A study on classifying imbalanced datasets

Lakshmi, T. Jaya; Prasad, CH. Vaikunta Krishna

doi:10.1109/cnsc.2014.6906652

Cited by 20 publications

(10 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As for TNR, it also gets high scores with most of the thresholds, and the highest score is around 0.9998 achieved by θ = 0.25. However, for the same reason as the ACC metric, because datasets of fraud detection problems are usually highly imbalanced, it tends to obtain a much higher TN value than FP value, which easily results in a high TNR [37]. Similarly, the threshold is then tuned to obtain the best ACC, TPR, TNR, and MCC, as demonstrated in Figure 6.…”

Section: Performance Evaluation Of Af-prfmentioning

confidence: 99%

Credit Card Fraud Detection with Autoencoder and Probabilistic Random Forest

Lin

Jiang

2021

Mathematics

View full text Add to dashboard Cite

This paper proposes a method, called autoencoder with probabilistic random forest (AEPRF), for detecting credit card frauds. The proposed AE-PRF method first utilizes the autoencoder to extract features of low-dimensionality from credit card transaction data features of high-dimensionality. It then relies on the random forest, an ensemble learning mechanism using the bootstrap aggregating (bagging) concept, with probabilistic classification to classify data as fraudulent or normal. The credit card fraud detection (CCFD) dataset is applied to AE-PRF for performance evaluation and comparison. The CCFD dataset contains large numbers of credit card transactions of European cardholders; it is highly imbalanced since its normal transactions far outnumber fraudulent transactions. Data resampling schemes like the synthetic minority oversampling technique (SMOTE), adaptive synthetic (ADASYN), and Tomek link (T-Link) are applied to the CCFD dataset to balance the numbers of normal and fraudulent transactions for improving AE-PRF performance. Experimental results show that the performance of AE-PRF does not vary much whether resampling schemes are applied to the dataset or not. This indicates that AE-PRF is naturally suitable for dealing with imbalanced datasets. When compared with related methods, AE-PRF has relatively excellent performance in terms of accuracy, the true positive rate, the true negative rate, the Matthews correlation coefficient, and the area under the receiver operating characteristic curve.

show abstract

Section: Performance Evaluation Of Af-prfmentioning

confidence: 99%

Credit Card Fraud Detection with Autoencoder and Probabilistic Random Forest

Lin

Jiang

2021

Mathematics

View full text Add to dashboard Cite

show abstract

“…In real-world contexts, activity periods might vary greatly, and there can be a significant class imbalance, which can impede the learning process of machine learning algorithms. Numerous datalevel and algorithmic-level solutions are adopted to solve it [21]. One of the most popular approaches is to resample data and generate synthetic data points for the minority classes [22]- [24].…”

Section: Related Workmentioning

confidence: 99%

Nurse Care Activity Recognition: A Cost-Sensitive Ensemble Approach to Handle Imbalanced Class Problem in the Wild

Rahman

Hassan

Ahad

2021

Adjunct Proceedings of the 2021 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of The

View full text Add to dashboard Cite

Nurse care activity recognition is a recent but demanding study topic in human activity recognition (HAR) since it has high class imbalance and intra-class variability problem depending on both the patient and the receiver. Although traditional imbalance learning approaches are offered to address this issue, they have several limitations: 1) important information is lost while using undersampling approaches, and 2) oversampling methods shifts the class distribution, exposing the model to overfitting and over-optimism. To tackle this problem, we present a framework for a cost-sensitive hybrid optimum ensemble classifier. Our technique modified the class weights to impose higher misclassification costs on minority classes. Then we combined different weighted base classifier outputs with a stacked generalization method. This technique allowed us to leverage the skills of a varied collection of high-performing base learners while balancing each other's limitations and strengths. By utilizing our hybrid technique, we achieved 70.8% average cross-validation balanced accuracy in classifying 28 activities conducted by nurses in real-life settings. We developed this algorithmic pipeline for "The 3rd Nurse Care Activity Recognition Challenge" at ACM UbiComp 2021.

show abstract

“…As an extreme case in imbalance dataset, even if all the samples in the minor class are mispredicted, as long as most samples of the major class are predicted correctly, a high accuracy can still be achieved because of the proportion of the major class [22]. Therefore, we conclude that class imbalance plays a vital role in the classification process [23]- [25], and the accuracy rate alone cannot sufficiently represent classification performance.…”

Section: Related Workmentioning

confidence: 99%

DMP_MI: An Effective Diabetes Mellitus Classification Algorithm on Imbalanced Data With Missing Values

et al. 2019

View full text Add to dashboard Cite

As a widely known chronic disease, diabetes mellitus is called a silent killer. It makes the body produce less insulin and causes increased blood sugar, which leads to many complications and affects the normal functioning of various organs, such as eyes, kidneys, and nerves. Although diabetes has attracted high attention in research, due to the existence of missing values and class imbalance in the data, the overall performance of diabetes classification using machine learning is relatively low. In this paper, we propose an effective Prediction algorithm for Diabetes Mellitus classification on Imbalanced data with Missing values (DMP_MI). First, the missing values are compensated by the Naïve Bayes (NB) method for data normalization. Then, an adaptive synthetic sampling method (ADASYN) is adopted to reduce the influence of class imbalance on the prediction performance. Finally, a random forest (RF) classifier is used to generate predictions and evaluated using comprehensive set of evaluation indicators. Experiments performed on Pima Indians diabetes dataset from the University of California at Irvine, Irvine (UCI) Repository, have demonstrated the effectiveness and superiority of our proposed DMP_MI.

show abstract

A study on classifying imbalanced datasets

Cited by 20 publications

References 20 publications

Credit Card Fraud Detection with Autoencoder and Probabilistic Random Forest

Credit Card Fraud Detection with Autoencoder and Probabilistic Random Forest

Nurse Care Activity Recognition: A Cost-Sensitive Ensemble Approach to Handle Imbalanced Class Problem in the Wild

DMP_MI: An Effective Diabetes Mellitus Classification Algorithm on Imbalanced Data With Missing Values

Contact Info

Product

Resources

About