卡尔曼滤波(Kalman filter,KF)和最大化后验概率法(maximum a posteriori,MAP)是结构荷载识别中常见的两类广义贝叶斯滤波算法,KF法计算效率高但数值稳定性较差,MAP法适用性强却需要复杂的矩阵求逆运算,加之这两类方法对荷载形式和测点...卡尔曼滤波(Kalman filter,KF)和最大化后验概率法(maximum a posteriori,MAP)是结构荷载识别中常见的两类广义贝叶斯滤波算法,KF法计算效率高但数值稳定性较差,MAP法适用性强却需要复杂的矩阵求逆运算,加之这两类方法对荷载形式和测点布置的苛刻要求,目前仅适用于简单荷载的识别。为此,该研究提出了针对任意分布式荷载的贝叶斯全局响应重构方法,从在线和离线两个角度改进了现有方法。针对在线KF方法,该研究从结构动力特性中导出等效荷载向量来降低未知荷载的维度,得到满足可控性条件的等效系统模型,并采用输入状态联合估计方法同时识别等效荷载和全局响应。针对离线MAP方法,引入考虑了空间相关性的荷载先验分布,采用MAP策略同时对等效荷载和观测噪声进行迭代估计,随后根据识别得到的等效荷载重构全局响应。改进后的在线和离线方法均不需要提前获取荷载位置或分布形式。通过青州大桥在风荷载和交通荷载下采集的响应数据对所提方法的精度和适用性进行了验证。展开更多
Objectives Medical knowledge extraction (MKE) plays a key role in natural language processing (NLP) research in electronic medical records (EMR),which are the important digital carriers for recording medical activitie...Objectives Medical knowledge extraction (MKE) plays a key role in natural language processing (NLP) research in electronic medical records (EMR),which are the important digital carriers for recording medical activities of patients.Named entity recognition (NER) and medical relation extraction (MRE) are two basic tasks of MKE.This study aims to improve the recognition accuracy of these two tasks by exploring deep learning methods.Methods This study discussed and built two application scenes of bidirectional long short-term memory combined conditional random field (BiLSTM-CRF) model for NER and MRE tasks.In the data preprocessing of both tasks,a GloVe word embedding model was used to vectorize words.In the NER task,a sequence labeling strategy was used to classify each word tag by the joint probability distribution through the CRF layer.In the MRE task,the medical entity relation category was predicted by transforming the classification problem of a single entity into a sequence classification problem and linking the feature combinations between entities also through the CRF layer.Results Through the validation on the I2B2 2010 public dataset,the BiLSTM-CRF models built in this study got much better results than the baseline methods in the two tasks,where the F1-measure was up to 0.88 in NER task and 0.78 in MRE task.Moreover,the model converged faster and avoided problems such as overfitting.Conclusion This study proved the good performance of deep learning on medical knowledge extraction.It also verified the feasibility of the BiLSTM-CRF model in different application scenarios,laying the foundation for the subsequent work in the EMR field.展开更多
Enterprise applications utilize relational databases and structured business processes, requiring slow and expensive conversion of inputs and outputs, from business documents such as invoices, purchase orders, and rec...Enterprise applications utilize relational databases and structured business processes, requiring slow and expensive conversion of inputs and outputs, from business documents such as invoices, purchase orders, and receipts, into known templates and schemas before processing. We propose a new LLM Agent-based intelligent data extraction, transformation, and load (IntelligentETL) pipeline that not only ingests PDFs and detects inputs within it but also addresses the extraction of structured and unstructured data by developing tools that most efficiently and securely deal with respective data types. We study the efficiency of our proposed pipeline and compare it with enterprise solutions that also utilize LLMs. We establish the supremacy in timely and accurate data extraction and transformation capabilities of our approach for analyzing the data from varied sources based on nested and/or interlinked input constraints.展开更多
As a core resource of scientific knowledge,academic documents have been frequently used by scholars,especially newcomers to a given field.In the era of big data,scientific documents such as academic articles,patents,t...As a core resource of scientific knowledge,academic documents have been frequently used by scholars,especially newcomers to a given field.In the era of big data,scientific documents such as academic articles,patents,technical reports,and webpages are booming.The rapid daily growth of scientific documents indicates that a large amount of knowledge is proposed,improved,and used(Zhang et al.,2021).展开更多
A new algorithm for the knowledge discovery based on statistic inductionlogic is proposed, and the validity of the methods is verified by examples. The method is suitablefor a large range of knowledge discovery applic...A new algorithm for the knowledge discovery based on statistic inductionlogic is proposed, and the validity of the methods is verified by examples. The method is suitablefor a large range of knowledge discovery applications in the studying of causal relation,uncertainty knowledge acquisition and principal factors analyzing. The language filed description ofthe state space makes the algorithm robust in the adaptation with easier understandable results,which are isomotopy with natural language in the topologic space.展开更多
