An Ensemble Method Based on SVC and Euclidean Distance for Classification Binary Imbalanced Data

In recent years, ensemble methods have been widely applied to classify binary imbalanced data. Traditional ensemble rules have manifested performance in dealing with imbalanced data. However, shortage appears that only the results of base classifiers is considered, while these traditional ensemble rules ignore the Euclidean distance between the new data and train data as well as the relations of majority and minority classes in the train data. So we proposed a novel ensemble rule which take Support Vector Classification (SVC) as base classifier. Moreover, the distance between the new data and train data and relations of majority classes and minority classes are taken into account to overcome conventional drawbacks. Simulation results are provided to confirm that the proposed method has better performance than existing ensemble methods.