4.6. Support Vector Machines

Support vector algorithms are a relatively recent concept, like so many other machine learning techniques. Cortes and Vapnik (Cortes, 1995) provided one of the first formal introductions to the concept while investigating algorithms for optical character recognition at the AT&T Bell Labs.
The term “support vector machine” is a confusing name for a predictive analytics algorithm. The fact is this term is very much a misnomer: there is really no specialized hardware. But it is a powerful algorithm that has been very successful in applications ranging from pattern recognition to text mining. A support vector machine (SVM) emphasizes the interdisciplinary nature of today’s advanced analytics by drawing equally from three ...

Get Predictive Analytics and Data Mining now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.