Support Vector Machine
Support vector machines are a type of classification technique that works well to classy linearly separable data with few samples. The above is done by generating a hyperplane between the labeled classes that best separates the classes. Support Vector Machines can also handle non linearly separable data. To do the above this method used a function commonly known as the kernel, to map the data points into a higher dimension space where the labeled data can be linearly separated. Similarly, as K-NN, 5-fold cross-validation was used as the resampling method for the data of the EOG features. The metrics results by performing a support vector machine classifier can be seen in Table \ref{604746}. \cite{Faul_2019}