TY - JOUR A2 - Melin, P. A2 - Pavón, J. AU - Li, Fengqi AU - Yu, Chuang AU - Yang, Nanhai AU - Xia, Feng AU - Li, Guangming AU - Kaveh-Yazdy, Fatemeh PY - 2013 DA - 2013/07/10 TI - Iterative Nearest Neighborhood Oversampling in Semisupervised Learning from Imbalanced Data SP - 875450 VL - 2013 AB - Transductive graph-based semisupervised learning methods usually build an undirected graph utilizing both labeled and unlabeled samples as vertices. Those methods propagate label information of labeled samples to neighbors through their edges in order to get the predicted labels of unlabeled samples. Most popular semi-supervised learning approaches are sensitive to initial label distribution which happened in imbalanced labeled datasets. The class boundary will be severely skewed by the majority classes in an imbalanced classification. In this paper, we proposed a simple and effective approach to alleviate the unfavorable influence of imbalance problem by iteratively selecting a few unlabeled samples and adding them into the minority classes to form a balanced labeled dataset for the learning methods afterwards. The experiments on UCI datasets and MNIST handwritten digits dataset showed that the proposed approach outperforms other existing state-of-art methods. SN - 2356-6140 UR - https://doi.org/10.1155/2013/875450 DO - 10.1155/2013/875450 JF - The Scientific World Journal PB - Hindawi Publishing Corporation KW - ER -