The spatial scale(?shing grid) of ?sheries research af fects the observed spatial patterns of?sheries resources such as catch-per-unit-ef fort(CPUE) and ?shing ef fort. We examined the scale impact of high value(HH) c...The spatial scale(?shing grid) of ?sheries research af fects the observed spatial patterns of?sheries resources such as catch-per-unit-ef fort(CPUE) and ?shing ef fort. We examined the scale impact of high value(HH) clusters of the annual ?shing ef fort for Dosidicus gigas of fshore Peru from 2009 to 2012.For a multi-scale analysis, the original commercial ?shery data were tessellated to twelve spatial scales from 6′ to 72′ with an interval of 6′. Under these spatial scales, D. gigas clusters were identi?ed using the Anselin Local Moran's I. Statistics including the number of points, mean CPUE, standard deviation(SD),skewness, kurtosis, area and centroid were calculated for these HH clusters. We found that the z-score of global Moran's I and the number of points for HH clusters follow a power law scaling relationship from2009 to 2012. The mean ef fort and its SD also follow a power law scaling relationship from 2009 to 2012.The skewness follows a linear scaling relationship in 2010 and 2011 but ?uctuates with spatial scale in2009 and 2012; kurtosis follows a logarithmic scale relationship in 2009, 2011 and 2012 but a linear scale relationship in 2010. Cluster area follows a power law scaling relationship in 2010 and 2012, a linear scaling relationship in 2009, and a quadratic scaling relationship in 2011. Based on the peaks of Moran's I indices and the multi-scale analysis, we conclude that the optimum scales are 12′ in 2009 ? 2011 and 6′ in 2012, while the coarsest allowable scales are 48′ in 2009, 2010 and 2012, and 60′ in 2011. Our research provides the best spatial scales for conducting spatial analysis of this pelagic species, and provides a better understanding of scaling behavior for the ?shing ef fort of D. gigas in the of fshore Peruvian waters.展开更多
Recently a new clustering algorithm called 'affinity propagation' (AP) has been proposed, which efficiently clustered sparsely related data by passing messages between data points. However, we want to cluster ...Recently a new clustering algorithm called 'affinity propagation' (AP) has been proposed, which efficiently clustered sparsely related data by passing messages between data points. However, we want to cluster large scale data where the similarities are not sparse in many cases. This paper presents two variants of AP for grouping large scale data with a dense similarity matrix. The local approach is partition affinity propagation (PAP) and the global method is landmark affinity propagation (LAP). PAP passes messages in the subsets of data first and then merges them as the number of initial step of iterations; it can effectively reduce the number of iterations of clustering. LAP passes messages between the landmark data points first and then clusters non-landmark data points; it is a large global approximation method to speed up clustering. Experiments are conducted on many datasets, such as random data points, manifold subspaces, images of faces and Chinese calligraphy, and the results demonstrate that the two ap-proaches are feasible and practicable.展开更多
