说明：  机器学习matlab源代码，包括多分类SVM，模式识别，特征选择，回归等算法。
(The spider is intended to be a complete object orientated environment for machine learning in Matlab. Aside from easy use of base learning algorithms, algorithms can be plugged together and can be compared with, e.g model selection, statistical tests and visual plots. Core library objects wilcoxon Wilcoxon object- significance test of results joint_kernel Joint kernel object- for calculating inner products in joint feature spaces joint_data Joint_data object corrt_test corrected resampled t-test object- significance test of results Unsupervised objects spectral Spectral clustering ppca Probabilistic Principal Components Analysis nmf Non-negative Matrix Factorization object mrank Manifold Ranking Density Estimation objects indep Feature selection by independent density estimation Feature Selection objects r2w2_sel Feature scaling/selection via SVMs and r^2/w^2 bound. nfe Non-Linear Feature Elimination mutin)