Nowadays,the internal structure of spacecraft has been increasingly complex.As its“lifeline”,cables require extensive manpower and resources for manual testing,and it is challenging to quickly and accurately locate ...Nowadays,the internal structure of spacecraft has been increasingly complex.As its“lifeline”,cables require extensive manpower and resources for manual testing,and it is challenging to quickly and accurately locate quality problems and find solutions.To address this problem,a knowledge graph based method is employed to extract multi-source heterogeneous cable knowledge entities.The method utilizes the bidirectional encoder representations from transformers(BERT)network to embed word vectors into the input text,then extracts the contextual features of the input sequence through the bidirectional long short-term memory(BiLSTM)network,and finally inputs them into the conditional random field(CRF)network to predict entity categories.Simultaneously,by using the entities extracted by this model as the data layer,a knowledge graph based method has been constructed.Compared to other traditional extraction methods,the entity extraction method used in this study demonstrates significant improvements in metrics such as precision,recall and an F1 score.Ultimately,employing cable test data from a particular aerospace precision machining company,the study has constructed the knowledge graph based method in the field to achieve visualized queries and the traceability and localization of quality problems.展开更多
Chinese organization name recognition is hard and important in natural language processing. To reduce tagged corpus and use untagged corpus,we presented combing Co-training with support vector machines (SVM) and condi...Chinese organization name recognition is hard and important in natural language processing. To reduce tagged corpus and use untagged corpus,we presented combing Co-training with support vector machines (SVM) and conditional random fields (CRF) to improve recognition results. Based on principles of uncorrelated and compatible,we constructed different classifiers from different views within SVM or CRF alone and combination of these two models. And we modified a heuristic untagged samples selection algorithm to reduce time complexity. Experimental results show that under the same tagged data,Co-training has 10% F-measure higher than using SVM or CRF alone; under the same F-measure,Co-training saves at most 70% of tagged data to achieve the same performance.展开更多
In this work,an approach is proposed to acquire synonymous attribute phrases of named entities(NEs) from an online encyclopedia.Synonymous attribute phrases are the phrases that express the same attribute with differe...In this work,an approach is proposed to acquire synonymous attribute phrases of named entities(NEs) from an online encyclopedia.Synonymous attribute phrases are the phrases that express the same attribute with different surface forms for a class of NEs.Specifically,the proposed approach is composed of three stages.Firstly,the entries related to a given NE class are automatically selected from an online encyclopedia.Secondly,attribute phrases are extracted based on the statistics of phrase frequency.Thirdly,synonymous attributes are identified in a pairwise manner through a classification framework combining multiple features.The proposed approach is applied on Baidu Baike,a Chinese online encyclopedia,for four different NE classes.Experimental results show that the approach obtains an average precision of 74%and an average F-value of 65%for the four NE classes.In particular,thousands of synonymous attribute phrase pairs are acquired for each class,which demonstrates the effectiveness of the proposed approach.展开更多
The issue of proper names recognition in Chinese text was discussed. An automatic approach based on association analysis to extract rules from corpus was presented. The method tries to discover rules relevant to exter...The issue of proper names recognition in Chinese text was discussed. An automatic approach based on association analysis to extract rules from corpus was presented. The method tries to discover rules relevant to external evidence by association analysis, without additional manual effort. These rules can be used to recognize the proper nouns in Chinese texts. The experimental result shows that our method is practical in some applications. Moreover, the method is language independent.展开更多
In China, Warring States Period was imbued with chaos of wars and the whole society was in a state of ferment. The disordered phase in ideological area, as the most direct reflection of the change, was characterized b...In China, Warring States Period was imbued with chaos of wars and the whole society was in a state of ferment. The disordered phase in ideological area, as the most direct reflection of the change, was characterized by that all the old ethical standard had been badly weaken, even failed to explain the moral facts. In terms of common people, they were confused and puzzled by the questions as "what is language?" and "what is the standard of speaking and behavior?" Some ancient scholars pinned these problems upon the reversal of "ming" (名) and "shi" (实). As a result, to rectify names and to clarify the name of ethics became a must for reconstructing the old proprieties and system. Among those scholars and thinkers, Xun Zi firstly proposed Zhengming to state the relationship between a name and the thing it represents in a systematical way. This essay only sheds light upon the reason of rectifying names: to distinguish the rank and differentiate the similarities and differences("明贵贱,辨同异")which is the must of "rectifying names".展开更多
