Multi-grain chips processed by screw extrusion processing have high nutrition and production value with a low glycemic index.To analyze the effects of particle sizes on the qualities of multi-grain chips extrusion pro...Multi-grain chips processed by screw extrusion processing have high nutrition and production value with a low glycemic index.To analyze the effects of particle sizes on the qualities of multi-grain chips extrusion processing by using a single screw extruder,mesh numbers were selected as 80,100 and 120 to describe different grain particle sizes.It was found that the particle sizes of the raw materials had effects on the basic components,sensory properties,texture properties,antioxidant activities and in vitro digestibilities of extruded chips.The results showed that with the decrease of particle sizes,the moisture contents,starch contents of the chips decreased,and fat contents,dietary fiber contents increased.The edible qualities of the chips increased with the decrease of the grain sizes of raw materials.The antioxidant capacities and estimated glycemic indexes of the three kinds of chips showed a trend of decreasing first,and then increasing with the decrease of particle sizes.Correlation analysis showed that the total antioxidant capacities of chips were negatively correlated with the estimated glycemic indexes.The research results provided valuable guidance for the quality processing of multi-grain chips under extrusion processing.展开更多
To effectively address the complexity of the environment,information uncertainty,and variability among decision-makers in the event of an enterprise emergency,a multi-granularity binary semantic-based emergency decisi...To effectively address the complexity of the environment,information uncertainty,and variability among decision-makers in the event of an enterprise emergency,a multi-granularity binary semantic-based emergency decision-making method is proposed.Decision-makers use preferred multi-granularity non-uniform linguistic scales combined with binary semantics to represent the evaluation information of key influencing factors.Secondly,the weights were determined based on the proposed method.Finally,the proposed method’s effectiveness is validated using a case study of a fire incident in a chemical company.展开更多
Dissolved gas analysis(DGA)is an effective online fault diagnosis technique for large oil-immersed transformers.However,due to the limited number of DGA data,most deep learning models will be overfitted and the classi...Dissolved gas analysis(DGA)is an effective online fault diagnosis technique for large oil-immersed transformers.However,due to the limited number of DGA data,most deep learning models will be overfitted and the classification accuracy cannot be guaranteed.Therefore,this paper has introduced the idea of deep neural networks into the multi-grained cascade forest(gcForest),which is a tree-based deep learning model,and proposed an improved gcForest that can be accelerated by GPU.Firstly,in order to extract features more effectively and reduce memory consumption,the multi-grained scanning of gcForest is replaced by convolutional neural networks.Secondly,the cascade forest(CasForest)is replaced by cascade eXtreme gradient boosting(CasXGBoost)to improve the classification ability.Finally,235 DGA samples are used to train and evaluate the proposed model.The average fault diagnosis accuracy of the improved gcForest is 88.08%,while the average recall,precision,and Fl-score are 0.89,0.90,0.89,respectively.Moreover,the proposed method still has high fault diagnosis accuracy for datasets of different sizes.展开更多