With the application of artificial intelligence technology in the power industry,the knowledge graph is expected to play a key role in power grid dispatch processes,intelligent maintenance,and customer service respons...With the application of artificial intelligence technology in the power industry,the knowledge graph is expected to play a key role in power grid dispatch processes,intelligent maintenance,and customer service response provision.Knowledge graphs are usually constructed based on entity recognition.Specifically,based on the mining of entity attributes and relationships,domain knowledge graphs can be constructed through knowledge fusion.In this work,the entities and characteristics of power entity recognition are analyzed,the mechanism of entity recognition is clarified,and entity recognition techniques are analyzed in the context of the power domain.Power entity recognition based on the conditional random fields (CRF) and bidirectional long short-term memory (BLSTM) models is investigated,and the two methods are comparatively analyzed.The results indicated that the CRF model,with an accuracy of 83%,can better identify the power entities compared to the BLSTM.The CRF approach can thus be applied to the entity extraction for knowledge graph construction in the power field.展开更多
针对中文文本检错纠错研究任务,提出了基于知识增强的自然语言表示模型(enhanced representation through knowledge integration, ERNIE)与序列标注结合的中文文本检错纠错模型。该模型由检错和纠错两部分组成,检错阶段ERNIE使用全局...针对中文文本检错纠错研究任务,提出了基于知识增强的自然语言表示模型(enhanced representation through knowledge integration, ERNIE)与序列标注结合的中文文本检错纠错模型。该模型由检错和纠错两部分组成,检错阶段ERNIE使用全局注意力机制进行词向量编码输入到BiLSTM-CRF序列标注模型中,双向长短期记忆网络(bi-directional long short-term memory, BiLSTM)提取上下文的信息进行拼接生成双向的词向量,再通过条件随机场(conditional random field, CRF)计算联合概率增加对邻近词标签的依赖性优化整个序列,从而解决标注偏置等问题给出的错误标注。纠错阶段根据检错模型输出的结果采用不同策略分类纠错,将标注为错字、缺字的错误使用ERNIE掩码语言模型和混淆集匹配进行预测,对多字、乱序错误直接纠正。实验结果表明,引入序列标注根据错误类型进行分类纠错有效提升了纠错率,在SIGHAN数据集上测试F1达到了81.8%。展开更多
Purpose: This paper explores a method of knowledge discovery by visualizing and analyzing co-occurrence relations among three or more entities in collections of journal articles.Design/methodology/approach: A variety ...Purpose: This paper explores a method of knowledge discovery by visualizing and analyzing co-occurrence relations among three or more entities in collections of journal articles.Design/methodology/approach: A variety of methods such as the model construction,system analysis and experiments are used. The author has improved Morris' crossmapping technique and developed a technique for directly describing,visualizing and analyzing co-occurrence relations among three or more entities in collections of journal articles.Findings: The visualization tools and the knowledge discovery method can efficiently reveal the multiple co-occurrence relations among three entities in collections of journal papers. It can reveal more and in-depth information than analyzing co-occurrence relations between two entities. Therefore,this method can be used for mapping knowledge domain that is manifested in association with the entities from multi-dimensional perspectives and in an all-round way.Research limitations: The technique could only be used to analyze co-occurrence relations of less than three entities at present.Practical implications: This research has expanded the study scope of co-occurrence analysis.The research result has provided a theoretical support for co-occurrence analysis.Originality/value: There has not been a systematic study on co-occurrence relations among multiple entities in collections of journal articles. This research defines multiple co-occurrence and the research scope,develops the visualization analysis tool and designs the analysis model of the knowledge discovery method.展开更多
A study on knowledge transfer in a mutli-agent organization is performed by applying the basic principle in physics such as the kinetic theory.Based on the theoretical analysis of the knowledge accumulation process an...A study on knowledge transfer in a mutli-agent organization is performed by applying the basic principle in physics such as the kinetic theory.Based on the theoretical analysis of the knowledge accumulation process and knowledge transfer attributes,a special type of knowledge field(KF)is introduced and the knowledge diffusion equation(KDE)is developed.The evolution of knowledge potential is modeled by lattice kinetic equation and verified by numerical experiments.The new equation-based modeling developed in this paper is meaningful to simulate and predict the knowledge transfer process in firms.The development of the lattice kinetic model(LKM)for knowledge transfer can contribute to the knowledge management theory,and the managers can also simulate the knowledge accumulation process by using the LKM.展开更多