In 2008 remuneration reform for public sector was introduced in Russia. Its main idea was to implement P4P principle well-known in business, including more flexible approach to wage setting. This paper presents an att...In 2008 remuneration reform for public sector was introduced in Russia. Its main idea was to implement P4P principle well-known in business, including more flexible approach to wage setting. This paper presents an attempt to estimate an influence of the new remuneration system （NRS） on earnings level, inequalities and job motivation. The estimates are based on microdata of monitoring survey of healthcare economic problems conducted in Russia in 2009 and 2010. The extended specification of Mincer earning equation and probit-models were used. One could observe increasing wage rates and earning inequalities within healthcare institutions adopted NRS though worker＇s experience and regional economic differences remain significant wage determining factors. As it occurs, NRS is widely adopted by large regional and central hospitals while smaller health care institutions show less enthusiasm in implementing reform. Obviously, the larger institutions have more money and better educated administrative staff to introduce the new system. Those chief physicians who adopted NRS point out a positive correlation between earnings and individual input. At the same time, those committed to old principles of wage setting more often note declining job turnover. This latter result could possibly indicate negative personnel sorting, less productive workers tending to stay with employer who doesn＇t assess their performance. As concerns anticipated NRS results such as increasing motivation and quality of health services, the evidence is still ambiguous.展开更多
Named entity disambiguation （NED） is the task of linking mentions of ambiguous entities to their referenced entities in a knowledge base such as Wikipedia. We propose an approach to effectively disentangle the discr...Named entity disambiguation （NED） is the task of linking mentions of ambiguous entities to their referenced entities in a knowledge base such as Wikipedia. We propose an approach to effectively disentangle the discriminative features in the manner of collaborative utilization of collective wisdom （via human-labeled crowd labels） and deep learning （via human-generated data） for the NED task. In particular, we devise a crowd model to elicit the underlying features （crowd features） from crowd labels that indicate a matching candidate for each mention, and then use the crowd features to fine-tune a dynamic convolutional neural network （DCNN）. The learned DCNN is employed to obtain deep crowd features to enhance traditional hand-crafted features for the NED task. The proposed method substantially benefits from the utilization of crowd knowledge （via crowd labels） into a generic deep learning for the NED task. Experimental analysis demonstrates that the proposed approach is superior to the traditional hand-crafted features when enough crowd labels are gathered.展开更多
A relatively simple plug-and-play control system of quantum key distribution （QKD） based on PCI7300 card is demonstrated, including mechanism design, key generation and key acquisition. The system works very well at...A relatively simple plug-and-play control system of quantum key distribution （QKD） based on PCI7300 card is demonstrated, including mechanism design, key generation and key acquisition. The system works very well at the repetition frequency of 1 MHz, and the key generation rate is 100 k/s. A visibility of better than 95% over 50 km-long fiber at 1.31 gm is obtained, which is stable under ordinary lab conditions for 24 h without any feedback control or adjustment. The presented system is a quite promising candidate to realize the QKD in the future.展开更多
We use decomposition and regression to examine the reasons for the changes in nominal and real rates of return of China＇s foreign exchange reserves between 2002 and 2009. The results show that the US financial market...We use decomposition and regression to examine the reasons for the changes in nominal and real rates of return of China＇s foreign exchange reserves between 2002 and 2009. The results show that the US financial market risk premium is the most important determinant of changes in the nominal rate of return, while the US dollar exchange rate and the bulk commodity price are the two key determinants of changes in the real rate of return. From empirically based research, one may conclude that the loose monetary policy of the US Federal Reserve increases China＇s foreign exchange reserves＇ nominal rate of return but decreases the real rate of return and that the European debt crisis has an uncertain impact on China＇s foreign exchange reserves＇ nominal rate of return but may well raise the real rate of return.展开更多
文摘Nowadays,the internal structure of spacecraft has been increasingly complex.As its“lifeline”,cables require extensive manpower and resources for manual testing,and it is challenging to quickly and accurately locate quality problems and find solutions.To address this problem,a knowledge graph based method is employed to extract multi-source heterogeneous cable knowledge entities.The method utilizes the bidirectional encoder representations from transformers(BERT)network to embed word vectors into the input text,then extracts the contextual features of the input sequence through the bidirectional long short-term memory(BiLSTM)network,and finally inputs them into the conditional random field(CRF)network to predict entity categories.Simultaneously,by using the entities extracted by this model as the data layer,a knowledge graph based method has been constructed.Compared to other traditional extraction methods,the entity extraction method used in this study demonstrates significant improvements in metrics such as precision,recall and an F1 score.Ultimately,employing cable test data from a particular aerospace precision machining company,the study has constructed the knowledge graph based method in the field to achieve visualized queries and the traceability and localization of quality problems.
