Mobile malware detection using a novel convolutional neural networks (CNN) algorithm compared with long short-term memory (LSTM) algorithm for improved accuracy

This research aims to enhance the accuracy of mobile malware detection using machine learning by evaluating the performance of a Novel Convolutional Neural Network (CNN) algorithm against a Long Short-Term Memory (LSTM) algorithm. The study employed both CNN and LSTM methods, with each being tested on a sample size of 20. The accuracy of mobile malware detection was assessed through multiple evaluations. The findings show that the CNN algorithm achieved a larger accuracy of 98.95%, significantly outperforming the LSTM algorithm, which had an accuracy of 64.45%. A sample t-test revealed a statistically significant difference between the two models, with a p-value of 0.001 (p