Persistent memory(PM)file systems have been developed to achieve high performance by exploiting the advanced features of PMs,including nonvolatility,byte addressability,and dynamic random access memory(DRAM)like perfo...Persistent memory(PM)file systems have been developed to achieve high performance by exploiting the advanced features of PMs,including nonvolatility,byte addressability,and dynamic random access memory(DRAM)like performance.Unfortunately,these PMs suffer from limited write endurance.Existing space management strategies of PM file systems can induce a severely unbalanced wear problem,which can damage the underlying PMs quickly.In this paper,we propose a Wear-leveling-aware Multi-grained Allocator,called WMAlloc,to achieve the wear leveling of PMs while improving the performance of file systems.WMAlloc adopts multiple min-heaps to manage the unused space of PMs.Each heap represents an allocation granularity.Then,WMAlloc allocates less-worn blocks from the corresponding min-heap for allocation requests.Moreover,to avoid recursive split and inefficient heap locations in WMAlloc,we further propose a bitmap-based multi-heap tree(BMT)to enhance WMAlloc,namely,WMAlloc-BMT.We implement WMAlloc and WMAlloc-BMT in the Linux kernel based on NOVA,a typical PM file system.Experimental results show that,compared with the original NOVA and dynamic wear-aware range management(DWARM),which is the state-of-the-art wear-leveling-aware allocator of PM file systems,WMAlloc can,respectively,achieve 4.11×and 1.81×maximum write number reduction and 1.02×and 1.64×performance with four workloads on average.Furthermore,WMAlloc-BMT outperforms WMAlloc with 1.08×performance and achieves 1.17×maximum write number reduction with four workloads on average.展开更多
在细粒度图像分类中,现有的小样本学习算法未能充分结合通道和空间信息提取细粒度图像的判别性特征,导致仅依靠单一类型的特征不足以准确捕捉细粒度对象的类间差异.针对这一难题,提出了一种基于通道先验感知的多尺度细化网络,旨在有效...在细粒度图像分类中,现有的小样本学习算法未能充分结合通道和空间信息提取细粒度图像的判别性特征,导致仅依靠单一类型的特征不足以准确捕捉细粒度对象的类间差异.针对这一难题,提出了一种基于通道先验感知的多尺度细化网络,旨在有效融合通道信息和空间信息,提升小样本细粒度图像分类的性能.通道先验感知模块实现了通道维度上注意力权重的动态分配,从而高效地捕捉通道先验信息;多尺度特征聚合过程充分利用细粒度图像中丰富的上下文信息,获取丰富的空间和边界细节特征;最后,特征细化模块对上述提取的通道和空间维度信息进行优化,实现了对关键区域的动态选择和强调,进而融合形成更精细、更具代表性的混合特征表示.所提算法在以Conv-4作为骨干网络时,在Stanford Cars、Stanford Dogs和CUB-200-2011三个细粒度数据集上的实验分类性能显著提升.在5 way 1 shot分类任务中,三个数据集的准确率分别达到了79.95%、66.97%和81.91%;在5 way 5 shot分类任务中,准确率则分别为93.42%、82.48%和93.19%.展开更多
现有的下一个兴趣点(point of interest,PoI)推荐技术存在三个主要问题:使用过于简单的方法构建用户兴趣模型、忽略用户和PoI之间在时空维度上的互动以及未能充分挖掘用户间复杂的高阶交互信息。针对这些问题,提出一种新颖的超图学习模...现有的下一个兴趣点(point of interest,PoI)推荐技术存在三个主要问题:使用过于简单的方法构建用户兴趣模型、忽略用户和PoI之间在时空维度上的互动以及未能充分挖掘用户间复杂的高阶交互信息。针对这些问题,提出一种新颖的超图学习模型FSTMH,细粒度地融合时间、空间和语义信息,用于下一个PoI推荐。FSTMH包括细粒度嵌入模块和多层次嵌入模块。前者通过使用地理图卷积网络和有向超图卷积网络进行学习,获取对应的嵌入信息,并通过对比学习提升PoI表示的质量,使用细粒度超图卷积网络学习该模块的PoI嵌入;后者将多层语义超图输入到多层超图卷积网络,学习多层次语义的PoI嵌入表示。最后,模型将两个模块的PoI嵌入向量进行组合,生成最终的top-K预测结果。通过在广泛使用的三个社交网络公共数据集上进行多种实验,结果均表明FSTMH模型表现出色,说明该新模型可作为提高下一个PoI推荐的有效方法。展开更多
基金Support by the National Key Research and Development Project of the 13th Five-Year Plan,China(2017YFD0401204)。
文摘Multi-grain chips processed by screw extrusion processing have high nutrition and production value with a low glycemic index.To analyze the effects of particle sizes on the qualities of multi-grain chips extrusion processing by using a single screw extruder,mesh numbers were selected as 80,100 and 120 to describe different grain particle sizes.It was found that the particle sizes of the raw materials had effects on the basic components,sensory properties,texture properties,antioxidant activities and in vitro digestibilities of extruded chips.The results showed that with the decrease of particle sizes,the moisture contents,starch contents of the chips decreased,and fat contents,dietary fiber contents increased.The edible qualities of the chips increased with the decrease of the grain sizes of raw materials.The antioxidant capacities and estimated glycemic indexes of the three kinds of chips showed a trend of decreasing first,and then increasing with the decrease of particle sizes.Correlation analysis showed that the total antioxidant capacities of chips were negatively correlated with the estimated glycemic indexes.The research results provided valuable guidance for the quality processing of multi-grain chips under extrusion processing.
文摘To effectively address the complexity of the environment,information uncertainty,and variability among decision-makers in the event of an enterprise emergency,a multi-granularity binary semantic-based emergency decision-making method is proposed.Decision-makers use preferred multi-granularity non-uniform linguistic scales combined with binary semantics to represent the evaluation information of key influencing factors.Secondly,the weights were determined based on the proposed method.Finally,the proposed method’s effectiveness is validated using a case study of a fire incident in a chemical company.
