With the escalating complexity in production scenarios, vast amounts of production information are retained within enterprises in the industrial domain. Probing questions of how to meticulously excavate value from com...With the escalating complexity in production scenarios, vast amounts of production information are retained within enterprises in the industrial domain. Probing questions of how to meticulously excavate value from complex document information and establish coherent information links arise. In this work, we present a framework for knowledge graph construction in the industrial domain, predicated on knowledge-enhanced document-level entity and relation extraction. This approach alleviates the shortage of annotated data in the industrial domain and models the interplay of industrial documents. To augment the accuracy of named entity recognition, domain-specific knowledge is incorporated into the initialization of the word embedding matrix within the bidirectional long short-term memory conditional random field (BiLSTM-CRF) framework. For relation extraction, this paper introduces the knowledge-enhanced graph inference (KEGI) network, a pioneering method designed for long paragraphs in the industrial domain. This method discerns intricate interactions among entities by constructing a document graph and innovatively integrates knowledge representation into both node construction and path inference through TransR. On the application stratum, BiLSTM-CRF and KEGI are utilized to craft a knowledge graph from a knowledge representation model and Chinese fault reports for a steel production line, specifically SPOnto and SPFRDoc. The F1 value for entity and relation extraction has been enhanced by 2% to 6%. The quality of the extracted knowledge graph complies with the requirements of real-world production environment applications. The results demonstrate that KEGI can profoundly delve into production reports, extracting a wealth of knowledge and patterns, thereby providing a comprehensive solution for production management.展开更多
An efficient and accurate prediction of a precise tidal level in estuaries and coastal areas is indispensable for the management and decision-making of human activity in the field wok of marine engineering. The variat...An efficient and accurate prediction of a precise tidal level in estuaries and coastal areas is indispensable for the management and decision-making of human activity in the field wok of marine engineering. The variation of the tidal level is a time-varying process. The time-varying factors including interference from the external environment that cause the change of tides are fairly complicated. Furthermore, tidal variations are affected not only by periodic movement of celestial bodies but also by time-varying interference from the external environment. Consequently, for the efficient and precise tidal level prediction, a neuro-fuzzy hybrid technology based on the combination of harmonic analysis and adaptive network-based fuzzy inference system（ANFIS）model is utilized to construct a precise tidal level prediction system, which takes both advantages of the harmonic analysis method and the ANFIS network. The proposed prediction model is composed of two modules： the astronomical tide module caused by celestial bodies’ movement and the non-astronomical tide module caused by various meteorological and other environmental factors. To generate a fuzzy inference system（FIS） structure,three approaches which include grid partition（GP）, fuzzy c-means（FCM） and sub-clustering（SC） are used in the ANFIS network constructing process. Furthermore, to obtain the optimal ANFIS based prediction model, large numbers of simulation experiments are implemented for each FIS generating approach. In this tidal prediction study, the optimal ANFIS model is used to predict the non-astronomical tide module, while the conventional harmonic analysis model is used to predict the astronomical tide module. The final prediction result is performed by combining the estimation outputs of the harmonious analysis model and the optimal ANFIS model. To demonstrate the applicability and capability of the proposed novel prediction model, measured tidal level samples of Fort Pulaski tidal station are selected as the testing databas展开更多
事件抽取旨在从非结构化文本中检测事件类型并抽取事件要素。现有方法在处理文档级文本时仍存在局限性。这是因为文档级文本可能由多个事件组成,并且构成某一事件的事件要素通常分散在不同句子中。为应对上述挑战,提出了一种文档级事件...事件抽取旨在从非结构化文本中检测事件类型并抽取事件要素。现有方法在处理文档级文本时仍存在局限性。这是因为文档级文本可能由多个事件组成,并且构成某一事件的事件要素通常分散在不同句子中。为应对上述挑战,提出了一种文档级事件抽取反向推理模型(reverse inference model for document-level event extraction,RIDEE)。基于无触发词的设计,将文档级事件抽取转化为候选事件要素抽取和事件触发推理两个子任务,并行式抽取事件要素并检测事件类型。此外,设计了一种用于存储历史事件的事件依赖池,使得模型在处理多事件文本时可以充分利用事件之间的依赖关系。公开数据集上的实验结果表明,与现有事件抽取模型相比,RIDEE在进行文档级事件抽取时具有更优的性能。展开更多
基金supported by the National Science and Technology Innovation 2030 New Generation Artificial Intelligence Major Project(Grant No.2018AAA0101800)the National Natural Science Foundation of China(Grant No.72271188).