Density-based algorithm for discovering clusters in large spatial databases with noise(DBSCAN) is a classic kind of density-based spatial clustering algorithm and is widely applied in several aspects due to good perfo...Density-based algorithm for discovering clusters in large spatial databases with noise(DBSCAN) is a classic kind of density-based spatial clustering algorithm and is widely applied in several aspects due to good performance in capturing arbitrary shapes and detecting outliers. However, in practice, datasets are always too massive to fit the serial DBSCAN. And a new parallel algorithm-Parallel DBSCAN(PDBSCAN) was proposed to solve the problem which DBSCAN faced. The proposed parallel algorithm bases on MapReduce mechanism. The usage of parallel mechanism in the algorithm focuses on region query and candidate queue processing which needed substantive computation resources. As a result, PDBSCAN is scalable for large-scale dataset clustering and is extremely suitable for applications in E-Commence, especially for recommendation.展开更多
It is a challenging topic to develop an efficient algorithm for large scale classification problems in many applications of machine learning. In this paper, a hierarchical clustering and fixed-layer local learning (HC...It is a challenging topic to develop an efficient algorithm for large scale classification problems in many applications of machine learning. In this paper, a hierarchical clustering and fixed-layer local learning (HCFLL) based support vector machine(SVM) algorithm is proposed to deal with this problem. Firstly, HCFLL hierarchically clusters a given dataset into a modified clustering feature tree based on the ideas of unsupervised clustering and supervised clustering. Then it locally trains SVM on each labeled subtree at a fixed-layer of the tree. The experimental results show that compared with the existing popular algorithms such as core vector machine and decision-tree support vector machine, HCFLL can significantly improve the training and testing speeds with comparable testing accuracy.展开更多
Based on the scope economic theory of "resource curse" and industrial clusters,the scale of sugar cluster is calculated by the output of sugarcane and sugar while the scale benefit of sugar cluster is measur...Based on the scope economic theory of "resource curse" and industrial clusters,the scale of sugar cluster is calculated by the output of sugarcane and sugar while the scale benefit of sugar cluster is measured by the productivity(rate of sugar production),sales output ratio,industrial output value as well as profit margin.Positive analysis of the scale merit of sugar clusters in resource-rich area of southwestern Guangxi is conducted according to related statistics of Chongzuo City.And the primary problems of sugar clusters are pointed out.The profit created by sugar for the sugar industry in Chongzuo City has already been near capacity.The sugar industry is big but not strong.With much governmental interfernce,there is no effective connections and inadequte competitive forces among subjects of the clusters.The extention of industrial chain is limited.Therefore,measures for developing sugar clusters in resource-rich area of southwestern Guangxi are proposed.Industrial structure is to be adjusted to improve the sugarcane cultivation techniques.The industrial chain should be extended to increase the economic benefits of sugar industry.Industrial support is to be strengthened and capital output for sicence and technology increased.Price regualtion fund of grain sugar is to be established with coordination with the superior region.The transformation from savings to investment should be quickened to evade "resource curse".展开更多
