Chance constrained conic-segmentation support vector machine with uncertain data

Support vector machines (SVM) is one of the well known supervised machine learning model. The standard SVM models are dealing with situation where exact values data points known. This paper studies model when set contains uncertain or mislabelled points. To ensure small probability misclassification for data, a chance constrained conic-segmentation proposed multiclass classification. Based on set, mixed integer programming formulation derived. Kernelization also exploited nonlinear geometric interpretation presented to show how works data. Finally, experimental results demonstrate effectiveness both artificial and real-world