基金supported in part by the National Natural Science Foundation of China under Grant(52277138)Natural Science Foundation of Guangxi under Grant(2018JJB160064,2018JJA160176)。
文摘Dissolved gas analysis(DGA)is an effective online fault diagnosis technique for large oil-immersed transformers.However,due to the limited number of DGA data,most deep learning models will be overfitted and the classification accuracy cannot be guaranteed.Therefore,this paper has introduced the idea of deep neural networks into the multi-grained cascade forest(gcForest),which is a tree-based deep learning model,and proposed an improved gcForest that can be accelerated by GPU.Firstly,in order to extract features more effectively and reduce memory consumption,the multi-grained scanning of gcForest is replaced by convolutional neural networks.Secondly,the cascade forest(CasForest)is replaced by cascade eXtreme gradient boosting(CasXGBoost)to improve the classification ability.Finally,235 DGA samples are used to train and evaluate the proposed model.The average fault diagnosis accuracy of the improved gcForest is 88.08%,while the average recall,precision,and Fl-score are 0.89,0.90,0.89,respectively.Moreover,the proposed method still has high fault diagnosis accuracy for datasets of different sizes.
基金Project supported by the National Natural Science Foundation of China(No.62162011)the Doctor Funds of Guizhou University,China(Nos.2020(13)and 2022(44))。
文摘Persistent memory(PM)file systems have been developed to achieve high performance by exploiting the advanced features of PMs,including nonvolatility,byte addressability,and dynamic random access memory(DRAM)like performance.Unfortunately,these PMs suffer from limited write endurance.Existing space management strategies of PM file systems can induce a severely unbalanced wear problem,which can damage the underlying PMs quickly.In this paper,we propose a Wear-leveling-aware Multi-grained Allocator,called WMAlloc,to achieve the wear leveling of PMs while improving the performance of file systems.WMAlloc adopts multiple min-heaps to manage the unused space of PMs.Each heap represents an allocation granularity.Then,WMAlloc allocates less-worn blocks from the corresponding min-heap for allocation requests.Moreover,to avoid recursive split and inefficient heap locations in WMAlloc,we further propose a bitmap-based multi-heap tree(BMT)to enhance WMAlloc,namely,WMAlloc-BMT.We implement WMAlloc and WMAlloc-BMT in the Linux kernel based on NOVA,a typical PM file system.Experimental results show that,compared with the original NOVA and dynamic wear-aware range management(DWARM),which is the state-of-the-art wear-leveling-aware allocator of PM file systems,WMAlloc can,respectively,achieve 4.11×and 1.81×maximum write number reduction and 1.02×and 1.64×performance with four workloads on average.Furthermore,WMAlloc-BMT outperforms WMAlloc with 1.08×performance and achieves 1.17×maximum write number reduction with four workloads on average.
文摘在细粒度图像分类中,现有的小样本学习算法未能充分结合通道和空间信息提取细粒度图像的判别性特征,导致仅依靠单一类型的特征不足以准确捕捉细粒度对象的类间差异.针对这一难题,提出了一种基于通道先验感知的多尺度细化网络,旨在有效融合通道信息和空间信息,提升小样本细粒度图像分类的性能.通道先验感知模块实现了通道维度上注意力权重的动态分配,从而高效地捕捉通道先验信息;多尺度特征聚合过程充分利用细粒度图像中丰富的上下文信息,获取丰富的空间和边界细节特征;最后,特征细化模块对上述提取的通道和空间维度信息进行优化,实现了对关键区域的动态选择和强调,进而融合形成更精细、更具代表性的混合特征表示.所提算法在以Conv-4作为骨干网络时,在Stanford Cars、Stanford Dogs和CUB-200-2011三个细粒度数据集上的实验分类性能显著提升.在5 way 1 shot分类任务中,三个数据集的准确率分别达到了79.95%、66.97%和81.91%;在5 way 5 shot分类任务中,准确率则分别为93.42%、82.48%和93.19%.
文摘现有的下一个兴趣点(point of interest,PoI)推荐技术存在三个主要问题:使用过于简单的方法构建用户兴趣模型、忽略用户和PoI之间在时空维度上的互动以及未能充分挖掘用户间复杂的高阶交互信息。针对这些问题,提出一种新颖的超图学习模型FSTMH,细粒度地融合时间、空间和语义信息,用于下一个PoI推荐。FSTMH包括细粒度嵌入模块和多层次嵌入模块。前者通过使用地理图卷积网络和有向超图卷积网络进行学习,获取对应的嵌入信息,并通过对比学习提升PoI表示的质量,使用细粒度超图卷积网络学习该模块的PoI嵌入;后者将多层语义超图输入到多层超图卷积网络,学习多层次语义的PoI嵌入表示。最后,模型将两个模块的PoI嵌入向量进行组合,生成最终的top-K预测结果。通过在广泛使用的三个社交网络公共数据集上进行多种实验,结果均表明FSTMH模型表现出色,说明该新模型可作为提高下一个PoI推荐的有效方法。
