Evaluating the Performance of Data Level Methods Using KEEL Tool to Address Class Imbalance Problem

Upadhyay, Kamlesh J.; Kaur, Prabhjot; Verma, Deepak Kumar

doi:10.1007/s13369-021-06377-x

Cited by 14 publications

(15 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ROS technique is a data-level approach aimed at addressing the issue of imbalanced data by increasing the number of minority classes. This is accomplished by randomly replicating instances to balance the majority classes [10]. The oversampling technique is also implemented by examining the training data for one class, with a similar probability assigned to both Y0 and Y1.…”

Section: Random Over Sampling (Ros)mentioning

confidence: 99%

“…Among these approaches, the data-level approach is the most commonly used method to address imbalanced information, due to its advantage of increasing data validity and reducing training errors [9]. The data-level approach can also be advanced through the use of hybrid sampling, which balances the data and reduces noise by combining oversampling and undersampling [10].…”

Section: Introductionmentioning

confidence: 99%

“…In light of the presented approaches, it can be concluded that MCMTD, SMOTE-RkNN, HybridDA, and CSMOUTE are particularly suitable for addressing imbalanced data classification problems in the context of binary classification with numerical data types [10]. However, the MC-CCR approach is designed for multi-class classification, although it has only been tested with numerical data types.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Hybrid Sampling Approach for Improving the Classification of Imbalanced Data Using ROS and NCL Methods

2023

IJIES

View full text Add to dashboard Cite

This research presents a novel hybrid sampling technique, implemented at the data level, to effectively address imbalanced and noisy data in classification processes. The proposed technique expertly combines two established methods, namely, the random over sampling (ROS) and neighbourhood cleaning rule (NCL) approaches, to tackle imbalance and noise issues, respectively. The study carried out an empirical evaluation of the proposed approach using crowdsourced text data that primarily emphasized the triple bottom line (TBL) dimension of a smart social, economic, and environmental city. The study used the long short-term memory (LSTM), convolutional neural networks (CNN), and CNN-LSTM classification models to validate the efficacy of the proposed hybrid sampling technique and compare its performance with other existing approaches, including ROS oversampling, NCL undersampling, synthetic minority over sampling & tomek links (SMOTE-Tomek), and synthetic minority oversampling and edited nearest neighbours (SMOTE-ENN) hybrid sampling. The results are impressive, with the ROS-NCL hybrid sampling technique achieving high accuracy rates across all three classification models, at 97.71%, 98.01%, and 98.11%, respectively. This approach provides a robust and effective solution for handling impure data and holds great promise in identifying complex data patterns in real-world classification problems.

show abstract

Section: Random Over Sampling (Ros)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Hybrid Sampling Approach for Improving the Classification of Imbalanced Data Using ROS and NCL Methods

2023

IJIES

View full text Add to dashboard Cite

show abstract

“…The objective is to minimize the overall cost of the training data set. Since they depend on numerous factors, cost values are challenging to determine [15,18,19,21,25,26].…”

Section: Introductionmentioning

confidence: 99%

“…In the hybrid approach, the data level and the algorithm level are added together [15,18,19,21,25,26].…”

Section: Introductionmentioning

confidence: 99%

A Ranged-based Association Rule and GAN-based Hybrid Approach for Imbalanced Classification

Ahmed

Das

2023

Preprint

View full text Add to dashboard Cite

Machine learning has been extensively used in the field of automation systems, and in machine learning, imbalanced data is a prevalent word in fact it is a challenging element to deal with. How to deal with this imbalanced data is a major focus for the majority of studies. In terms of balancing the data, at the data level point under-sampling, over-sampling, and their variants are widely used. Since over-sampling creates precise replicas of examples from the minority class, it may increase the risk of over-fitting. Under-sampling wipes out a significant quantity of data, making it more difficult to determine where the decision boundary between minority and majority classes lies. In this work, a novel method has been proposed that combines both under-sampling and over-sampling strategies based on the Heuristic Range-Basd Association rule and a modified Tabular Generative Adversarial Network (TGAN) known as the Ranged-based Association rule and GAN-based Hybrid (RAMGANH) to avoid those kinds of problem scores and produce a well-balanced data set. The proposed approach has been tested by using existing standard classifiers with a few standard data sets, and the results demonstrate appreciable improvements in the classifier performance than the other state of-art-method.

show abstract

Credit scoring prediction leveraging interpretable ensemble learning

Liu,

Huang,

et al. 2023

Journal of Forecasting

View full text Add to dashboard Cite

Credit scoring models based on machine learning often need to work on accuracy and interpretability in practical applications. Original KCDWU has a more prominent adaptive property but ignores intra‐class and inter‐class distances in the clustering process, resulting in the possibility of inaccurate identification of class features and cluster structure of data, which compromises the clustering effect. Therefore, we improve the automatic K‐means clustering based on the Calinski–Harabasz index, thus achieving a clustering output for improved results. We also scrutinize representative five single classification models and six ensemble learning models for credit scoring prediction. We empirically test the superior performance of ensemble learning models and identify the best model CatBoost by comparing them based on multiple evaluation indicators. Empirical results reveal that the SHAP method conforms well to CatBoost and delivers a global and local interpretation of the predictions. This work provides financial institutions with a promising candidate for interpretable credit scoring models.

show abstract

Evaluating the Performance of Data Level Methods Using KEEL Tool to Address Class Imbalance Problem

Cited by 14 publications

References 55 publications

A Hybrid Sampling Approach for Improving the Classification of Imbalanced Data Using ROS and NCL Methods

A Hybrid Sampling Approach for Improving the Classification of Imbalanced Data Using ROS and NCL Methods

A Ranged-based Association Rule and GAN-based Hybrid Approach for Imbalanced Classification

Credit scoring prediction leveraging interpretable ensemble learning

Contact Info

Product

Resources

About