Turbulent motion could be regarded as the superposition of fluctuations with different scales. It's of great theoretical and practical importance to determine the classification of turbulent scales quantitatively ...Turbulent motion could be regarded as the superposition of fluctuations with different scales. It's of great theoretical and practical importance to determine the classification of turbulent scales quantitatively to the better description of vortex motions with different scales, and to the research of the interaction among different sclaes of vortex and the construction of better turbulent models. The mathematical method, which carries out the classification on a certain requirement, is called cluster analysis. In this paper, fuzzy cluster analysis method is used to study the classification of turbulent scales quantitatively in smooth and rough wall boundary conditions. Furthermore, the properties and interactions among all kinds of flow structures are also studied. The results are helpful to gain some insight into the properties and interactions of all kinds of turbulent scales in wall turbulent shear flow.展开更多
A new algorithm for segmentation of suspected lung ROI(regions of interest)by mean-shift clustering and multi-scale HESSIAN matrix dot filtering was proposed.Original image was firstly filtered by multi-scale HESSIAN ...A new algorithm for segmentation of suspected lung ROI(regions of interest)by mean-shift clustering and multi-scale HESSIAN matrix dot filtering was proposed.Original image was firstly filtered by multi-scale HESSIAN matrix dot filters,round suspected nodular lesions in the image were enhanced,and linear shape regions of the trachea and vascular were suppressed.Then,three types of information,such as,shape filtering value of HESSIAN matrix,gray value,and spatial location,were introduced to feature space.The kernel function of mean-shift clustering was divided into product form of three kinds of kernel functions corresponding to the three feature information.Finally,bandwidths were calculated adaptively to determine the bandwidth of each suspected area,and they were used in mean-shift clustering segmentation.Experimental results show that by the introduction of HESSIAN matrix of dot filtering information to mean-shift clustering,nodular regions can be segmented from blood vessels,trachea,or cross regions connected to the nodule,non-nodular areas can be removed from ROIs properly,and ground glass object(GGO)nodular areas can also be segmented.For the experimental data set of 127 different forms of nodules,the average accuracy of the proposed algorithm is more than 90%.展开更多
场地-城市相互作用(site-city interaction,SCI)效应会显著改变场地地震波场分布及建筑反应,基于SCI效应理论计算研究方法的发展现状,发挥谱元(spectral element,SE)法可快速高效求解三维地震波场传播和多自由度(multi-degree of freedo...场地-城市相互作用(site-city interaction,SCI)效应会显著改变场地地震波场分布及建筑反应,基于SCI效应理论计算研究方法的发展现状,发挥谱元(spectral element,SE)法可快速高效求解三维地震波场传播和多自由度(multi-degree of freedom,MDOF)模型计算量小且可同时模拟大量建筑的优势,同时,结合频率波数域(frequency wave number analysis,FK)方法,以等效地震荷载的方式施加地震波场,建立了FK-SE-MDOF耦合方法,实现了SE-MDOF耦合模型中多种波型(P波、SV波和SH波)的斜入射输入,解决了当前三维SCI效应研究方法中未能同时考虑建筑非线性、频谱特性、地震波波型及入射角度影响的问题。首先对方法原理进行了介绍;然后,通过与振动台试验的对比,验证了方法的正确性;进而,采用该方法建立理想场地-城市建筑群相互作用耦合模型,主要探讨了入射角度和地震波波型对SCI效应的影响,得到了一些有益结论。该方法较为真实地反映SCI效应影响的同时,可反映建筑基础轮廓对地震波场的影响,适用于需考虑建筑轮廓信息的社区尺度SCI效应研究,可为城市规划、抗震设计、风险评估以及震后救援等工作提供定量指导。展开更多
大规模多视图聚类旨在解决传统多视图聚类算法中计算速度慢、空间复杂度高,以致无法扩展到大规模数据的问题.其中,基于锚点的多视图聚类方法通过使用整体数据集合的锚点集构建后者对于前者的重构矩阵,利用重构矩阵进行聚类,有效地降低...大规模多视图聚类旨在解决传统多视图聚类算法中计算速度慢、空间复杂度高,以致无法扩展到大规模数据的问题.其中,基于锚点的多视图聚类方法通过使用整体数据集合的锚点集构建后者对于前者的重构矩阵,利用重构矩阵进行聚类,有效地降低了算法的时间和空间复杂度.然而,现有的方法忽视了锚点之间的差异,均等地看待所有锚点,导致聚类结果受到低质量锚点的限制.为定位更具有判别性的锚点,加强高质量锚点对聚类的影响,提出一种基于加权锚点的大规模多视图聚类算法(Multi-view clustering with weighted anchors,MVC-WA).通过引入自适应锚点加权机制,所提方法在统一框架下确定锚点的权重,进行锚图的构建.同时,为增加锚点的多样性,根据锚点之间的相似度进一步调整锚点的权重.在9个基准数据集上与现有最先进的大规模多视图聚类算法的对比实验结果验证了所提方法的高效性与有效性.展开更多
利用深度学习对声呐图像进行目标检测是近年来的研究热点,然而声呐图像存在目标尺度分布集中、数据获取难等问题,导致检测效果难以满足需求。针对该问题,提出了一种基于可变尺度先验框的目标检测方法。首先,考虑到声呐图像中目标的尺度...利用深度学习对声呐图像进行目标检测是近年来的研究热点,然而声呐图像存在目标尺度分布集中、数据获取难等问题,导致检测效果难以满足需求。针对该问题,提出了一种基于可变尺度先验框的目标检测方法。首先,考虑到声呐图像中目标的尺度分布具有其特殊性,基于先验统计生成可变尺度先验框。其次,为了解决声呐图像稀缺的难题,采用数据增强的方法对训练集进行扩充。最后,探索了模型的轻量化,通过删减模型的大目标检测层,在不降低模型精度的同时简化模型结构。为了评估算法的有效性,以前视声呐图像为例进行了综合试验,平均精度(mean average precision,mAP)@0.75和mAP@0.5:0.95分别达0.585和0.559,较原Yolov5网络分别提升了5.8%和3.1%,同时每秒10亿次浮点运算次数下降到14.9。结果表明,所提算法具有更高的精度和更轻量化的模型结构。展开更多
