Purpose:Mo ve recognition in scientific abstracts is an NLP task of classifying sentences of the abstracts into different types of language units.To improve the performance of move recognition in scientific abstracts,...Purpose:Mo ve recognition in scientific abstracts is an NLP task of classifying sentences of the abstracts into different types of language units.To improve the performance of move recognition in scientific abstracts,a novel model of move recognition is proposed that outperforms the BERT-based method.Design/methodology/approach:Prevalent models based on BERT for sentence classification often classify sentences without considering the context of the sentences.In this paper,inspired by the BERT masked language model(MLM),we propose a novel model called the masked sentence model that integrates the content and contextual information of the sentences in move recognition.Experiments are conducted on the benchmark dataset PubMed 20K RCT in three steps.Then,we compare our model with HSLN-RNN,BERT-based and SciBERT using the same dataset.Findings:Compared with the BERT-based and SciBERT models,the F1 score of our model outperforms them by 4.96%and 4.34%,respectively,which shows the feasibility and effectiveness of the novel model and the result of our model comes closest to the state-of-theart results of HSLN-RNN at present.Research limitations:The sequential features of move labels are not considered,which might be one of the reasons why HSLN-RNN has better performance.Our model is restricted to dealing with biomedical English literature because we use a dataset from PubMed,which is a typical biomedical database,to fine-tune our model.Practical implications:The proposed model is better and simpler in identifying move structures in scientific abstracts and is worthy of text classification experiments for capturing contextual features of sentences.Originality/value:T he study proposes a masked sentence model based on BERT that considers the contextual features of the sentences in abstracts in a new way.The performance of this classification model is significantly improved by rebuilding the input layer without changing the structure of neural networks.展开更多
Vertical transport is critical to the movement of oil spills in seawater. Breaking waves play an important role by developing a well-defined mixing layer in the upper part of the water column. A three-dimensional （3-...Vertical transport is critical to the movement of oil spills in seawater. Breaking waves play an important role by developing a well-defined mixing layer in the upper part of the water column. A three-dimensional （3-D） Lagrangian random walk oil spill model was used here to study the influence of sea surface waves on the vertical turbulence movement of oil particles. Three vertical diffusion schemes were utilized in the model to compare their impact on oil dispersion and transportation. The first scheme calculated the vertical eddy viscosity semi-empirically. In the second scheme, the vertical diffusion coefficient was obtained directly from an Eulerian hydrodynamic model （Princeton Ocean Model, POM2k） while considering wave- caused turbulence. The third scheme was formulated by solving the Langevin equation. The trajectories, percentages of oil particles intruding into water, and the vertical distribution structures of oil particles were analyzed for a series of numerical experiments with different wind magnitudes. The results showed that the different vertical diffusion schemes could generate different horizontal trajectories and spatial distributions of oil spills on the sea surface. The vertical diffusion schemes caused different water-intruding and resurfacing oil particle behaviors, leading to different horizontal transport of oil particles at the surface and subsurface of the ocean. The vertical diffusion schemes were also applied to a realistic oil spill simulation, and these results were compared to satellite observations. All three schemes yielded acceptable results, and those of the third scheme most closely simulated the observed data.展开更多
基金supported by the project “The demonstration system of rich semantic search application in scientific literature” (Grant No. 1734) from the Chinese Academy of Sciences
文摘Purpose:Mo ve recognition in scientific abstracts is an NLP task of classifying sentences of the abstracts into different types of language units.To improve the performance of move recognition in scientific abstracts,a novel model of move recognition is proposed that outperforms the BERT-based method.Design/methodology/approach:Prevalent models based on BERT for sentence classification often classify sentences without considering the context of the sentences.In this paper,inspired by the BERT masked language model(MLM),we propose a novel model called the masked sentence model that integrates the content and contextual information of the sentences in move recognition.Experiments are conducted on the benchmark dataset PubMed 20K RCT in three steps.Then,we compare our model with HSLN-RNN,BERT-based and SciBERT using the same dataset.Findings:Compared with the BERT-based and SciBERT models,the F1 score of our model outperforms them by 4.96%and 4.34%,respectively,which shows the feasibility and effectiveness of the novel model and the result of our model comes closest to the state-of-theart results of HSLN-RNN at present.Research limitations:The sequential features of move labels are not considered,which might be one of the reasons why HSLN-RNN has better performance.Our model is restricted to dealing with biomedical English literature because we use a dataset from PubMed,which is a typical biomedical database,to fine-tune our model.Practical implications:The proposed model is better and simpler in identifying move structures in scientific abstracts and is worthy of text classification experiments for capturing contextual features of sentences.Originality/value:T he study proposes a masked sentence model based on BERT that considers the contextual features of the sentences in abstracts in a new way.The performance of this classification model is significantly improved by rebuilding the input layer without changing the structure of neural networks.
基金supported by Marine Industry Scientific Research Special Funds for Public Welfare Project-The development and application of fine-scale high precision comprehensive forecast system on the key protection coastal area(Grant No.201305031)The modular construction and application of marine forecasting operational system(Grant No.201205017)
文摘Vertical transport is critical to the movement of oil spills in seawater. Breaking waves play an important role by developing a well-defined mixing layer in the upper part of the water column. A three-dimensional （3-D） Lagrangian random walk oil spill model was used here to study the influence of sea surface waves on the vertical turbulence movement of oil particles. Three vertical diffusion schemes were utilized in the model to compare their impact on oil dispersion and transportation. The first scheme calculated the vertical eddy viscosity semi-empirically. In the second scheme, the vertical diffusion coefficient was obtained directly from an Eulerian hydrodynamic model （Princeton Ocean Model, POM2k） while considering wave- caused turbulence. The third scheme was formulated by solving the Langevin equation. The trajectories, percentages of oil particles intruding into water, and the vertical distribution structures of oil particles were analyzed for a series of numerical experiments with different wind magnitudes. The results showed that the different vertical diffusion schemes could generate different horizontal trajectories and spatial distributions of oil spills on the sea surface. The vertical diffusion schemes caused different water-intruding and resurfacing oil particle behaviors, leading to different horizontal transport of oil particles at the surface and subsurface of the ocean. The vertical diffusion schemes were also applied to a realistic oil spill simulation, and these results were compared to satellite observations. All three schemes yielded acceptable results, and those of the third scheme most closely simulated the observed data.
