reference：[非常好的两本书。再加上libsvm的源码与调参的论文。]
[1]http://files2.syncfusion.com/Downloads/Ebooks/support_vector_machines_succinctly.pdf
[2]An Introduction to Support Vector Machines and Other Kernel-based Learning Methods
[3]https://pan.baidu.com/share/link?shareid=262520779&uk=1378138793干货
首先，SVM是解决supervised learning 中classification问题。有两种情况，看是否linearly separable。线性不可分则引入kernel，想法为先做transformation到其他空间进而转为可分问题。
对于线性可分的监督分类问题，SVM的目标是什么呢? find the optimal separating hyperplane which maximizes the margin of the training data
为什么以最大化间隔为目标？因为it correctly classifies the training data and because it is the one which will generalize better with unseen data
SVM
字数：3.2k字 | 预计阅读时长：14分钟
