文章针对混合教学数据的特征分析与分级问题,提出了一种基于核主成分分析(Kernel Principal Component Analysis,KPCA)的方法。文章融合传统教学数据和在线教育数据,形成完整的教育特征数据链;利用复杂网络的研究方法和理论模型,构建教...文章针对混合教学数据的特征分析与分级问题,提出了一种基于核主成分分析(Kernel Principal Component Analysis,KPCA)的方法。文章融合传统教学数据和在线教育数据,形成完整的教育特征数据链;利用复杂网络的研究方法和理论模型,构建教育特征数据网络模型;采用非线性KPCA特征抽取算法提取出影响教育质量的群集特征;建立针对教育质量、学生级别进行综合评价的评价模型,并对学生进行分级,从而更好地提供个性化教育干预,提高教学质量。通过实验验证了该方法的有效性和优越性,表明该方法能够有效地提取混合教学数据的非线性特征,并能够实现混合教学数据的分级。展开更多
Phytoplankton and environment factors were investigated in 2015 and phytoplankton functional groups were used to understand their temporal and spatial distribution and their driving factors in Wanfeng Reservoir. Seven...Phytoplankton and environment factors were investigated in 2015 and phytoplankton functional groups were used to understand their temporal and spatial distribution and their driving factors in Wanfeng Reservoir. Seventeen functional groups(B, D, E, F, G, J, Lo, MP, P, S1, T, W1, W2, X1, X2, Xph, Y) were identified based on 34 species. The dominant groups were: J/B/P/D in dry season, X1/J/Xph/G/T in normal season and J in flood season. Phytoplankton abundance ranged from 5.33×10~4 cells/L to 3.65×10~7 cells/L, with the highest value occurring in flood season and lowest in dry season. The vertical profi le of dominant groups showed little differentiation except for P, which dominated surface layers over 20 m as a result of mixing water masses and higher transparency during dry season. However, the surface waters presented higher values of phytoplankton abundance than other layers, possibly because of greater irradiance. The significant explaining variables and their ability to describe the spatial distribution of the phytoplankton community in RDA diff ered seasonally as follows: dry season, NH4-N, NO_3-N, NO_2-N, TN:TP ratio and transparency(SD); normal season, temperature(WT), water depth, TN, NH4-N and NO_3-N; flood season, WT, water depth, NO_3-N and NO_2-N. Furthermore, nitrogen, water temperature, SD and water depth were significant variables explaining the variance of phytoplankton communities when datasets included all samples. The results indicated that water physical conditions and hydrology were important in phytoplankton community dynamics, and nitrogen was more important than phosphorus in modifying phytoplankton communities. Seasonal differences in the relationship between the environment and phytoplankton community should be considered in water quality management.展开更多
文摘针对近场条件下数字阵列雷达导向矢量幅相非一致性对自适应波束形成(adaptive beamforming,ADBF)算法性能的影响,通过构建近场多通道数字阵列雷达回波信号模型,分析近场多通道信号二维频谱,发现在近场条件下带限干扰信号的频谱会出现非均匀分布,呈现周期性栅格分布特征,造成算法性能下降.本文提出一种具有全新干扰样本选择策略的近场ADBF(near field ADBF, NF-ADBF)算法,通过寻优干扰信号频谱栅格边界,在栅格区间进行多门限样本筛选,离散提取干扰信号样本,构建完备的干扰信号协方差矩阵,提升近场条件下的自适应处理性能.通过在地面搭建仿真试验环境,模拟典型的数字阵列近场工作环境,通过录取试验数据分析并与理论仿真进行对比,验证了近场干扰样本筛选策略和NF-ADBF算法模型的有效性.
文摘文章针对混合教学数据的特征分析与分级问题,提出了一种基于核主成分分析(Kernel Principal Component Analysis,KPCA)的方法。文章融合传统教学数据和在线教育数据,形成完整的教育特征数据链;利用复杂网络的研究方法和理论模型,构建教育特征数据网络模型;采用非线性KPCA特征抽取算法提取出影响教育质量的群集特征;建立针对教育质量、学生级别进行综合评价的评价模型,并对学生进行分级,从而更好地提供个性化教育干预,提高教学质量。通过实验验证了该方法的有效性和优越性,表明该方法能够有效地提取混合教学数据的非线性特征,并能够实现混合教学数据的分级。
基金Supported by the Department of Science and Technology of Guizhou Province(Nos.[2014]7001,[2015]2001,[2015]10)the Water Resources Department of Guizhou Province(No.KT201401)
文摘Phytoplankton and environment factors were investigated in 2015 and phytoplankton functional groups were used to understand their temporal and spatial distribution and their driving factors in Wanfeng Reservoir. Seventeen functional groups(B, D, E, F, G, J, Lo, MP, P, S1, T, W1, W2, X1, X2, Xph, Y) were identified based on 34 species. The dominant groups were: J/B/P/D in dry season, X1/J/Xph/G/T in normal season and J in flood season. Phytoplankton abundance ranged from 5.33×10~4 cells/L to 3.65×10~7 cells/L, with the highest value occurring in flood season and lowest in dry season. The vertical profi le of dominant groups showed little differentiation except for P, which dominated surface layers over 20 m as a result of mixing water masses and higher transparency during dry season. However, the surface waters presented higher values of phytoplankton abundance than other layers, possibly because of greater irradiance. The significant explaining variables and their ability to describe the spatial distribution of the phytoplankton community in RDA diff ered seasonally as follows: dry season, NH4-N, NO_3-N, NO_2-N, TN:TP ratio and transparency(SD); normal season, temperature(WT), water depth, TN, NH4-N and NO_3-N; flood season, WT, water depth, NO_3-N and NO_2-N. Furthermore, nitrogen, water temperature, SD and water depth were significant variables explaining the variance of phytoplankton communities when datasets included all samples. The results indicated that water physical conditions and hydrology were important in phytoplankton community dynamics, and nitrogen was more important than phosphorus in modifying phytoplankton communities. Seasonal differences in the relationship between the environment and phytoplankton community should be considered in water quality management.
