China implemented the public hospital reform in 2012. This study utilized bootstrapping data envelopment analysis（DEA） to evaluate the technical efficiency（TE） and productivity of county public hospitals in Easter...China implemented the public hospital reform in 2012. This study utilized bootstrapping data envelopment analysis（DEA） to evaluate the technical efficiency（TE） and productivity of county public hospitals in Eastern, Central, and Western China after the 2012 public hospital reform. Data from 127 county public hospitals（39, 45, and 43 in Eastern, Central, and Western China, respectively） were collected during 2012–2015. Changes of TE and productivity over time were estimated by bootstrapping DEA and bootstrapping Malmquist. The disparities in TE and productivity among public hospitals in the three regions of China were compared by Kruskal–Wallis H test and Mann–Whitney U test. The average bias-corrected TE values for the four-year period were 0.6442, 0.5785, 0.6099, and 0.6094 in Eastern, Central, and Western China, and the entire country respectively, with average non-technical efficiency, low pure technical efficiency（PTE）, and high scale efficiency found. Productivity increased by 8.12%, 0.25%, 12.11%, and 11.58% in China and its three regions during 2012–2015, and such increase in productivity resulted from progressive technological changes by 16.42%, 6.32%, 21.08%, and 21.42%, respectively. The TE and PTE of the county hospitals significantly differed among the three regions of China. Eastern and Western China showed significantly higher TE and PTE than Central China. More than 60% of county public hospitals in China and its three areas operated at decreasing return scales. There was a considerable space for TE improvement in county hospitals in China and its three regions. During 2012–2015, the hospitals experienced progressive productivity; however, the PTE changed adversely. Moreover, Central China continuously achieved a significantly lower efficiency score than Eastern and Western China. Decision makers and administrators in China should identify the causes of the observed inefficiencies and take appropriate measures to increase the efficiency of county public hospitals in the three areas of China, especially in Central China.展开更多
A new approach of relation extraction is described in this paper. It adopts a bootstrap- ping model with a novel iteration strategy, which generates more precise examples of specific relation. Compared with previous m...A new approach of relation extraction is described in this paper. It adopts a bootstrap- ping model with a novel iteration strategy, which generates more precise examples of specific relation. Compared with previous methods, the proposed method has three main advantages: first, it needs less manual intervention; second, more abundant and reasonable information are introduced to represent a relation pattern; third, it reduces the risk of circular dependency occurrence in bootstrapping. Scalable evaluation methodology and metrics are developed for our task with comparable techniques over TianWang 100G corpus. The experimental results show that it can get 90% precision and have excellent expansibility.展开更多
Purpose： Our study proposes a bootstrapping-based method to automatically extract data- usage statements from academic texts. Design/methodology/approach： The method for data-usage statements extraction starts with ...Purpose： Our study proposes a bootstrapping-based method to automatically extract data- usage statements from academic texts. Design/methodology/approach： The method for data-usage statements extraction starts with seed entities and iteratively learns patterns and data-usage statements from unlabeled text. In each iteration, new patterns are constructed and added to the pattern list based on their calculated score. Three seed-selection strategies are also proposed in this paper. Findings： The performance of the method is verified by means of experiments on real data collected from computer science journals. The results show that the method can achieve satisfactory performance regarding precision of extraction and extensibility of obtained patterns. Research limitations： While the triple representation of sentences is effective and efficient for extracting data-usage statements, it is unable to handle complex sentences. Additional features that can address complex sentences should thus be explored in the future. Practical implications： Data-usage statements extraction is beneficial for data-repository construction and facilitates research on data-usage tracking, dataset-based scholar search, and dataset evaluation. Originality/value： To the best of our knowledge, this paper is among the first to address the important task of automatically extracting data-usage statements from real data.展开更多
This paper presents a method to learn semantic lexicons using a new bootstrapping method based on graph mutual reinforcement(CMR).The approach uses only unlabeled data and a few seed words to learn new words for each ...This paper presents a method to learn semantic lexicons using a new bootstrapping method based on graph mutual reinforcement(CMR).The approach uses only unlabeled data and a few seed words to learn new words for each semantic category.Different from other bootstrapping methods,we use GMR-based bootstrapping to sort the candidate words and patterns.Experi-mental results show that the GMR.-based bootstrapping approach outperforms the existing algorithms both in in-domain data and out-domain data.Furthermore,it shows that the result depends on not only the size of the corpus but also the quality.展开更多
In order to predict the extreme load of the mechanical components during the entire life,an automatic method based on the bootstrapping technology(BT)is proposed to determine the most suitable threshold.Based on all t...In order to predict the extreme load of the mechanical components during the entire life,an automatic method based on the bootstrapping technology(BT)is proposed to determine the most suitable threshold.Based on all the turning points of the load history and a series of thresholds estimated in advance,the generalized Pareto distribution is established to fit the exceedances.The corresponding distribution parameters are estimated with the maximum likelihood method.Then,BT is employed to calculate the mean squared error(MSE)of each estimated threshold based on the exceedances and the specific distribution parameters.Finally,the threshold with the smallest MSE will be the optimal one.Compared to the kurtosis method and the mean excess function method,the average deviation of the probability density function of exceedances determined by BT reduces by 38.52%and 29.25%,respectively.Moreover,the quantile-quantile plot of the exceedances determined by BT is closer to a straight line.The results suggest the improvement of the modeling flexibility and the determined threshold precision.If the exceedances are insufficient,BT will enlarge their amount by resampling to solve the instability problem of the original distribution parameters.展开更多