基金National Natural Science Foundations of China (No.60873179, No.60803078)
文摘Chinese organization name recognition is hard and important in natural language processing. To reduce tagged corpus and use untagged corpus,we presented combing Co-training with support vector machines (SVM) and conditional random fields (CRF) to improve recognition results. Based on principles of uncorrelated and compatible,we constructed different classifiers from different views within SVM or CRF alone and combination of these two models. And we modified a heuristic untagged samples selection algorithm to reduce time complexity. Experimental results show that under the same tagged data,Co-training has 10% F-measure higher than using SVM or CRF alone; under the same F-measure,Co-training saves at most 70% of tagged data to achieve the same performance.
基金Supported by the National High Technology Research and Development Programme of China(No.2008AA01Z144)the National NaturalScience Foundation of China(No.61073126,61073129)
文摘In this work,an approach is proposed to acquire synonymous attribute phrases of named entities(NEs) from an online encyclopedia.Synonymous attribute phrases are the phrases that express the same attribute with different surface forms for a class of NEs.Specifically,the proposed approach is composed of three stages.Firstly,the entries related to a given NE class are automatically selected from an online encyclopedia.Secondly,attribute phrases are extracted based on the statistics of phrase frequency.Thirdly,synonymous attributes are identified in a pairwise manner through a classification framework combining multiple features.The proposed approach is applied on Baidu Baike,a Chinese online encyclopedia,for four different NE classes.Experimental results show that the approach obtains an average precision of 74%and an average F-value of 65%for the four NE classes.In particular,thousands of synonymous attribute phrase pairs are acquired for each class,which demonstrates the effectiveness of the proposed approach.
基金The National Hi-Tech Research and Development Program ( 863 )of China ( No2002AA119050)
文摘The issue of proper names recognition in Chinese text was discussed. An automatic approach based on association analysis to extract rules from corpus was presented. The method tries to discover rules relevant to external evidence by association analysis, without additional manual effort. These rules can be used to recognize the proper nouns in Chinese texts. The experimental result shows that our method is practical in some applications. Moreover, the method is language independent.
文摘In China, Warring States Period was imbued with chaos of wars and the whole society was in a state of ferment. The disordered phase in ideological area, as the most direct reflection of the change, was characterized by that all the old ethical standard had been badly weaken, even failed to explain the moral facts. In terms of common people, they were confused and puzzled by the questions as "what is language?" and "what is the standard of speaking and behavior?" Some ancient scholars pinned these problems upon the reversal of "ming" (名) and "shi" (实). As a result, to rectify names and to clarify the name of ethics became a must for reconstructing the old proprieties and system. Among those scholars and thinkers, Xun Zi firstly proposed Zhengming to state the relationship between a name and the thing it represents in a systematical way. This essay only sheds light upon the reason of rectifying names: to distinguish the rank and differentiate the similarities and differences("明贵贱,辨同异")which is the must of "rectifying names".