To increase the resilience of farmers’livelihood systems,detailed knowledge of adaptation strategies for dealing with the impacts of climate change is required.Knowledge co-production approach is an adaptation strate...To increase the resilience of farmers’livelihood systems,detailed knowledge of adaptation strategies for dealing with the impacts of climate change is required.Knowledge co-production approach is an adaptation strategy that is considered appropriate in the context of the increasing frequency of disasters caused by climate change.Previous research of knowledge co-production on climate change adaptation in Indonesia is insufficient,particularly at local level,so we examined the flow of climate change adaptation knowledge in the knowledge co-production process through climate field school(CFS)activities in this study.We interviewed 120 people living in Bulukumba Regency,South Sulawesi Province,Indonesia,involving 12 crowds including male and female farmers participated in CFS and not participated in CFS,local government officials,agriculture extension workers,agricultural traders,farmers’family members and neighbors,etc.In brief,the 12 groups of people mainly include two categories of people,i.e.,people involved in CFS activities and outside CFS.We applied descriptive method and Social network analysis(SNA)to determine how knowledge flow in the community network and which groups of actors are important for knowledge flow.The findings of this study reveal that participants in CFS activities convey the knowledge they acquired formally(i.e.,from TV,radio,government,etc.)and informally(i.e.,from market,friends,relatives,etc.)to other actors,especially to their families and neighbors.The results also show that the acquisition and sharing of knowledge facilitate the flow of climate change adaptation knowledge based on knowledge co-operation.In addition,the findings highlight the key role of actors in the knowledge transfer process,and key actors involved in disseminating information about climate change adaptation.To be specific,among all the actors,family member and neighbor of CFS actor are the most common actors in disseminating climate knowledge information and closest to other actors in the network;agricultural trader and family member of CFS actor collaborate most with other actors in the community network;and farmers participated in CFS,including those heads of farmer groups,agricultural extension workers,and local government officials are more willing to contact with other actors in the network.To facilitate the flow of knowledge on climate change adaptation,CFS activities should be conducted regularly and CFS models that fit the situation of farmers’vulnerability to climate change should be developed.展开更多
文摘卡尔曼滤波(Kalman filter,KF)和最大化后验概率法(maximum a posteriori,MAP)是结构荷载识别中常见的两类广义贝叶斯滤波算法,KF法计算效率高但数值稳定性较差,MAP法适用性强却需要复杂的矩阵求逆运算,加之这两类方法对荷载形式和测点布置的苛刻要求,目前仅适用于简单荷载的识别。为此,该研究提出了针对任意分布式荷载的贝叶斯全局响应重构方法,从在线和离线两个角度改进了现有方法。针对在线KF方法,该研究从结构动力特性中导出等效荷载向量来降低未知荷载的维度,得到满足可控性条件的等效系统模型,并采用输入状态联合估计方法同时识别等效荷载和全局响应。针对离线MAP方法,引入考虑了空间相关性的荷载先验分布,采用MAP策略同时对等效荷载和观测噪声进行迭代估计,随后根据识别得到的等效荷载重构全局响应。改进后的在线和离线方法均不需要提前获取荷载位置或分布形式。通过青州大桥在风荷载和交通荷载下采集的响应数据对所提方法的精度和适用性进行了验证。
基金Supported by the Zhejiang Provincial Natural Science Foundation(No.LQ16H180004)~~
文摘Objectives Medical knowledge extraction (MKE) plays a key role in natural language processing (NLP) research in electronic medical records (EMR),which are the important digital carriers for recording medical activities of patients.Named entity recognition (NER) and medical relation extraction (MRE) are two basic tasks of MKE.This study aims to improve the recognition accuracy of these two tasks by exploring deep learning methods.Methods This study discussed and built two application scenes of bidirectional long short-term memory combined conditional random field (BiLSTM-CRF) model for NER and MRE tasks.In the data preprocessing of both tasks,a GloVe word embedding model was used to vectorize words.In the NER task,a sequence labeling strategy was used to classify each word tag by the joint probability distribution through the CRF layer.In the MRE task,the medical entity relation category was predicted by transforming the classification problem of a single entity into a sequence classification problem and linking the feature combinations between entities also through the CRF layer.Results Through the validation on the I2B2 2010 public dataset,the BiLSTM-CRF models built in this study got much better results than the baseline methods in the two tasks,where the F1-measure was up to 0.88 in NER task and 0.78 in MRE task.Moreover,the model converged faster and avoided problems such as overfitting.Conclusion This study proved the good performance of deep learning on medical knowledge extraction.It also verified the feasibility of the BiLSTM-CRF model in different application scenarios,laying the foundation for the subsequent work in the EMR field.
