based on different membership functions,scaling factors and con-trol rules were done separately.The experiment results and the influence of different know- ledge base on the control performance were analysed in theory so that criteria of selcting knowledge base can be summarized correctly.Results Knowledge base,including membershipfunctions, scaling factors and control rules,has a crucial effect on the fuzzy control system.Suitably selected knowledge base can lead to good control performance of fuzzy control sys-tem. Conclusion Being symmetric,having an intersection ratio of 1 and satisfying width con- dition are three necessities for selecting membership functions.Selecting scaling factors dependson both the system requirement and a comprehensive analysis in the overshoot,oscillation, rising time and stability. Integrity and continuity must be guaranteed when determining control rules.
基金Program for Changjiang Scholars and Innovative Research Team in University (NoIRT0652)the National High Technology Research and Development Program of China (863 Program) ( No2006AA01A123)
文摘To semantically integrate heterogeneous resources and provide a unified intelligent access interface, semantic web technology is exploited to publish and interlink machineunderstandable resources so that intelligent search can be supported. TCMSearch, a deployed intelligent search engine for traditional Chinese medicine （TCM）, is presented. The core of the system is an integrated knowledge base that uses a TCM domain ontology to represent the instances and relationships in TCM. Machine-learning techniques are used to generate semantic annotations for texts and semantic mappings for relational databases, and then a semantic index is constructed for these resources. The major benefit of representing the semantic index in RDF/OWL is to support some powerful reasoning functions, such as class hierarchies and relation inferences. By combining resource integration with reasoning, the knowledge base can support some intelligent search paradigms besides keyword search, such as correlated search, semantic graph navigation and concept recommendation.
基金supported by National Natural Science Foundation of China (No. 61773239)Shenzhen Future Industry Special Fund (No. JCYJ20160331174814755)
文摘In order to improve the learning ability of robots, we present a reinforcement learning approach with a knowledge base for mapping natural language instructions to executable action sequences. A simulated platform with physical engine is built as interactive environment. Based on the knowledge base, a reward function with immediate rewards and delayed rewards is designed to handle sparse reward problems. Also, a list of object states is produced by retrieving the knowledge base, as a standard to define the quality of action sequences. Experimental results demonstrate that our approach yields good performance on accuracy of action sequences production.
文摘QNET-CFD is a thematic network on quality and trust for the industrial applications of Computational Fluid Dynamics （CFD）, developed under the European Union R＆D program. The main objectives of QNET-CFD were to collect CFD and experimental data in a systematic and quality controlled way and to set the basis for a consistent Knowledge Base in support of CFD guidance and validation. The QNET-CFD activity was organized around six Thematic Areas （TAs） covering the following industry sectors： external aerodynamics; combustion ＆ heat transfer; chemical process, thermal hydraulics and nuclear safety; civil construction ＆ HVAC; environment; turbomachinery internal flows. The main outcome of the QNET-CFD actions is the Knowledge Base （KB） with contains in a user oriented interface, extensive experimental and CFD data for a large number of test cases subdivided into 53 Application Challenges （AC） and 43 Underlying Flow Regimes （UFR）. The KB contains, in addition to state-of-the-art reviews for each of the six thematic areas, Best Practice Advice （BPA） in the use of CFD for most of AC. This is considered as a significant contribution form the QNET-CFD activities and it is expected that the level of the thrust and quality in CFD will hereby be improved.
基金Supported by the China National Science and Technology Major Project(2016ZX05014-002,2017ZX05005)Chinese Academy of Sciences Pilot A Special Project(XDA14010205)。
文摘To improve the efficiency and accuracy of carbonate reservoir research,a unified reservoir knowledge base linking geological knowledge management with reservoir research is proposed.The reservoir knowledge base serves high-quality analysis,evaluation,description and geological modeling of reservoirs.The knowledge framework is divided into three categories:technical service standard,technical research method and professional knowledge and cases related to geological objects.In order to build a knowledge base,first of all,it is necessary to form a knowledge classification system and knowledge description standards;secondly,to sort out theoretical understandings and various technical methods for different geologic objects and work out a technical service standard package according to the technical standard;thirdly,to collect typical outcrop and reservoir cases,constantly expand the content of the knowledge base through systematic extraction,sorting and saving,and construct professional knowledge about geological objects.Through the use of encyclopedia based collaborative editing architecture,knowledge construction and sharing can be realized.Geological objects and related attribute parameters can be automatically extracted by using natural language processing(NLP)technology,and outcrop data can be collected by using modern fine measurement technology,to enhance the efficiency of knowledge acquisition,extraction and sorting.In this paper,the geological modeling of fracture-cavity reservoir in the Tarim Basin is taken as an example to illustrate the construction of knowledge base of carbonate reservoir and its application in geological modeling of fracture-cavity carbonate reservoir.
