Cost-Sensitive SPFCNN Miner for Classification of Imbalanced Data

Abstract

Since the target data are high-dimensional, limited and class-unbalanced distribution in most real-world classification, most conventional classification methods can hardly achieve good classification results on these data. To explore an effective solution, this paper proposes the Siamese Parallel Fully-connected Neural Network (SPFCNN) as a binary classifier and uses the SMOTE method to deal with the problem of class-unbalanced data distribution. Given that classified cases naturally come with costs, cost-sensitive learning is used to improve the performance of the proposed SPFCNN. An extensive computational study is also performed on cost-sensitive SPFCNN, and the results show that the performance of the proposed approach is better than that of the comparison methods.