Support vector machines are a set of algorithms that learn from data by creating models that maximize their margin of error.<\/p>\n
Support vector machines<\/a>\u00a0(SVMs) are a family of algorithms for\u00a0classification<\/a>,\u00a0regression<\/a>,\u00a0transduction<\/a>, novelty detection<\/a>, and\u00a0semi-supervised learning<\/a>. They work by choosing a model that\u00a0maximizes the error margin<\/a> of a training set.<\/p>\n SVMs\u00a0were originally developed by\u00a0Vladimir Vapnik<\/a> in 1963. Since the mid-90s, a energetic research community has grown around them. If you want to learn more about SVMs, you can read Chris Burges’ tutorial<\/a>.\u00a0Nello Cristianini<\/a> and\u00a0John Shawe-Taylor<\/a> have written\u00a0a textbook<\/a> about them.\u00a0Bernhard Sch\u00f6lkopf<\/a> and\u00a0Alex Smola<\/a> wrote a textbook about kernel methods<\/a>, which are a closely-related set of methods.<\/p>\n Since 1998, we’ve done basic research into making SVMs be more user-friendly. Our research has resulted in:<\/p>\n