Class imbalance is a common challenge when dealing with pattern classification of real-world medical data-sets. An effective counter-measure typically used is a method known as re-sampling. In this paper we implement an ANN with different re-sampling techniques to subsequently compare and evaluate the performances. Re-sampling strategies included a control, under-sampling, over-sampling, and a combination of the two. We found that over-sampling and the combination of under- and over-sampling both led to a significantly superior classifier performance compared to under-sampling only in correctly predicting labelled classes.
CITATION STYLE
Saul, M. A., & Rostami, S. (2019). A comparison of re-sampling techniques for pattern classification in imbalanced data-sets. In Advances in Intelligent Systems and Computing (Vol. 840, pp. 240–251). Springer Verlag. https://doi.org/10.1007/978-3-319-97982-3_20
Mendeley helps you to discover research relevant for your work.