Reports Outline Support Vector Machines Study Results from Nanjing Normal University

摘要

According to the authors of a study from Jiangsu, People's Republic of China, "Existing multi-labelnsupport vector machine (Rank-SVM) has an extremely high computational complexity and lacks annintrinsic zero point to determine relevant labels. In this paper, we propose a novel support vector machinenfor multi-label classification through both simplifying Rank-SVM and adding a zero label, resulting into anquadratic programming problem in which each class has an independent equality constraint."