A Simple and Fast Multi-instance Classification via Support Vector Machine

Abstract

In this paper, we proposed a Simple and Fast Multi-Instance Classification Via Support Vector Machine(called Fast MI-SVM). Compared with the other conventional Multi-Instance learning method, our method is able to deal with multi-instance learning problem by only solving a quadratic programming problem. So the training time of Fast MI-SVM is very fast. All numerical experiments on benchmark datasets show the feasibility and validity of the proposed method.