Search

Mitigating the impact of class imbalance datasets on classifiers poses a challenge to the machine learning community. Conventional classifiers do not perform well as they are habitually biased toward the majority class. Among existing solutions, the synthetic minority oversampling technique (SMOTE) has shown great potential, aiming to improve the dataset rather than the classifier. However, SMOTE still needs improvement because of its equal oversampling to each minority instance. Based on the consensus that instances far from the borderline contribute less to classification, a refined method for oversampling borderline minority instances (OBMI) is proposed in this paper using a two-stage Tomek link-finding procedure. In the oversampling stage, the pairs of between-class instances nearest to each other are first found to form Tomek links. Then, these minority instances in Tomek links are extracted as base instances. Finally, new minority instances are generated, each of which is linearly interpolated between a base instance and one minority neighbor of the base instance. To address the overlap caused by oversampling, in the cleaning stage, Tomek links are employed again to remove the borderline instances from both classes. The OBMI is compared with ten baseline methods on 17 benchmark datasets. The results show that it performs better on most of the selected datasets in terms of the F1-score and G-mean. Statistical analysis also indicates its higher-level Friedman ranking. [ABSTRACT FROM AUTHOR]

Abstract Mitigating the impact of class imbalance datasets on classifiers poses a challenge to the machine learning community. Conventional classifiers do not perform well as they are habitually biased toward the majority class. Among existing solutions, the synthetic minority oversampling technique (SMOTE) has shown great potential, aiming to improve the dataset rather than the classifier. However, SMOTE still needs improvement because of its equal oversampling to each minority instance. Based on the consensus that instances far from the borderline contribute less to classification, a refined method for oversampling borderline minority instances (OBMI) is proposed in this paper using a two-stage Tomek link-finding procedure. In the oversampling stage, the pairs of between-class instances nearest to each other are first found to form Tomek links. Then, these minority instances in Tomek links are extracted as base instances. Finally, new minority instances are generated, each of which is linearly interpolated between a base instance and one minority neighbor of the base instance. To address the overlap caused by oversampling, in the cleaning stage, Tomek links are employed again to remove the borderline instances from both classes. The OBMI is compared with ten baseline methods on 17 benchmark datasets. The results show that it performs better on most of the selected datasets in terms of the F1-score and G-mean. Statistical analysis also indicates its higher-level Friedman ranking.

Imbalanced data classification, as a challenging task, has drawn a significant interest in numerous scientific areas. One popular strategy to balance the instance quantities between two classes is oversampling via generating synthetic instances. However, it still suffers from two key issues: where and how many synthetic instances should be generated. In this paper, we propose a L ocal distribution-based A daptive M inority O versampling method (LAMO) to deal with the imbalance classification problem. LAMO first identifies the informative borderline minority instances as sampling seeds according to their neighbors and the corresponding class distribution. Then, LAMO captures the local distribution of each seed according to its Euclidean distances from the nearest majority instance and nearest minority instance.Finally, LAMO generates synthetic instances around seeds via a Gaussian Mixture Model (GMM). For each component of GMM, the mixing coefficient and bandwidth are adaptively set with the aid of seeds' local distribution. Extensive experiments have been conducted on both simulated and real data sets under varying the imbalance ratio and data size. By comparing with the state-of-the-art oversampling methods, the proposed LAMO obtains promising results in terms of several widely used evaluation metrics. [ABSTRACT FROM AUTHOR]

Learning class-imbalance data has become a challenging task in machine learning. Oversampling is an effective way to achieve rebalancing between classes by generating new minority instances. However, most existing oversampling methods exhibit significant ill-posedness due to the involvement of k -nearest neighbors. Recently, several novel techniques have been developed using natural neighbors instead of k -nearest neighbors. But these budding techniques fail to take into account the natural distribution of data, resulting in degraded adaptability of natural neighbors to data characteristics. In this paper, we propose an adaptive and completely parameter-free synthetic oversampling method called NanBDOS (short for borderline oversampling via natural neighbor search). NanBDOS first determines the intrinsic neighbor relationship between instances through a natural neighbor search procedure. Then, those informative borderline instances are identified and used as base instances, i.e., sampling seeds. NanBDOS assigns dynamic sampling weights to the base instances so that the data complexity can be well represented. Finally, new synthetic instances are generated by interpolating between the base instances and their natural neighbors. NanBDOS is experimentally compared with seven baseline methods on twenty-four real-world datasets. The results confirm the effectiveness of the proposed method. Moreover, the statistical analysis also indicates its higher-level Friedman ranking. [ABSTRACT FROM AUTHOR]