是用遗传算法求函数：f(x1,x2)=100(x1*x1

神经网络中神经元线性模型的C语言实现源程序！-Neural network linear model neurons realize the C language source code!

自适应小生境递阶遗传算法求解函数优化问题 想对其他函数求解就修改function函数-adaptive hierarchical genetic algorithm optimization function to function for the other amendments on the funct ion function

bayes network 的分三角形的算法-bayes network breakdown triangle algorithm

--- -一个介绍支持向量机的文档。 认真读一读，会有收获的。-------- introduced a Support Vector Machine documents. Seriously read, there will be rewarding.

图像检测模板匹配的Visual C++代码-Image Detection template matching Visual C code

利用遗传算法进行文本聚类的特征选择.把一种特征组合看作一个染色体，对其进行二进制编码，引入文本集密度作为适应度函数进行特征个体适应度的评价. -The use of genetic algorithm feature selection in text clustering. To a combination of characteristics as a chromosome, its binary code, the introduction of text set density as a characteristic of fitness function for evaluation of individual fitness.

改进遗传算法－郭涛算法做最优化问题很管用,算法的基本思想是 先任意产生n个随机数，然后从n个数里随机选择m个数，再有这m个 数合成一个新数，将这个新数同n个数中间适应值函数值的最差的比较， 如果好的话就取代最差的那个，如果它比最好的还要好的话，则把最好的 也取代。如果比最差的坏，则重新合成一个新数。依次循环下去。 程序的奇妙之处是GA_crossover()函数，产生的新数确实比较好，看看 那位大侠能改进一下，产生比这跟好的数。-improved genetic algorithm- Guo Tao done optimization algorithm is very effective, and the algorithm is the first basic idea arbitrary n generated random number and then the number li n m randomly selected number, and this number m of a new synthesis of the new middle with n number fitness function of the worst by comparison, if a good case to replace the worst, um, if it even than the best, then ruled the best have replaced. If worse than the worst, a de novo synthesis of new. Followed by the cycle continues. The magic of procedures is GA_crossover () function, the new really good to see heroes who can be improved upon, it generated more than a few with a good.

本文采用红外光电传感器技术, 建立了一个基于神经网络的客流计数系统, 与传 统的仅利用光电传感器计数方法相比, 该方法提高了实时客流计数准确率。为了更好的分 类识别计数, 针对采集的连续的时空序列客流数据特点, 改进了已有的分割方法和针对客 流实际情况的特征提取方法。最后经过稀疏和密集客流模拟仿真测试, 结果表明, 本文所 提出的模型能够识别多人并排的情况, 误差率较低, 具有理论和实际意义。-RBF

This toolbox contains re-implementations of four different multi-instance learners, i.e. Diverse Density, Citation-kNN, Iterated-discrim APR, and EM-DD. Ensembles of these single multi-instance learners can be built with this toolbox

多层神经网络范例 http://www.codeproject.com/cpp/MLP.asp?df=100&forumid=148477&exp=0&select=1141594#xx1141594-multilayer neural network model http :// www.codeproject.com/cpp/MLP.asp df = 1