文摘With the escalating complexity in production scenarios, vast amounts of production information are retained within enterprises in the industrial domain. Probing questions of how to meticulously excavate value from complex document information and establish coherent information links arise. In this work, we present a framework for knowledge graph construction in the industrial domain, predicated on knowledge-enhanced document-level entity and relation extraction. This approach alleviates the shortage of annotated data in the industrial domain and models the interplay of industrial documents. To augment the accuracy of named entity recognition, domain-specific knowledge is incorporated into the initialization of the word embedding matrix within the bidirectional long short-term memory conditional random field (BiLSTM-CRF) framework. For relation extraction, this paper introduces the knowledge-enhanced graph inference (KEGI) network, a pioneering method designed for long paragraphs in the industrial domain. This method discerns intricate interactions among entities by constructing a document graph and innovatively integrates knowledge representation into both node construction and path inference through TransR. On the application stratum, BiLSTM-CRF and KEGI are utilized to craft a knowledge graph from a knowledge representation model and Chinese fault reports for a steel production line, specifically SPOnto and SPFRDoc. The F1 value for entity and relation extraction has been enhanced by 2% to 6%. The quality of the extracted knowledge graph complies with the requirements of real-world production environment applications. The results demonstrate that KEGI can profoundly delve into production reports, extracting a wealth of knowledge and patterns, thereby providing a comprehensive solution for production management.
基金The National Natural Science Foundation of China under contract No.51379002the Fundamental Research Funds for the Central Universities of China under contract Nos 3132016322 and 3132016314the Applied Basic Research Project Fund of the Chinese Ministry of Transport of China under contract No.2014329225010
文摘An efficient and accurate prediction of a precise tidal level in estuaries and coastal areas is indispensable for the management and decision-making of human activity in the field wok of marine engineering. The variation of the tidal level is a time-varying process. The time-varying factors including interference from the external environment that cause the change of tides are fairly complicated. Furthermore, tidal variations are affected not only by periodic movement of celestial bodies but also by time-varying interference from the external environment. Consequently, for the efficient and precise tidal level prediction, a neuro-fuzzy hybrid technology based on the combination of harmonic analysis and adaptive network-based fuzzy inference system（ANFIS）model is utilized to construct a precise tidal level prediction system, which takes both advantages of the harmonic analysis method and the ANFIS network. The proposed prediction model is composed of two modules： the astronomical tide module caused by celestial bodies’ movement and the non-astronomical tide module caused by various meteorological and other environmental factors. To generate a fuzzy inference system（FIS） structure,three approaches which include grid partition（GP）, fuzzy c-means（FCM） and sub-clustering（SC） are used in the ANFIS network constructing process. Furthermore, to obtain the optimal ANFIS based prediction model, large numbers of simulation experiments are implemented for each FIS generating approach. In this tidal prediction study, the optimal ANFIS model is used to predict the non-astronomical tide module, while the conventional harmonic analysis model is used to predict the astronomical tide module. The final prediction result is performed by combining the estimation outputs of the harmonious analysis model and the optimal ANFIS model. To demonstrate the applicability and capability of the proposed novel prediction model, measured tidal level samples of Fort Pulaski tidal station are selected as the testing databas
文摘事件抽取旨在从非结构化文本中检测事件类型并抽取事件要素。现有方法在处理文档级文本时仍存在局限性。这是因为文档级文本可能由多个事件组成,并且构成某一事件的事件要素通常分散在不同句子中。为应对上述挑战,提出了一种文档级事件抽取反向推理模型(reverse inference model for document-level event extraction,RIDEE)。基于无触发词的设计,将文档级事件抽取转化为候选事件要素抽取和事件触发推理两个子任务,并行式抽取事件要素并检测事件类型。此外,设计了一种用于存储历史事件的事件依赖池,使得模型在处理多事件文本时可以充分利用事件之间的依赖关系。公开数据集上的实验结果表明,与现有事件抽取模型相比,RIDEE在进行文档级事件抽取时具有更优的性能。