基金Supported by the National Natural Science Foundation of China(No.41406146)the Laboratory for Marine Fisheries Science and Food Production Processes at Qingdao National Laboratory for Marine Science and Technology of China(No.2017-1A02)the Shanghai Universities First-class Disciplines Project-Fisheries(A)
文摘The spatial scale(?shing grid) of ?sheries research af fects the observed spatial patterns of?sheries resources such as catch-per-unit-ef fort(CPUE) and ?shing ef fort. We examined the scale impact of high value(HH) clusters of the annual ?shing ef fort for Dosidicus gigas of fshore Peru from 2009 to 2012.For a multi-scale analysis, the original commercial ?shery data were tessellated to twelve spatial scales from 6′ to 72′ with an interval of 6′. Under these spatial scales, D. gigas clusters were identi?ed using the Anselin Local Moran's I. Statistics including the number of points, mean CPUE, standard deviation(SD),skewness, kurtosis, area and centroid were calculated for these HH clusters. We found that the z-score of global Moran's I and the number of points for HH clusters follow a power law scaling relationship from2009 to 2012. The mean ef fort and its SD also follow a power law scaling relationship from 2009 to 2012.The skewness follows a linear scaling relationship in 2010 and 2011 but ?uctuates with spatial scale in2009 and 2012; kurtosis follows a logarithmic scale relationship in 2009, 2011 and 2012 but a linear scale relationship in 2010. Cluster area follows a power law scaling relationship in 2010 and 2012, a linear scaling relationship in 2009, and a quadratic scaling relationship in 2011. Based on the peaks of Moran's I indices and the multi-scale analysis, we conclude that the optimum scales are 12′ in 2009 ? 2011 and 6′ in 2012, while the coarsest allowable scales are 48′ in 2009, 2010 and 2012, and 60′ in 2011. Our research provides the best spatial scales for conducting spatial analysis of this pelagic species, and provides a better understanding of scaling behavior for the ?shing ef fort of D. gigas in the of fshore Peruvian waters.
基金the National Natural Science Foundation of China (Nos. 60533090 and 60603096)the National Hi-Tech Research and Development Program (863) of China (No. 2006AA010107)+2 种基金the Key Technology R&D Program of China (No. 2006BAH02A13-4)the Program for Changjiang Scholars and Innovative Research Team in University of China (No. IRT0652)the Cultivation Fund of the Key Scientific and Technical Innovation Project of MOE, China (No. 706033)
文摘Recently a new clustering algorithm called 'affinity propagation' (AP) has been proposed, which efficiently clustered sparsely related data by passing messages between data points. However, we want to cluster large scale data where the similarities are not sparse in many cases. This paper presents two variants of AP for grouping large scale data with a dense similarity matrix. The local approach is partition affinity propagation (PAP) and the global method is landmark affinity propagation (LAP). PAP passes messages in the subsets of data first and then merges them as the number of initial step of iterations; it can effectively reduce the number of iterations of clustering. LAP passes messages between the landmark data points first and then clusters non-landmark data points; it is a large global approximation method to speed up clustering. Experiments are conducted on many datasets, such as random data points, manifold subspaces, images of faces and Chinese calligraphy, and the results demonstrate that the two ap-proaches are feasible and practicable.