文摘Enterprise applications utilize relational databases and structured business processes, requiring slow and expensive conversion of inputs and outputs, from business documents such as invoices, purchase orders, and receipts, into known templates and schemas before processing. We propose a new LLM Agent-based intelligent data extraction, transformation, and load (IntelligentETL) pipeline that not only ingests PDFs and detects inputs within it but also addresses the extraction of structured and unstructured data by developing tools that most efficiently and securely deal with respective data types. We study the efficiency of our proposed pipeline and compare it with enterprise solutions that also utilize LLMs. We establish the supremacy in timely and accurate data extraction and transformation capabilities of our approach for analyzing the data from varied sources based on nested and/or interlinked input constraints.
文摘As a core resource of scientific knowledge,academic documents have been frequently used by scholars,especially newcomers to a given field.In the era of big data,scientific documents such as academic articles,patents,technical reports,and webpages are booming.The rapid daily growth of scientific documents indicates that a large amount of knowledge is proposed,improved,and used(Zhang et al.,2021).
基金[This work was financially supported by the National Natural Science Foundation of China (No. 69835001).]
文摘A new algorithm for the knowledge discovery based on statistic inductionlogic is proposed, and the validity of the methods is verified by examples. The method is suitablefor a large range of knowledge discovery applications in the studying of causal relation,uncertainty knowledge acquisition and principal factors analyzing. The language filed description ofthe state space makes the algorithm robust in the adaptation with easier understandable results,which are isomotopy with natural language in the topologic space.
基金supported by Science and Technology Project of State Grid Corporation(Research and Application of Intelligent Energy Meter Quality Analysis and Evaluation Technology Based on Full Chain Data)
文摘With the application of artificial intelligence technology in the power industry,the knowledge graph is expected to play a key role in power grid dispatch processes,intelligent maintenance,and customer service response provision.Knowledge graphs are usually constructed based on entity recognition.Specifically,based on the mining of entity attributes and relationships,domain knowledge graphs can be constructed through knowledge fusion.In this work,the entities and characteristics of power entity recognition are analyzed,the mechanism of entity recognition is clarified,and entity recognition techniques are analyzed in the context of the power domain.Power entity recognition based on the conditional random fields (CRF) and bidirectional long short-term memory (BLSTM) models is investigated,and the two methods are comparatively analyzed.The results indicated that the CRF model,with an accuracy of 83%,can better identify the power entities compared to the BLSTM.The CRF approach can thus be applied to the entity extraction for knowledge graph construction in the power field.
