Constructing classifiers for imbalanced data using diversity optimisation

Khorshidi, Hadi Akbarzadeh; Aickelin, Uwe

doi:10.1016/j.ins.2021.02.069

Cited by 14 publications

(7 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bellinger et al [13] proposed ReMix to seamlessly utilize batch resampling to induce robust deep models from imbalanced and long-tailed dataset. Khorshidi et al [14] proposed adapting diversity optimization through DIWO and DADO to generate different synthetic instances close to instances in the minority class, but failed to apply extended diversity optimization more effectively in generating synthetic instances for multi-class dataset. By contrast, we perform data augmentation to expand data, and improve loss function to further alleviate class imbalanced problem.…”

Section: Class-imbalancedmentioning

confidence: 99%

Imbalance multiclass Problem: A robust Feature Enhancement-based Framework for Liver Lesion Classification

Rui

Song

Liu

et al. 2023

Preprint

View full text Add to dashboard Cite

The classification of liver lesions in CT images is essential for the diagnosis and treatment of liver diseases. Since the characteristics of different classes of lesions are similar and the degree of differentiation is not obvious, it is difficult to accurately classify different classes of liver lesions, especially imbalanced multiclass distribution. To this end, we propose a novel feature enhancement-based framework for imbalanced liver lesion classification. Specifically, a liver lesion processing method is introduced to expand data based on Augmentor. Besides, dual feature enhancement based on feature refinement extraction and feature global correlation is designed to enhance feature extraction ability. To further alleviate the problem of class imbalance, the improved loss function based on standard Cross-Entropy (CE) is also adopted to make the network pay more attention to the class with fewer samples. Experiments on 551 liver lesions in 120 patients showed that: 1) Our proposed framework improved the classification performance of liver tumors under imbalanced multiclass data distribution; 2) The dual feature enhancement was lightweight and efficient which enhanced the semantic expression of overall features without introducing additional parameters.

show abstract

Section: Class-imbalancedmentioning

confidence: 99%

Imbalance multiclass Problem: A robust Feature Enhancement-based Framework for Liver Lesion Classification

Rui

Song

Liu

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Most recently, Diversity-based Average Distance Oversampling (DADO) and Diversity-based Instance-Wise Oversampling (DIWO) have been proposed to promote diversity [16]. The objective of the two techniques is to generate welldiverse synthetic instances close to minority class instances.…”

Section: Related Workmentioning

confidence: 99%

“…The distance measures chosen for both objective function and diversity measure are the optimal distance measure based on experimental results [16]. Euclidean distance measure ( D Eu ) is chosen for DADO, and Canberra ( D c ) is chosen for DIWO.…”

Section: Parameter Selectionmentioning

confidence: 99%

A Diversity-Based Synthetic Oversampling Using Clustering for Handling Extreme Imbalance

Yang,

Akbarzadeh Khorshidi,

Aickelin

2023

SN COMPUT. SCI.

Self Cite

View full text Add to dashboard Cite

Imbalanced data are typically observed in many real-life classification problems. However, mainstream machine learning algorithms are mostly designed with the underlying assumption of a relatively well-balanced distribution of classes. The mismatch between reality and algorithm assumption results in a deterioration of classification performance. One form of approach to address this problem is through re-sampling methods, although its effectiveness is limited; most re-sampling methods fail to consider the distribution of minority and majority instances and the diversity within synthetically generated data. Diversity becomes increasingly important when minority data becomes more sparse, as each data point becomes more valuable. They should all be considered during the generation process instead of being regarded as noise. In this paper, we propose a cluster-based diversity re-sampling method, combined with NOAH algorithm. Neighbourhood-based Clustering Diversity Over-sampling (NBCDO) is introduced with the aim to complement our previous cluster-based diversity algorithm Density-based Clustering Diversity Over-sampling (DBCDO). It first uses a neighbourhood-based clustering algorithm to consider the distribution of both minority and majority class instances, before applying NOAH algorithm to encourage diversity optimisation during the generation of synthetic instances. We demonstrate the implementation of both cluster-based diversity methods by conducting experiments over 10 real-life datasets with ≤ 5% imbalance ratio and show that our proposed cluster-based diversity algorithm (NBCDO, DBCDO) brings performance improvements over its comparable methods (DB-SMOTE, MAHAKIL, KMEANS-SMOTE, MC-SMOTE).

show abstract

“…The undersampling method balances dataset by partially selecting the majority class dataset. Oversampling method is to artificially generate data to minority class to achieve a balance dataset 12 . Both methods are effective to imbalance classification.…”

Section: Introductionmentioning

confidence: 99%

“…Oversampling method is to artificially generate data to minority class to achieve a balance dataset. 12 Both methods are effective to imbalance classification.…”

Section: Introductionmentioning

confidence: 99%

Undersampling of approaching the classification boundary for imbalance problem

Jiang

Yuan

Liao

et al. 2022

Concurrency and Computation

View full text Add to dashboard Cite

Using imbalanced data in classification affect the accuracy. If the classification is based on imbalanced data directly, the results will have large deviations. A common approach to dealing with imbalanced data is to re-structure the raw dataset via undersampling method. The undersampling method usually uses random or clustering approaches to trimming the majority class in the dataset, since some data in the majority class makes not contribute to classification model. In this paper a revised undersampling approach is proposed. First, we perform space compression in the vertical direction of the separating hyperplane. Then, a weighted random sampling hybrid ensemble learning method is carried out to make the sampled objects spread more widely near the separating hyperplane. Experiments with 7 under-sampling methods on 21 imbalanced datasets show that our method has achieved good results.

show abstract

Constructing classifiers for imbalanced data using diversity optimisation

Cited by 14 publications

References 31 publications

Imbalance multiclass Problem: A robust Feature Enhancement-based Framework for Liver Lesion Classification

Imbalance multiclass Problem: A robust Feature Enhancement-based Framework for Liver Lesion Classification

A Diversity-Based Synthetic Oversampling Using Clustering for Handling Extreme Imbalance

Undersampling of approaching the classification boundary for imbalance problem

Contact Info

Product

Resources

About