Support vector classifier (SVC)

Support vector machines can be applied to classification, anomalies detection, and regression problems. Let's dive into the support vector classifiers first.

The binary SVC

The first classifier to be evaluated is the binary (2-class) support vector classifier. The implementation uses the LIBSVM library created by Chih-Chung Chang and Chih-Jen Lin from the National Taiwan University [8:9].

LIBSVM

The library was originally written in C and ported to Java. It can be downloaded from http://www.csie.ntu.edu.tw/~cjlin/libsvm as a .zip or tar.gzip file. The library includes the following classifier modes:

• Support vector classifiers (C-SVC, υ-SVC, and one-class SVC)
• Support vector regression (υ-SVR and ε-SVR)
• RBF, linear, sigmoid, polynomial, and precomputed kernels

LIBSVM has the distinct advantage of using Sequential Minimal Optimization (SMO), which reduces the time complexity of a training of n observations to O(n²). LIBSVM documentation covers...