Table 3 Count of frequency of metadata elements of methodologies in papers studied in this survey. Includes classifier, evaluation metric, dataset, year, and baseline method
From: The use of generative adversarial networks to alleviate class imbalance in tabular data: a survey
Element | Value | Count | Element | Value | Count |
|---|---|---|---|---|---|
Classifier | CNN | 4 | Dataset | MNIST | 4 |
DT | 3 | CIFAR-10 | 3 | ||
KNN | 3 | GM Sim | 2 | ||
SVM | 3 | ISCX | 2 | ||
LR | 2 | Blog Catalog | 2 | ||
MLP | 2 | CelebA | 2 | ||
GBC | 2 | CICIDS 2017 | 2 | ||
RF | 2 | SVHN | 2 | ||
SAE | 1 | UCI Kaggle | 2 | ||
NB | 1 | Wikipedia | 2 | ||
Evaluation Metric | Balanced Accuracy | 13 | Year | 2015 | 1 |
F1 Score | 11 | 2017 | 5 | ||
Precision | 7 | 2018 | 3 | ||
Recall | 7 | 2019 | 8 | ||
AUC | 5 | 2020 | 2 | ||
Geometric Mean | 3 | Baseline Method | SMOTE | 10 | |
Ranking | 2 | no treatment | 8 | ||
Specificity | 2 | Oversampling | 4 | ||
SSIM | 2 | VAE | 4 | ||
Likelihood fitness metric | 1 | ADASYN | 3 | ||
FID | 1 | Kmeans-SMOTE | 2 | ||
Euclidean Distance | 1 | RBM | 2 |