Support Vector Machine
Support vector machines (SVMs) represent a set of supervised learning techniques that create a function from training data. The training data usually consist of pairs of input objects (typically vectors) and desired outputs. The learned function can be used to predict the output of a new object. SVMs are typically used for classification where the function outputs one of finite classes. SVMs are also used for regression and preference learning, for which they are called support vector regression (SVR) and ranking SVM, respectively. SVMs belong to a family of generalized linear classifier where the classification (or boundary) function is a hyperplane in the feature space. Two special properties of SVMs are that SVMs achieve (i) high generalization (Generalization denotes the performance of the learned function on testing data or “unseen” data that are excluded in training.) by maximizing the margin (Margin denotes the distance between the hyperplane and the...
- 3.Herbrich R, Graepel T, Obermayer K, editors. Large margin rank boundaries for ordinal regression. Cambridge, MA: MIT Press; 2000.Google Scholar
- 4.Smola AJ, Scholkopf B. A tutorial on support vector regression. Technical report, NeuroCOLT2 technical report NC2-TR-1998-030. 1998.Google Scholar
- 5.Yu H. SVM selective sampling for ranking with application to data retrieval. In: Proceedings of the 11th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; 2005.Google Scholar