Table 1 Related work highlights on different traditional models.
From: Prediction of android ransomware with deep learning model using hybrid cryptography
Ref | Year | Method used | Accuracy | Dataset |
|---|---|---|---|---|
Jabbar and Bhaya19 | 2023 | LR & SGD | 98% | UNSW-NB15 & BBC dataset |
Attou et al.21 | 2023 | RF | 98.3 | Bot-IoT |
Ahmad et al.22 | 2022 | NB, RF, KNN, and SVM | 92.0% | – |
Singh et al.24 | 2023 | ResNet50 & VGG 16 | 99.1% for testing | Ransomware attack dataset |
Omar et al.23 | 2022 | ESOML-IDS | Denoising Autoencoder 83.09% | UNSW-NB 15 |
Omar et al.25 | 2023 | OELSTM-MDC | 97.14% | – |
Omar et al.33 | 2021 | SVM HHO | 94% | CICmalanal2017 |
Alzubi et al.35 | 2024 | CNN and LSTM | Above 90% | CSE-CIC-IDS-2018 |
Movassagh et al.16 | 2023 | ANN | – | – |