基金National Natural Science Foundations of China( No. 61070101,No. 60875029,No. 61175048)
文摘Density-based algorithm for discovering clusters in large spatial databases with noise(DBSCAN) is a classic kind of density-based spatial clustering algorithm and is widely applied in several aspects due to good performance in capturing arbitrary shapes and detecting outliers. However, in practice, datasets are always too massive to fit the serial DBSCAN. And a new parallel algorithm-Parallel DBSCAN(PDBSCAN) was proposed to solve the problem which DBSCAN faced. The proposed parallel algorithm bases on MapReduce mechanism. The usage of parallel mechanism in the algorithm focuses on region query and candidate queue processing which needed substantive computation resources. As a result, PDBSCAN is scalable for large-scale dataset clustering and is extremely suitable for applications in E-Commence, especially for recommendation.
基金National Natural Science Foundation of China ( No. 61070033 )Fundamental Research Funds for the Central Universities,China( No. 2012ZM0061)
文摘It is a challenging topic to develop an efficient algorithm for large scale classification problems in many applications of machine learning. In this paper, a hierarchical clustering and fixed-layer local learning (HCFLL) based support vector machine(SVM) algorithm is proposed to deal with this problem. Firstly, HCFLL hierarchically clusters a given dataset into a modified clustering feature tree based on the ideas of unsupervised clustering and supervised clustering. Then it locally trains SVM on each labeled subtree at a fixed-layer of the tree. The experimental results show that compared with the existing popular algorithms such as core vector machine and decision-tree support vector machine, HCFLL can significantly improve the training and testing speeds with comparable testing accuracy.
基金Supported by Project Launched by Guangxi Education Office (201012MS212)Special Project of"Borderland Question Research"launched by Research Center of Humanities and Social Science in Guangxi(XWSKYB2010006)Research Fund of Natural Science of Guangxi Normal University for Nationalities(XYYB2010-006)
文摘Based on the scope economic theory of "resource curse" and industrial clusters,the scale of sugar cluster is calculated by the output of sugarcane and sugar while the scale benefit of sugar cluster is measured by the productivity(rate of sugar production),sales output ratio,industrial output value as well as profit margin.Positive analysis of the scale merit of sugar clusters in resource-rich area of southwestern Guangxi is conducted according to related statistics of Chongzuo City.And the primary problems of sugar clusters are pointed out.The profit created by sugar for the sugar industry in Chongzuo City has already been near capacity.The sugar industry is big but not strong.With much governmental interfernce,there is no effective connections and inadequte competitive forces among subjects of the clusters.The extention of industrial chain is limited.Therefore,measures for developing sugar clusters in resource-rich area of southwestern Guangxi are proposed.Industrial structure is to be adjusted to improve the sugarcane cultivation techniques.The industrial chain should be extended to increase the economic benefits of sugar industry.Industrial support is to be strengthened and capital output for sicence and technology increased.Price regualtion fund of grain sugar is to be established with coordination with the superior region.The transformation from savings to investment should be quickened to evade "resource curse".
文摘Turbulent motion could be regarded as the superposition of fluctuations with different scales. It's of great theoretical and practical importance to determine the classification of turbulent scales quantitatively to the better description of vortex motions with different scales, and to the research of the interaction among different sclaes of vortex and the construction of better turbulent models. The mathematical method, which carries out the classification on a certain requirement, is called cluster analysis. In this paper, fuzzy cluster analysis method is used to study the classification of turbulent scales quantitatively in smooth and rough wall boundary conditions. Furthermore, the properties and interactions among all kinds of flow structures are also studied. The results are helpful to gain some insight into the properties and interactions of all kinds of turbulent scales in wall turbulent shear flow.
