AI 人工智能 在C#环境下 实现 过河问题 算法

数据挖掘中k-means聚类算法的实现源代码-clustering algorithm source code in data mining

基于人工神经网络求解TSP问题的论文,很不错.-Based on artificial neural network to solve TSP problem papers, is pretty good.

本多功能自控系统软实验平台包括经典控制理论、现代控制理论、系统辨识、最优控制、智能控制等的软实验内容，其功能接近和部分超过了Matlab的自动控制工具箱，其人性化的中文界面，让使用者能轻松完成各种控制系统实验，是学习控制理论、设计控制系统的有力工具。-the multifunction control system experimental platform including soft classical control theory, the modern control theory, system identification, optimal control, intelligent control of the soft experiments, its function and some close over the Matlab toolbox control, the human interface to help users easily End into a variety of experimental control system, control theory is the study and design of the control system powerful tool.

这是一个的遗传算法程序,开发平台是vc7.0-This is a genetic algorithm process, the development platform is vc7.0

用mfc写的一个五子棋游戏，不含人工智能-mfc written with a 331 game, which was non-AI

SOBI算法，采用二阶统计量进行盲分离。-SOBI algorithm, using second-order statistic for blind separation.

由于K－均值聚类算法局部最优的特点，而模拟退火算法理论上具有全局最优的特点。因此，用模拟退火算法对聚类进行了改进。20组聚类仿真表明，平均每次对K结果值改进8次左右，效果显著。下一步工作：实际上在高温区随机生成邻域是个组合爆炸问题（见本人上载软件‘k－均值聚类算法’所述），高温跳出局部解的概率几乎为0，因此正考虑采用凸包约束进行模拟聚类，相关工作正在进行。很快将奉献给各位朋友。-as K-means clustering algorithm for optimal local characteristics, and simulated annealing algorithm theory with the characteristics of the global optimum. Thus, simulated annealing algorithm for clustering improvements. Cluster Group of 20 simulations show that the average value of K results improved about eight times, the results are obvious. The next step : In fact, in high temperature generated random neighborhood is a combination of explosives (see my software on the "k-means clustering algorithm" mentioned above), high-temperature solution of partial out almost zero probability, it is considering the use of convex hull bound for simulation cluster, the work under way . Soon dedication to the ladies.

bp神经网络在人口预测上的应用，可以根据前几年的人口数值预测出以后的人口数值-bp neural network prediction in the population on the application, based on previous years, the population of numerical prediction of the future value of the population

基于梯度训练方法的rbf神经网络实现，采用随机训练及测试数据-Gradient-based training methods of rbf neural network, using a random training and test data

神经网络7种C语言实现代码，大家可以-Neural Networks 7 kinds of C language code, everyone can