Performance Analysis of SMOTE-based Oversampling Techniques When Dealing with Data Imbalance

Bajer, Dražen; Zonc, Bruno; Dudjak, Mario; Martinović, Goran

doi:10.1109/iwssip.2019.8787306

Cited by 23 publications

(20 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…1. Outline of SMOTE [7] While developing a synthetic instance creation mechanism, the authors of SMOTE focused on feature space, rather than data space, to make the algorithm appropriate for a general imbalance problem [5]. Therefore, the synthetic instances are created along the line segments joining selected class neighbors.…”

Section: Synthetic Minority Oversampling Techniquementioning

confidence: 99%

“…The experiment setup is similar to that proposed in [7], in terms of used datasets and performance evaluation method, but different metrics are compared and different classifiers are considered. The datasets were obtained from the Keel [22] and the UCI [23] machine learning repositories and represent real-world problems with various imbalanced ratios to support the comparison.…”

Section: Experimental Analysismentioning

confidence: 99%

“…In order to observe which oversampling algorithms operate best with different types of classifiers, we selected the k-nearest neighbors with k = 5 (5-NN), multi-layer perceptron (MLP), decision tree (DT) and support vector machine (SVM) algorithms. These classifiers are often used in the literature to compare oversampling techniques [7,8]. The classifiers were implemented using Scikit-learn library [24], leaving default settings for each, to avoid the impact of manually selected parameters.…”

Section: Experimental Analysismentioning

confidence: 99%

“…Considering the importance of achieving quality performance for all classes in imbalanced problems, these metrics are combined into more comprehensive metrics such as AUC [27], G-mean [28] and F-measure [29]. The latter has been widely used in the evaluation of oversampling algorithms in the literature [3,7,30], as a balanced F β -score with a β value of 1…”

Section: Performance Metrics For Imbalanced Problemsmentioning

confidence: 99%

“…The latest such studies have begun to place more emphasis on experimental analysis than on plain review. In analysis performed by Bajer et al [7], six SMOTE-based algorithms are statistically compared in terms of the obtained geometric mean across large number of diverse imbalanced datasets and three different classifiers. An interesting conclusion of this analysis is that even a slight change in the synthetic instance creation method can alter the performance of the overall algorithm.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

In-Depth Performance Analysis of SMOTE-Based Oversampling Algorithms in Binary Classification

Dudjak¹,

Martinović²

2020

Int. j. electr. comput. eng. syst. (Online)

Self Cite

View full text Add to dashboard Cite

In the field of machine learning, the problem of class imbalance considerably impairs the performance of classification algorithms. Various techniques have been proposed that seek to mitigate classifier bias with respect to the majority class, with simple oversampling approaches being one of the most effective. Their main representative is the well-known SMOTE algorithm, which introduces a synthetic instances creation mechanism as an interpolation procedure between minority instances. To date, an abundance of SMOTE-based extensions that intend to improve the original algorithm has been proposed. This paper aims to compare the performance of several such extensions. In addition to comparing the overall performance, the impact of the selected oversamplers on the per-class performance is also evaluated. Finally, this paper tries to interpret the obtained performance results with respect to the internal procedures of oversampling algorithms. Some interesting findings have been made in this regard.

show abstract

Section: Synthetic Minority Oversampling Techniquementioning

confidence: 99%

Section: Experimental Analysismentioning

confidence: 99%

Section: Experimental Analysismentioning

confidence: 99%

Section: Performance Metrics For Imbalanced Problemsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

In-Depth Performance Analysis of SMOTE-Based Oversampling Algorithms in Binary Classification

Dudjak¹,

Martinović²

2020

Int. j. electr. comput. eng. syst. (Online)

Self Cite

View full text Add to dashboard Cite

show abstract

Examining the classification performance of pre‐trained capsule networks on imbalanced bone marrow cell dataset

Aydin Atasoy,

Faris Abdulla Al Rahhawi

2024

Int J Imaging Syst Tech

View full text Add to dashboard Cite

The automatic detection of bone marrow (BM) cell diseases plays a vital role in the medical field; it helps to make diagnoses more precise and effective, which leads to early detection and can significantly improve patient outcomes and increase the chances of successful intervention. This study proposed a fully automated intelligent system for BM classification by developing and enhancing Capsule Neural Network (CapsNet) architecture. Although CapsNet has demonstrated success in many classification tasks, it still has some limitations and challenges associated with using Convolutional Neural Networks (CNNs), which suffer from information loss during the pooling and discarding of detailed spatial information, resulting in the loss of fine‐grained features. Additionally, CNNs must help capture hierarchical feature relationships and often learn them implicitly by stacking convolutional layers. In contrast, CapsNets are designed to capture hierarchical features through dynamic routing and relationships between capsules, resulting in a more explicit representation of spatial hierarchy. CapsNets manage transformations and offer equivariance, preserving spatial information through capsule routing mechanisms. Further, to improve how features are represented, pre‐trained models such as Residual Capsule Network (RES‐CapsNet), Visual Geometry Group Capsule Network (VGG‐CapsNet), and Google Network (Inception V3) (GN‐CapsNet) have been used. This helps the network obtain the low‐ and mid‐level features and information it has previously learned so that subsequent capsule layers receive better initial information. Additionally, the Synthetic Minority Over‐Sampling Technique (SMOTE) was implemented to mitigate class imbalance. It generates synthetic samples in feature space by over‐sampling the minority class, leading to improving model performance in accurately classifying rare instances. Fine‐tuning the hyperparameters and implementing these improvements resulted in remarkable accuracy rates on a large BM dataset, with reduced training time and trainable parameters. CapsNet achieved 96.99%, VGG‐CapsNet achieved 98.95%, RES‐CapsNet achieved 99.24%, and the GN‐CapsNet model demonstrated superior accuracy at 99.45%. GN‐Caps Net was the best because it requires a small number of epochs and has an effective deep inception architecture that efficiently extracts features at different scales to form a robust representation of the input. Our proposed models were compared with existing state‐of‐the‐art models using the BM dataset; the results showed that our models outperformed the existing approaches and demonstrated excellent performance. Further, this automated system can analyze large amounts of data and complex cells in images of the BM dataset. Thus, it gives healthcare professionals a detailed understanding of different diseases, which may take time to achieve manually.

show abstract

KSMOTEEN: A Cluster Based Hybrid Sampling Model for Imbalance Class Data

Dhamal,

Mehrotra

2023

Lecture Notes in Networks and Systems

View full text Add to dashboard Cite

Performance Analysis of SMOTE-based Oversampling Techniques When Dealing with Data Imbalance

Cited by 23 publications

References 25 publications

In-Depth Performance Analysis of SMOTE-Based Oversampling Algorithms in Binary Classification

In-Depth Performance Analysis of SMOTE-Based Oversampling Algorithms in Binary Classification

Examining the classification performance of pre‐trained capsule networks on imbalanced bone marrow cell dataset

KSMOTEEN: A Cluster Based Hybrid Sampling Model for Imbalance Class Data

Contact Info

Product

Resources

About