样本数据

基于小样本数据的航天器自主导航方法及其DSP实现研究

Research of Spacecraft Autonomous Navigation Algorithm and Its DSP Realization Based on Small Sample Data

在数据管理方面，采用XML技术实现对样本数据、预测模型、用户等信息的存储与管理，进一步提高系统运行效率和稳定性。

In the data management , we use XML to collect and store the sample data and forecast model .

用e-mail提交一份样本数据给我。

In the response body of an e-mail to me .

为此，通过BP神经网络对样本数据进行训练，得到非线性校正模型及精度值，然后与传统的曲线拟合方法进行比较分析。

The nonlinear correction model and precision value was obtained by training the sample data with BP neural network .

基于Bootstrap方法的正态分布样本数据生成研究

Study of Sampled Data Creation for Norm Distribution on Bootstrap Method

同时仅用较少数量的DCT系数就能有效的表征样本数据，进而可以达到加快整个人脸识别过程的速度的目的。

While only a small number of DCT coefficients can be effectively characterized by sample data .

首先，利用小波变换提取扰动信号的特征矢量样本数据；然后，应用模糊C均值聚类的方法将所提取的连续的特征矢量样本数据离散化，得到离散化后的分类知识规则表；

The continuous eigenvector sample data of disturbance signal is extracted by wavelet transform and then dispersed by fuzzy C-means clustering algorithm to achieve a rule table of classification knowledge .

论文在介绍非线性理论方法的基础上，讨论了ARCH类模型的参数估计及样本数据ARCH效应的检验方法。

On the basis of introducing non-linear theory , this paper discusses the methods of parameter estimating and ARCH test of the sample data .

结果表明，基于均匀设计、MATLAB建模与优化程序的小样本数据挖掘在氧化镍纳米晶合成试验中得到了成功的应用。

The results showed that the data mining base on uniform design and MATLAB programs of modeling and optimization was applied to synthesis of nanometer nickel oxide successfully .

对于大样本数据聚类算法，我们考察了其对Iris数据集和Shuttle数据集的聚类效果。

For Large Sample Clustering Algorithm , the experiments on the Iris data set and the Shuttle data set are studied .

利用SPSS软件对样本数据进行多元线性回归分析和描述性统计，验证了微利上市公司存在盈余管理行为的假设。

Sample data is analysised by using multiple linear regression analysis and descriptive statistics of spss software to verify the hypothesis .

基于C～2W模型与C～2WY模型的样本数据包络分析方法研究

Research on Sample Data Envelopment Analysis Method Based on C ~ 2 W Model and C ~ 2WY Model

运用神经网络对铝电解生产过程的环境负荷进行预测，在负荷预测过程中，首先对样本数据进行归一化处理，然后采用BP算法对神经网络进行训练。

Environmental load is predicted with neural networks . In the prediction process , sample data is firstly unitarily treated ; then the network is trained with BP arithmetic method ;

系统分为训练和诊断两部分，系统首先根据给定样本数据进行训练，然后保存BP网络权值矩阵。

The system is divided into two parts : Training and diagnostic . First , the system is trained according to sample data set , and then save trained weight matrix .

运用小波分析理论和MATLAB软件的小波分析工具，对61例异常病例的肢体血流图样本数据进行分析和对比，提取信号的频率、波幅等特征值，发现异常病例的肢体血流图变化的基本规律。

Analyzing and comparing 61 exceptional cases by using wavelet theory and MATLAB wavelet toolbox , including eigenvalue computing . And finding the basic rule of blood-flow curve of exceptional case .

然后介绍了如何使用模糊聚类算法和等价的前馈神经网络从样本数据中辨识离散的TS模型。

Then we introduce how to identify the TS model from sample data using fuzzy clustering algorithm and equivalent feedforward neural network .

文章选择单片机为实验环境，建立了一个全自动的边信道自动化采集平台，捕获算法信息，采集样本数据，在matlab中分析处理。

The article chooses microcontroller as experimental environment , establishes a fully automatic side channel acquisition platform , to catch algorithm information and collect sample data for analysis and processing in matlab .