文摘针对中文文本检错纠错研究任务,提出了基于知识增强的自然语言表示模型(enhanced representation through knowledge integration, ERNIE)与序列标注结合的中文文本检错纠错模型。该模型由检错和纠错两部分组成,检错阶段ERNIE使用全局注意力机制进行词向量编码输入到BiLSTM-CRF序列标注模型中,双向长短期记忆网络(bi-directional long short-term memory, BiLSTM)提取上下文的信息进行拼接生成双向的词向量,再通过条件随机场(conditional random field, CRF)计算联合概率增加对邻近词标签的依赖性优化整个序列,从而解决标注偏置等问题给出的错误标注。纠错阶段根据检错模型输出的结果采用不同策略分类纠错,将标注为错字、缺字的错误使用ERNIE掩码语言模型和混淆集匹配进行预测,对多字、乱序错误直接纠正。实验结果表明,引入序列标注根据错误类型进行分类纠错有效提升了纠错率,在SIGHAN数据集上测试F1达到了81.8%。
文摘Purpose: This paper explores a method of knowledge discovery by visualizing and analyzing co-occurrence relations among three or more entities in collections of journal articles.Design/methodology/approach: A variety of methods such as the model construction,system analysis and experiments are used. The author has improved Morris' crossmapping technique and developed a technique for directly describing,visualizing and analyzing co-occurrence relations among three or more entities in collections of journal articles.Findings: The visualization tools and the knowledge discovery method can efficiently reveal the multiple co-occurrence relations among three entities in collections of journal papers. It can reveal more and in-depth information than analyzing co-occurrence relations between two entities. Therefore,this method can be used for mapping knowledge domain that is manifested in association with the entities from multi-dimensional perspectives and in an all-round way.Research limitations: The technique could only be used to analyze co-occurrence relations of less than three entities at present.Practical implications: This research has expanded the study scope of co-occurrence analysis.The research result has provided a theoretical support for co-occurrence analysis.Originality/value: There has not been a systematic study on co-occurrence relations among multiple entities in collections of journal articles. This research defines multiple co-occurrence and the research scope,develops the visualization analysis tool and designs the analysis model of the knowledge discovery method.
基金supported by the National Natural Science Foundation of China(71472055 71871007)+2 种基金National Social Science Foundation of China(16AZD0006)Heilongjiang Philosophy and Social Science Research Project(19GLB087)the Fundamental Research Funds for the Central Universities(HIT.NSRIF.2019033)
文摘A study on knowledge transfer in a mutli-agent organization is performed by applying the basic principle in physics such as the kinetic theory.Based on the theoretical analysis of the knowledge accumulation process and knowledge transfer attributes,a special type of knowledge field(KF)is introduced and the knowledge diffusion equation(KDE)is developed.The evolution of knowledge potential is modeled by lattice kinetic equation and verified by numerical experiments.The new equation-based modeling developed in this paper is meaningful to simulate and predict the knowledge transfer process in firms.The development of the lattice kinetic model(LKM)for knowledge transfer can contribute to the knowledge management theory,and the managers can also simulate the knowledge accumulation process by using the LKM.
文摘To increase the resilience of farmers’livelihood systems,detailed knowledge of adaptation strategies for dealing with the impacts of climate change is required.Knowledge co-production approach is an adaptation strategy that is considered appropriate in the context of the increasing frequency of disasters caused by climate change.Previous research of knowledge co-production on climate change adaptation in Indonesia is insufficient,particularly at local level,so we examined the flow of climate change adaptation knowledge in the knowledge co-production process through climate field school(CFS)activities in this study.We interviewed 120 people living in Bulukumba Regency,South Sulawesi Province,Indonesia,involving 12 crowds including male and female farmers participated in CFS and not participated in CFS,local government officials,agriculture extension workers,agricultural traders,farmers’family members and neighbors,etc.In brief,the 12 groups of people mainly include two categories of people,i.e.,people involved in CFS activities and outside CFS.We applied descriptive method and Social network analysis(SNA)to determine how knowledge flow in the community network and which groups of actors are important for knowledge flow.The findings of this study reveal that participants in CFS activities convey the knowledge they acquired formally(i.e.,from TV,radio,government,etc.)and informally(i.e.,from market,friends,relatives,etc.)to other actors,especially to their families and neighbors.The results also show that the acquisition and sharing of knowledge facilitate the flow of climate change adaptation knowledge based on knowledge co-operation.In addition,the findings highlight the key role of actors in the knowledge transfer process,and key actors involved in disseminating information about climate change adaptation.To be specific,among all the actors,family member and neighbor of CFS actor are the most common actors in disseminating climate knowledge information and closest to other actors in the network;agricultural trader and family member of CFS actor collaborate most with other actors in the community network;and farmers participated in CFS,including those heads of farmer groups,agricultural extension workers,and local government officials are more willing to contact with other actors in the network.To facilitate the flow of knowledge on climate change adaptation,CFS activities should be conducted regularly and CFS models that fit the situation of farmers’vulnerability to climate change should be developed.