基金Projects(61172002,61001047,60671050)supported by the National Natural Science Foundation of ChinaProject(N100404010)supported by Fundamental Research Grant Scheme for the Central Universities,China
文摘A new algorithm for segmentation of suspected lung ROI(regions of interest)by mean-shift clustering and multi-scale HESSIAN matrix dot filtering was proposed.Original image was firstly filtered by multi-scale HESSIAN matrix dot filters,round suspected nodular lesions in the image were enhanced,and linear shape regions of the trachea and vascular were suppressed.Then,three types of information,such as,shape filtering value of HESSIAN matrix,gray value,and spatial location,were introduced to feature space.The kernel function of mean-shift clustering was divided into product form of three kinds of kernel functions corresponding to the three feature information.Finally,bandwidths were calculated adaptively to determine the bandwidth of each suspected area,and they were used in mean-shift clustering segmentation.Experimental results show that by the introduction of HESSIAN matrix of dot filtering information to mean-shift clustering,nodular regions can be segmented from blood vessels,trachea,or cross regions connected to the nodule,non-nodular areas can be removed from ROIs properly,and ground glass object(GGO)nodular areas can also be segmented.For the experimental data set of 127 different forms of nodules,the average accuracy of the proposed algorithm is more than 90%.
文摘场地-城市相互作用(site-city interaction,SCI)效应会显著改变场地地震波场分布及建筑反应,基于SCI效应理论计算研究方法的发展现状,发挥谱元(spectral element,SE)法可快速高效求解三维地震波场传播和多自由度(multi-degree of freedom,MDOF)模型计算量小且可同时模拟大量建筑的优势,同时,结合频率波数域(frequency wave number analysis,FK)方法,以等效地震荷载的方式施加地震波场,建立了FK-SE-MDOF耦合方法,实现了SE-MDOF耦合模型中多种波型(P波、SV波和SH波)的斜入射输入,解决了当前三维SCI效应研究方法中未能同时考虑建筑非线性、频谱特性、地震波波型及入射角度影响的问题。首先对方法原理进行了介绍;然后,通过与振动台试验的对比,验证了方法的正确性;进而,采用该方法建立理想场地-城市建筑群相互作用耦合模型,主要探讨了入射角度和地震波波型对SCI效应的影响,得到了一些有益结论。该方法较为真实地反映SCI效应影响的同时,可反映建筑基础轮廓对地震波场的影响,适用于需考虑建筑轮廓信息的社区尺度SCI效应研究,可为城市规划、抗震设计、风险评估以及震后救援等工作提供定量指导。
文摘大规模多视图聚类旨在解决传统多视图聚类算法中计算速度慢、空间复杂度高,以致无法扩展到大规模数据的问题.其中,基于锚点的多视图聚类方法通过使用整体数据集合的锚点集构建后者对于前者的重构矩阵,利用重构矩阵进行聚类,有效地降低了算法的时间和空间复杂度.然而,现有的方法忽视了锚点之间的差异,均等地看待所有锚点,导致聚类结果受到低质量锚点的限制.为定位更具有判别性的锚点,加强高质量锚点对聚类的影响,提出一种基于加权锚点的大规模多视图聚类算法(Multi-view clustering with weighted anchors,MVC-WA).通过引入自适应锚点加权机制,所提方法在统一框架下确定锚点的权重,进行锚图的构建.同时,为增加锚点的多样性,根据锚点之间的相似度进一步调整锚点的权重.在9个基准数据集上与现有最先进的大规模多视图聚类算法的对比实验结果验证了所提方法的高效性与有效性.
文摘利用深度学习对声呐图像进行目标检测是近年来的研究热点,然而声呐图像存在目标尺度分布集中、数据获取难等问题,导致检测效果难以满足需求。针对该问题,提出了一种基于可变尺度先验框的目标检测方法。首先,考虑到声呐图像中目标的尺度分布具有其特殊性,基于先验统计生成可变尺度先验框。其次,为了解决声呐图像稀缺的难题,采用数据增强的方法对训练集进行扩充。最后,探索了模型的轻量化,通过删减模型的大目标检测层,在不降低模型精度的同时简化模型结构。为了评估算法的有效性,以前视声呐图像为例进行了综合试验,平均精度(mean average precision,mAP)@0.75和mAP@0.5:0.95分别达0.585和0.559,较原Yolov5网络分别提升了5.8%和3.1%,同时每秒10亿次浮点运算次数下降到14.9。结果表明,所提算法具有更高的精度和更轻量化的模型结构。