以回代正确率为标准，利用因子组合对多个总体的样本数据建立Fisher线性判别方程。

We take the correct rate of back substitution as a standard , and use the combinatorial factor to establish Fisher distinguishing equation .

首先研究了CEO任期与企业绩效之间的直接关系。基于我国市场化背景的样本数据的检验结论表明，CEO任期与企业绩效之间存在着比较显著的正向关系。

Firstly , the test outcome about the direct relationship between CEO tenure and firm performance indicates that CEO tenure has positive influence on the firm performance significantly .

本文利用BP神经网络拟合电化领域的工艺模型，并通过正交实验获得实验样本数据，对BP网络进行演化训练。

A craft model of the electrochemistry industry is simulated by the BP neural network , and the BP neural network is trained by some sample data of the orthogonal test and evolutional algorithm .

算例表明，在相同（有限元）样本数据的情况下，进化神经网络通过自适应调节网络结构和权值，可获得比BP神经网络更高精度的映射模型，具有很强的泛化能力。

The results show that with the same FEM sample data , evolutionary neural networks can get more accurate mapping model than traditional BP neural network through self-adaptive adjustment grid structure and weight value .

最后，基于具有锥结构的综合DEA模型，给出了综合样本数据包络分析模型（Sam-C~2WY）和相应的SC~2WY-DEA有效性概念。

Finally , Base on the comprehensive data envelopment analysis with cone-structure , Sam-C ~ 2WY model and corresponding definition of SC ~ 2WY-DEA efficiency are given .

通过收集的合理的现场样本数据对爆破参数反向BP神经网络预测模型进行训练，经检验，该模型精度完全符合要求。

Through the collection of a reasonable sample data on the scene , for training blasting parameters reverse BP neural network forecasting model , inspection , the model of the accuracy fully meet the requirements .

针对非均匀分布的样本数据，提出了采用局部加权LS-SVM算法进行在线建模来提高预测精度。

For non-uniformly distributed training data , a local weighted LS-SVM method for the online modeling of continuous process is proposed .

经证明，假如由样本数据得到的联合概率函数严格为正，则该算法发现的Markov网一定是样本的最小I图。

It has been proved that if the joint probability obtained from the sample data is strictly positive , the found Markov network must be the minimal I map of the sample .

采用BP学习算法，通过对历史观测样本数据的训练，调整该神经网络的权值，建立非线性时间序列辨识模型，以此预测股票价格的变化。

The nonlinear identification model on the time series is proposed to predict the change of stock by introducing the BP learning algorithm , training the data of former sample and adjusting the weights of network .

仿真结果表明：固定尺度最小二乘支持向量机在训练各种样本数据集时，有效地避开了LS-SVM中的稀疏性问题，且训练速度快，同时具有良好的预测精度。

The simulation results indicate that fixed size LS_ - SVM shortens the training time enormously and possesses good predicting precision on different datasets .

仿真结果显示了该方法能够找到最优的参数组合。（2）K近邻方法简单直观，但样本数据过大时则计算耗时。

The result of simulation shows the algorithm can find the most suitable parameters . ( 2 ) K-Nearest Neighbor ( KNN ) is simple , but when the number of the data is larger , it needs many operations .

继而提出了利用比较行驶周期VSPBIN分布与总样本数据VSPBIN分布差异程度的行驶周期评价方法。

Further , it develops a method to evaluate driving cycles by comparing the VSP bin distribution of driving cycles with the VSP bin distribution of the entire sampled data .

本文选取了20支封闭式基金和1260支股票的2004年至2005年之间的数据，应用SPSS软件，对样本数据进行描述性统计、相关性分析和线性回归分析。

This paper has selected twenty funds and 1260 A - stocks from year 2004 to year 2005 as samples , applied descriptive statistic analysis , corresponding analysis and linear regression through SPSS .