Deep Ensemble Classifier for Ransomware Identification Using Digitalized DNA Genotyping System
International journal of intelligent engineering and systems · 2022
Ransomware is a kind of virus that enciphers data on a victim's machine and only allows it to be decrypted when the user makes the payment.Understanding the strategies employed by cyber thieves is necessary to design effective protection against them.So, an active learning-based digital DNA sequencing engine called DNAact-Ran was developed to predict and identify ransomware data.But, it was not suitable for temporally changed DNA sequences.Hence, this article proposes a deep ensemble ransomware prediction (DeepERPred) model to handle the temporally changed DNA sequences and identify ransomware effectively.In this model, the raw data is converted into the required form.After that, the most relevant attributes are chosen from the preprocessed data using optimization algorithms.The selected attributes are classified by using the ensemble convolutional neural network and long short-term memory (CNN-LSTM) model.This ensemble classification is introduced to handle temporal changes in the DNA sequences by learning dependencies among present and previous occurrences of a sequence for identifying ransomware data.Finally, the investigational outcomes reveal that the DeepERPred model achieves a 90.67% of accuracy, whereas the classical models such as the modified decision tree (MDT), LightGBM, random forest (RF), artificial neural network (ANN), RansomDroid and DNAact-Ran attain 75.83%, 78.52%, 83.22%, 85.48%, 87.91% and 89.14%, respectively to identify ransomware.