基金supported by the National Natural Science Foundation of China(No.71473099)
文摘China implemented the public hospital reform in 2012. This study utilized bootstrapping data envelopment analysis（DEA） to evaluate the technical efficiency（TE） and productivity of county public hospitals in Eastern, Central, and Western China after the 2012 public hospital reform. Data from 127 county public hospitals（39, 45, and 43 in Eastern, Central, and Western China, respectively） were collected during 2012–2015. Changes of TE and productivity over time were estimated by bootstrapping DEA and bootstrapping Malmquist. The disparities in TE and productivity among public hospitals in the three regions of China were compared by Kruskal–Wallis H test and Mann–Whitney U test. The average bias-corrected TE values for the four-year period were 0.6442, 0.5785, 0.6099, and 0.6094 in Eastern, Central, and Western China, and the entire country respectively, with average non-technical efficiency, low pure technical efficiency（PTE）, and high scale efficiency found. Productivity increased by 8.12%, 0.25%, 12.11%, and 11.58% in China and its three regions during 2012–2015, and such increase in productivity resulted from progressive technological changes by 16.42%, 6.32%, 21.08%, and 21.42%, respectively. The TE and PTE of the county hospitals significantly differed among the three regions of China. Eastern and Western China showed significantly higher TE and PTE than Central China. More than 60% of county public hospitals in China and its three areas operated at decreasing return scales. There was a considerable space for TE improvement in county hospitals in China and its three regions. During 2012–2015, the hospitals experienced progressive productivity; however, the PTE changed adversely. Moreover, Central China continuously achieved a significantly lower efficiency score than Eastern and Western China. Decision makers and administrators in China should identify the causes of the observed inefficiencies and take appropriate measures to increase the efficiency of county public hospitals in the three areas of China, especially in Central China.
基金Supported by the National Natural Science Foundation of China (No.60503072, No.60575042 and No.60435020).
文摘A new approach of relation extraction is described in this paper. It adopts a bootstrap- ping model with a novel iteration strategy, which generates more precise examples of specific relation. Compared with previous methods, the proposed method has three main advantages: first, it needs less manual intervention; second, more abundant and reasonable information are introduced to represent a relation pattern; third, it reduces the risk of circular dependency occurrence in bootstrapping. Scalable evaluation methodology and metrics are developed for our task with comparable techniques over TianWang 100G corpus. The experimental results show that it can get 90% precision and have excellent expansibility.
基金supported by the National Natural Science Foundation of China (Grant No.:71473183)
文摘Purpose： Our study proposes a bootstrapping-based method to automatically extract data- usage statements from academic texts. Design/methodology/approach： The method for data-usage statements extraction starts with seed entities and iteratively learns patterns and data-usage statements from unlabeled text. In each iteration, new patterns are constructed and added to the pattern list based on their calculated score. Three seed-selection strategies are also proposed in this paper. Findings： The performance of the method is verified by means of experiments on real data collected from computer science journals. The results show that the method can achieve satisfactory performance regarding precision of extraction and extensibility of obtained patterns. Research limitations： While the triple representation of sentences is effective and efficient for extracting data-usage statements, it is unable to handle complex sentences. Additional features that can address complex sentences should thus be explored in the future. Practical implications： Data-usage statements extraction is beneficial for data-repository construction and facilitates research on data-usage tracking, dataset-based scholar search, and dataset evaluation. Originality/value： To the best of our knowledge, this paper is among the first to address the important task of automatically extracting data-usage statements from real data.
基金Supported by National Natural Science Foundation of China(60673038,60503070)
文摘This paper presents a method to learn semantic lexicons using a new bootstrapping method based on graph mutual reinforcement(CMR).The approach uses only unlabeled data and a few seed words to learn new words for each semantic category.Different from other bootstrapping methods,we use GMR-based bootstrapping to sort the candidate words and patterns.Experi-mental results show that the GMR.-based bootstrapping approach outperforms the existing algorithms both in in-domain data and out-domain data.Furthermore,it shows that the result depends on not only the size of the corpus but also the quality.
基金The National Science and Technology Pillar Program of China(No.2015BAF07B00)
文摘In order to predict the extreme load of the mechanical components during the entire life,an automatic method based on the bootstrapping technology(BT)is proposed to determine the most suitable threshold.Based on all the turning points of the load history and a series of thresholds estimated in advance,the generalized Pareto distribution is established to fit the exceedances.The corresponding distribution parameters are estimated with the maximum likelihood method.Then,BT is employed to calculate the mean squared error(MSE)of each estimated threshold based on the exceedances and the specific distribution parameters.Finally,the threshold with the smallest MSE will be the optimal one.Compared to the kurtosis method and the mean excess function method,the average deviation of the probability density function of exceedances determined by BT reduces by 38.52%and 29.25%,respectively.Moreover,the quantile-quantile plot of the exceedances determined by BT is closer to a straight line.The results suggest the improvement of the modeling flexibility and the determined threshold precision.If the exceedances are insufficient,BT will enlarge their amount by resampling to solve the instability problem of the original distribution parameters.
