Voice-Pathology-Detection

shape context的MATLAB一个Demo，好像是从原作者网站上下下来的(shape context of MATLAB a Demo, seems to be down from the original author site down)

文献关于matlab遗传算法的应用，matlab遗传算法工具箱，实数编码(Literature on the application of genetic algorithm matlab, matlab genetic algorithm toolbox, real-coded)

两本关于matlab7.0的很好的教程书籍，主要是关于数值计算方面的例程(Two about matlab7.0 the good tutorial books, mainly on numerical calculation routines)

本文采用最小二乘支持向量机（LSSVM）模型，根据浙江台州某地区的历史负荷数据和气象数据，分析影响预测的各种因素，总结了负荷变化的规律性，对历史负荷数据中的“异常数据”进行修正，对负荷预测中要考虑的相关因素进行了归一化处理。LSSVM中的两个参数对模型有很大影响，而目前依然是基于经验的办法解决。对此，本文采用粒子群优化算法对模型参数进行寻优，以测试集误差作为判决依据，实现模型参数的优化选择，使得预测精度有所提高。实际算例表明，本文的预测方法收敛性好、有较高的预测精度和较快的训练速度。(this paper adopted particle swarm optimization algorithm to optimized the model parameters, make the test set error as the judgments, realized the optimization of model parameters, maked prediction accuracy improved. Practical examples show that convergence of prediction method was pefect, had a higher prediction accuracy and fast training speed.)

This file implement Gda algorithm for dimension reduction and clustering techniques. Dimension reduction, Clustering.

岩体稳定性研究在工程中十分重要，岩体破坏引起崩塌落石等。(Stability of rock mass is very important in engineering, rock mass damage caused by falling rock and so on collapse.)

说明：  A class for efficient manipulation of N-fold Kronecker products in terms of its operands only

说明：  MIMO信道SIMO信道下信道容量与天线数目等的关系，研究MIMO的好资料(This program demonstrates the convolution procedure)

说明：  arry elemments far fild

function pi = solve2(count,m,cc) 借助布冯投针实验 仿真求pi的值 pi=0 frq=0 randNo = binornd(1,cc,1,m) pro=zeros(1,count) for j=1:count for i = 1:m if randNo(1,i) == 1 frq = frq + 1 end pro(i) = frq/i end pi = (2*m)/frq pro(j) = pi end pro = pro,num =1:count plot(num,pro) (function pi = solve2 (count, m, cc) to vote with Buffon needle experiment simulation find the value of pi pi = 0 frq = 0 randNo = binornd (1, cc, 1, m) pro = zeros (1, count) for j = 1: count for i = 1: m if randNo (1, i) == 1 frq = frq+ 1 end pro (i) = frq/i end pi = (2* m )/frq pro (j) = pi end pro = pro, num = 1: count plot (num, pro) )