文摘In 2008 remuneration reform for public sector was introduced in Russia. Its main idea was to implement P4P principle well-known in business, including more flexible approach to wage setting. This paper presents an attempt to estimate an influence of the new remuneration system （NRS） on earnings level, inequalities and job motivation. The estimates are based on microdata of monitoring survey of healthcare economic problems conducted in Russia in 2009 and 2010. The extended specification of Mincer earning equation and probit-models were used. One could observe increasing wage rates and earning inequalities within healthcare institutions adopted NRS though worker＇s experience and regional economic differences remain significant wage determining factors. As it occurs, NRS is widely adopted by large regional and central hospitals while smaller health care institutions show less enthusiasm in implementing reform. Obviously, the larger institutions have more money and better educated administrative staff to introduce the new system. Those chief physicians who adopted NRS point out a positive correlation between earnings and individual input. At the same time, those committed to old principles of wage setting more often note declining job turnover. This latter result could possibly indicate negative personnel sorting, less productive workers tending to stay with employer who doesn＇t assess their performance. As concerns anticipated NRS results such as increasing motivation and quality of health services, the evidence is still ambiguous.
基金supported by the National Basic Research Program of China(No.2015CB352300)the National Natural Science Foundation of China(Nos.61402401 and U1509206)+3 种基金the Zhejiang Provincial Natural Science Foundation of China(No.LQ14F010004)the China Knowledge Centre for Engineering Sciences and Technologythe Fundamental Research Funds for the Central Universitiesthe Qianjiang Talents Program of Zhejiang Province,China
文摘Named entity disambiguation （NED） is the task of linking mentions of ambiguous entities to their referenced entities in a knowledge base such as Wikipedia. We propose an approach to effectively disentangle the discriminative features in the manner of collaborative utilization of collective wisdom （via human-labeled crowd labels） and deep learning （via human-generated data） for the NED task. In particular, we devise a crowd model to elicit the underlying features （crowd features） from crowd labels that indicate a matching candidate for each mention, and then use the crowd features to fine-tune a dynamic convolutional neural network （DCNN）. The learned DCNN is employed to obtain deep crowd features to enhance traditional hand-crafted features for the NED task. The proposed method substantially benefits from the utilization of crowd knowledge （via crowd labels） into a generic deep learning for the NED task. Experimental analysis demonstrates that the proposed approach is superior to the traditional hand-crafted features when enough crowd labels are gathered.
基金supported by the National Natural Science Foundation of China(Nos.61178010and10805006)the Fundamental Research Funds for the Central Universities(No.bupt2010zx04)the China Postdoctoral Science Foundation(No.20100480508)
文摘A relatively simple plug-and-play control system of quantum key distribution （QKD） based on PCI7300 card is demonstrated, including mechanism design, key generation and key acquisition. The system works very well at the repetition frequency of 1 MHz, and the key generation rate is 100 k/s. A visibility of better than 95% over 50 km-long fiber at 1.31 gm is obtained, which is stable under ordinary lab conditions for 24 h without any feedback control or adjustment. The presented system is a quite promising candidate to realize the QKD in the future.
基金part of the key program of the 2011"Strategic Studies on the Diversification of China’s Foreign Exchange Reserves"of the Chinese Academy of Social SciencesCentral Foreign Exchange Business Center for its support
文摘We use decomposition and regression to examine the reasons for the changes in nominal and real rates of return of China＇s foreign exchange reserves between 2002 and 2009. The results show that the US financial market risk premium is the most important determinant of changes in the nominal rate of return, while the US dollar exchange rate and the bulk commodity price are the two key determinants of changes in the real rate of return. From empirically based research, one may conclude that the loose monetary policy of the US Federal Reserve increases China＇s foreign exchange reserves＇ nominal rate of return but decreases the real rate of return and that the European debt crisis has an uncertain impact on China＇s foreign exchange reserves＇ nominal rate of return but may well raise the real rate of return.
