A hybrid under-sampling approach for mining unbalanced datasets: applications to banking and insurance

摘要: In solving unbalanced classification problems, machine learning algorithms are overwhelmed by the majority class and consequently misclassify minority observations. Here, we propose a hybrid under-sampling approach to improve performance of classifiers. The proposed first employs k -reverse nearest neighbour (kRNN) method detect outliers from class. After removing outliers, using K-means clustering, K-clusters selected further reduce influence Then, employed support vector (SVM), logistic regression (LR), multi layer perceptron (MLP), radial basis function network (RBF), group data handling (GMDH), genetic programming (GP) decision tree (J48) for purpose. effectiveness was demonstrated on datasets taken insurance fraud detection credit card churn in banking domain. Ten-fold cross validation used study. It is observed that improved