Multiclass Support Vector Machines for Articulatory Feature Classification

of somewhat abstracting away from the literal physiological measurements of articulation that are so closely tied to the acoustic signal, and with some additional computational burden. The SVM is a binary classifier which has demonstrated excellent performance on a variety of classification tasks (Bennett and Campbell 2000). The learned decision boundary corresponds to the optimal separating hyperplane, which maximizes the margin between two sets of linearly separable vectors. Several variations of the SVM, designed to classify linearly inseparable data, are discussed in (Sch

• Type: Conference
• Year: 2006
• Where: AAAI
• Authors: Brian Hutchinson, Jianna Zhang
• Keywords: AAAI 2006, Additional Computational Burden, Intelligent Agents, Literal Physiological Measurements, Optimal Separating Hyperplane