Multi-Class Support Vector Machines forLarge DataSetsviaMinimumEnclosing BallClustering

for the second stage SVM Then we extend binary SVM classification tocase ofmulticlass. The noveltwo-stage multi-class SVM hasdistinctive advantages on dealing withhugedatasets.Finally, we apply theproposed method on severalbenchmarkproblems, experimen talresults demonstratethatour approachhavefgrood classification accuracy whilethetraining issignificantly fasterthanotherSVM classifiers.