基向量

Next, kernel PCA is used to construct a set of orthogonal coordinate base in feature space.

其次，在特征空间中，利用协方差矩阵的非零特征向量构造了一组正交基，并将变换后的光谱向量投影到特征空间中的正交基上。

In this paper, the performance of SVM based on Fourier kernel is studied which aims at the regression in signal processing questions, and the influence of parameter q on performance of SVM is analyzed. A conclusion is drew that the integral of Fourier kernel in one period is a constant and the concept of equivalent kernel function width is proposed.

本文针对信号处理中的回归问题，研究了基于傅立叶核函数支持向量机的性能，并分析了参数q对其性能的影响；根据推导得出傅立叶核函数在单个周期内积分为常数的结论，并提出了等效核宽度的概念；最后的仿真结果证实了，在信号处理领域，采用傅立叶核函数的支持向量机比采用径向基核函数的有着更好的性能。

This paper firstly reports three kinds of technique of text automatic category based on statistic (k nearest neighbor, support vector machine and nave bayes), and analyses their advantages and disadvantages.

本文首先介绍三种基於统计的自动分类技术（k近邻分类器、支持向量机分类器和朴素贝叶斯分类器），剖析了基於统计的自动分类的优势及不足。

Shape and color characteristics were used to form the separated pattern vector and RBF network was used as classifier for each pattern. Each classifier was linearly combined to give the final identification result. Shape pattern consists of rectangularity, elongation, roundness, sphericity, lobation and seven movement invariants. R, G, B, L histogram and R.G chromaticity histogram were treated as separated color characteristic vector.

昆虫图像的形态特征值由矩形度、延长度、球状型、叶状型、似圆度和 7个Hu不变矩等12个特征值组成，颜色特征值由红、绿、蓝、灰度真方图及基于红、绿的二维色度直方图特征值分别组成，然后建立径向基神经网络分类器，每一特征向量由独立的径向基神经网络做为分类器，最终识别由每个分类器识别结果的线性组合而成。

The normalized vector is used as the input of RBF NN, and target recognition is performed based on this, which leads to a satisfactory recognition result.

并对该特征向量进行对数归一化，将归一化的特征向量作为径向基函数神经网络的输入，在此基础上进行识别，达到较好的识别效果。

Based on analyzing regression analysis, neural network and the least square support vector machines that the three soft computing methods fitting for data fitting, in the paper try to use nonlinear regression analysis, the BP neural network based on Levenberg-Marquardt training algorithm, radial basis function neural network on orthogonal least square training algorithm and least square support vector machines for tracking the position of glass plate.

本文在分析适于数据拟合的回归分析、神经网络、最小二乘支持向量机这三种"软计算"方法的基础上，尝试将非线性回归分析方法，基于Levenberg-Marquardt算法的BP神经网络，基于正交最小二乘训练算法的径向基函数神经网络和最小二乘支持向量机应用于玻璃板位的跟踪中。

The characteristic vector of the real symmetric matrix should be found out, which is orthogonalized and normalized to a standard orthogonal base and is used as row vector to construct the transformation matrix P, so the P~(-1)AP can be made into diagonal matrix.

实对称矩阵A经相似变换P-1AP可化为对角矩阵，在x =Py 下，不一定能化A的二次型为标准型；应寻求对称矩阵A的特征向量，将其正交化并单位化作为标准正交基，作为列向量构造变换矩阵P，可使P-1AP=Λ为对角阵，在x =Py 下，要将二次型化为标准型，且二次项系数即为对角阵Λ主对角线上元素。

Based on this result,the intrinsic reasons that singular value vector-based recognition methods have the low recognition accauracy are follows: singular value vectors of arbitrary two face images have not the equivalent basis space in general;there is no one-to-one correspondence between singular value vector and face image;singular value vectors are not separated.

进一步研究发现，导致基于奇异值向量人脸识别算法识别率低的根本原因是：不同人脸图像对应的奇异值向量所在的基空间不一致、奇异值向量与人脸图像之间并不存在一一对应关系、奇异值向量具有不可分割性。

On CHE programming, the higher coupling ratio of control-gate makes the higher electrical field across TOX in Si than Ge. Also because of the continuity of displacement vector, the higher permittivity of Ge would cause the lower electrical field at interface. We get the higher gate current in Si than Ge. On CFN programming, the higher CT in Ge would show the higher electrical field across TOX. However, the parameters of F-N tunneling are calculated and showing the gate current in Si is larger than Ge. On the same mechanism of F-N tunneling erasing, the parameters also show the higher electrical filed of Si would cause the higher erasing speed. The continuity of displacement vector also explains the higher electrical field at interface for F-N tunneling programming/erasing.

从通道热电子穿隧写入的模拟结果发现，由於控制闸极耦合电容的影响，加上电位移向量在半导体-氧化层界面连续的观念，拥有较高介电常数的锗反而得到较小的等效电场，决定了穿隧电流反倒是不如矽基板；在F-N穿隧写入的模拟中，即便锗基板拥有较大的总耦合电容，使得在浮闸的耦合电压大於矽基板，但仍旧是半导体-氧化层界面的电场扮演了穿隧电流的决定性因素，得到的结果仍旧是矽基板的写入速度高於锗基板；在F-N抹除的模拟中，运用与F-N写入相同的数学模型，仍旧看见在锗基板上未能得到速度上的改善，同时用数学的计算展示了合理的解释。

The time-history velocity response signal of pile can be decomposed by Sym wavelet. The power spectrum value can be extracted from some specified spectrum range. These values from one signal makes up the characteristic vector representing this signal. The relationship between characteristic vector of pile and pile defect type can he established by using BP artificial neural network. Abundant time-history velocity response signals of pile can be acquired by numerical simulation method.

利用Sym小波对基桩速度响应时程曲线进行小波分解，对指定频带上的信号分量进行特征值提取，提取的特征值为反映各频带范围内体现能量分布的功率谱均值，提取的特征值可构成反映信号特征的特征向量，同时利用BP人工神经网络的非线性映射特性建立特征向量和基桩缺陷类别之间的一种对应关系。

We have no common name for a mime of Sophron or Xenarchus and a Socratic Conversation; and we should still be without one even if the imitation in the two instances were in trimeters or elegiacs or some other kind of verse--though it is the way with people to tack on 'poet' to the name of a metre, and talk of elegiac-poets and epic-poets, thinking that they call them poets not by reason of the imitative nature of their work, but indiscriminately by reason of the metre they write in.

索夫农 、森那库斯和苏格拉底式的对话采用的模仿没有一个公共的名称；三音步诗、挽歌体或其他类型的诗的模仿也没有——人们把&诗人&这一名词和格律名称结合到一起，称之为挽歌体诗人或者史诗诗人，他们被称为诗人，似乎只是因为遵守格律写作，而非他们作品的模仿本质。

The relationship between communicative competence and grammar teaching should be that of the ends and the means.

交际能力和语法的关系应该是目标与途径的关系。