数据挖掘中CART算法实现

使用支持向量机SVM神经网络分类数据的预测(the prediction of data classification by SVM neural network)

FUNCTION SQUARE WITH SIMULINK

这是讲凸优化与非线性优化的很好的书。国外大牛力作，做移动通信、金融、运筹的同学可好好看此书(This is a book on convex optimization and non-linear optimization which contains many detailed examples. It can be useful for Who are major in wireless communication,financial and operation research)

it is a matlab code for takagi-sugeno modeling

快速模态分解，用以处理电流信号的奇异性，找到行波突变点，用在故障输电线路故障定位上(Fast mode decomposition to handle current signal singularity point mutation found traveling wave, with the failure of transmission line fault location)

Allow to read UFF file (all format) in matlab

分析带通采样定理的原理和实现过程，说明带通信号重构原始信号的方法，仿真给出不同采样频率和低通采样性能对比(By analyzing the principle and implementation process of the bandpass sampling theorem, the method of reconstructing the original signal with the signal is explained, and the performance comparison of different sampling frequency and low pass sampling is given.)

一种用MATLAB编写的能够实现机械连杆的运动分析，并且能自动画出其速度、加速度等的图像(Using MATLAB to achieve mechanical linkage written motion analysis, and can automatically draw its speed and acceleration of image)

本程序利用奇异值分解对3通道彩色图像进行压缩分解，具体步骤如下： 压缩过程： 1. 选取子图像大小K值，把图像分解成M×M个子图像，IMG(s),s=1,2,…, M2，其中M=N/K，原始图像IMG大小为N×N。 2. 计算这M2个子图像的平均值average，对每幅子图像减去均值图像得到新图像。 3. 计算相关矩阵R，其元素定义为 。 4. 计算R的特征值与特征向量，计算每幅子图像与最大特征向量的内积，便得到编码，即压缩后的图像。 (This procedure using singular value decomposition of 3-channel color image compression decomposition, concrete steps are as follows: compression process: 1. Select the sub-image size K value, the image is decomposed into M × M sub-image, IMG (s), s = 1, 2, ..., M2, in which M = N/K, the original image IMG size N × N. 2. M2 calculate the average of sub-image average, for each sub-image minus the new images mean images. 3. Calculation of correlation matrix R, defined as its elements. 4. Calculation of the characteristics of R value and eigenvector calculated each sub-image with the largest eigenvector of the inner product, they will have to code, that is, the compressed images.)

说明：  独立成分分析算法降低原始数据噪声，并提取特征值，非常有用得数据去噪程序。(Independent component analysis algorithm reduces the raw data noise, and extract characteristic value, is more helpful data denoising procedure.)