In machine learning, support vector machines (SVM) are supervised learning models with associated learning algorithms that analyze data used for classification and regression analysis (Wikipedia).
These slides show the background of the approach in the classification context. We address the binary classification problem, the soft-margin principle, the construction of the nonlinear classifiers by means of the kernel functions, the feature selection process, the multiclass SVM.
The presentation is complemented by the implementation of the approach under the open source software Python (Scikit-Learn), R (e1071) and Tanagra (SVM and C-SVC).
Keywords: svm, e1071 package, R software, Python, scikit-learn package, sklearn
Components: SVM, C-SVC
Slides: Support Vector Machine (SVM)
Dataset: svm exemples.xlsx
References:
Abe S., "Support Vector Machines for Pattern Classification", Springer, 2010.
