Hybrid Classifier Ensemble for Imbalanced Data

Yang, Kaixiang; Yu, Zhiwen; Wen, Xin; Cao, Wenming; Chen, C. L. Philip; Wong, Hau-San; You, Jane

doi:10.1109/tnnls.2019.2920246

Cited by 82 publications

(20 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several informed over-sampling techniques have also been developed to reduce over-fitting and strength class boundaries. Over-sampling methods tend to be more efficient than under-sampling techniques when handling extremely imbalanced big data problem with large imbalanced ratio [12], [23]- [25]. Chawla et al [11] have introduced a Synthetic Minority Over-sampling Technique (SMOTE); a method for creating synthetic data of minority samples by identifying the feature space for the minority samples and considering their k nearest neighbours.…”

Section: A Data-driven Methodsmentioning

confidence: 99%

Learning With Imbalanced Data in Smart Manufacturing: A Comparative Analysis

2021

View full text Add to dashboard Cite

Section: A Data-driven Methodsmentioning

confidence: 99%

Learning With Imbalanced Data in Smart Manufacturing: A Comparative Analysis

2021

View full text Add to dashboard Cite

“…Zhu et al [31] proposed a geometric structural ensemble learning framework, which partitions and eliminates redundant majority samples by generating hyper-sphere through the Euclidean metric and learns basic classifiers to enclose the minority samples. Yang et al [32] proposed a hybrid ensemble classifier framework that combines density-based undersampling and cost-effective methods using multi-objective optimization algorithm to handle two issues: (1) undersampling methods suffer from losing important information; (2) cost-sensitive methods are sensitive to outliers and noise.…”

Section: Related Workmentioning

confidence: 99%

Binary Imbalanced Data Classification Based on Modified D2GAN Oversampling and Classifier Fusion

Zhai

Zhang

2020

IEEE Access

View full text Add to dashboard Cite

Binary imbalance problem refers to such a classification scenario where one class contains a large number of samples while another class contains only a few samples. When traditional classifiers face with imbalanced datasets, they usually bias towards majority class resulting in poor classification performance. Oversampling is an effective method to address this problem, yet how to conduct diversity oversampling is a challenge. In this paper, we proposed a diversity oversampling method based on a modified D2GAN model, and on the basis of diversity oversampling, we also proposed a binary imbalanced data classification approach based on classifier fusion by fuzzy integral. Extensive experiments are conducted on 8 data sets to compare the proposed methods with 7 state-of-the-art methods on 5 aspects: MMD-score, Silhouette-score, F-measure, G-means, and AUC-area. The 7 methods include 4 SMOTE related approaches and 3 GAN related approaches. The experimental results demonstrate that the proposed methods are more effective and efficient than the compared approaches. INDEX TERMS Binary class imbalance, diversity oversampling, generative adversarial network, classifier fusion, fuzzy integral

show abstract

“…Machine learning and deep learning algorithms are strongly affected by the class imbalance problem [11][12][13][14][15]. The latter refers to some difficulties that appear when the number of samples in one or more classes in the dataset is fewer than another class (or classes), thereby producing an important deterioration of the classifier performance [16].…”

Section: Introductionmentioning

confidence: 99%

“…Another way to deal with the class imbalance problem has been through the Cost Sensitive (CS) approach [34], which has become an important topic in deep learning research in recent years [13][14][15]35]. CS considers the costs associated with misclassifying samples; i.e., it uses different cost matrices describing the costs of misclassifying any particular data sample [29].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Data Sampling Methods to Deal With the Big Data Multi-Class Imbalance Problem

et al. 2020

View full text Add to dashboard Cite

The class imbalance problem has been a hot topic in the machine learning community in recent years. Nowadays, in the time of big data and deep learning, this problem remains in force. Much work has been performed to deal to the class imbalance problem, the random sampling methods (over and under sampling) being the most widely employed approaches. Moreover, sophisticated sampling methods have been developed, including the Synthetic Minority Over-sampling Technique (SMOTE), and also they have been combined with cleaning techniques such as Editing Nearest Neighbor or Tomek's Links (SMOTE+ENN and SMOTE+TL, respectively). In the big data context, it is noticeable that the class imbalance problem has been addressed by adaptation of traditional techniques, relatively ignoring intelligent approaches. Thus, the capabilities and possibilities of heuristic sampling methods on deep learning neural networks in big data domain are analyzed in this work, and the cleaning strategies are particularly analyzed. This study is developed on big data, multi-class imbalanced datasets obtained from hyper-spectral remote sensing images. The effectiveness of a hybrid approach on these datasets is analyzed, in which the dataset is cleaned by SMOTE followed by the training of an Artificial Neural Network (ANN) with those data, while the neural network output noise is processed with ENN to eliminate output noise; after that, the ANN is trained again with the resultant dataset. Obtained results suggest that best classification outcome is achieved when the cleaning strategies are applied on an ANN output instead of input feature space only. Consequently, the need to consider the classifier's nature when the classical class imbalance approaches are adapted in deep learning and big data scenarios is clear.

show abstract

Hybrid Classifier Ensemble for Imbalanced Data

Cited by 82 publications

References 53 publications

Learning With Imbalanced Data in Smart Manufacturing: A Comparative Analysis

Learning With Imbalanced Data in Smart Manufacturing: A Comparative Analysis

Binary Imbalanced Data Classification Based on Modified D2GAN Oversampling and Classifier Fusion

Data Sampling Methods to Deal With the Big Data Multi-Class Imbalance Problem

Contact Info

Product

Resources

About