Unformatted text preview: lasses are not linearly separable a
hyperplane is selected such that as few as possible document vectors are located
on the “wrong” side.
SVMs can be used with non-linear predictors by transforming the usual input
features in a non-linear way, e.g. by deﬁning a feature map
φ ( t 1 , . . . , t N ) = t 1 , . . . , t N , t 2 , t 1 t 2 , . . . , t N t N −1 , t 2
Subsequently a hyperplane may be deﬁned in the expanded input space. Obviously such non-linear transformations may be deﬁned in a large number of
ways. Band 20 – 2005 35 Hotho, Nürnberger, and Paaß
The most important property of SVMs is that learning is nearly independent
of the dimensionality of the feature space. It rarely requires feature selection
as it inherently selects data points (the support vectors) required for a good
classiﬁcation. This allows good generalization even in the presence of a large
number of features and makes SVM especially suitable for the classiﬁcation
of texts (Joachims 1998). In the case of textual data the choice of the kernel
function has a minimal effect on the accuracy of classiﬁ...
View Full Document