H. Manoj T. Gadiyar
Department of Computer Science and Business System, Canara Engineering College, Benjanapadavu - 574219
Pradeep G S
Assistant Professor, Department of Computer Science and Engineering, Sri Dharmasthala Manjunatheshwara Institute of Technology, Ujire-574240, Karnataka, India
Minal Moharir
Professor, Department of Computer Science and Engineering, R V College of Engineering, Bangalore-560074, Karnataka, India
Thyagaraju G S
Professor and Head, Department of Computer Science and Engineering, Sri Dharmasthala Manjunatheshwara Institute of
Abstract
In the current scenario, there is a significant rise in the amount as well as the sort of malicious mobile apps, notably on Android. This trend is expected to continue. Because of this, identifying counterfeit mobile apps has become a far more challenging task. Researchers have the capacity to discover potentially harmful programmes by doing an analysis of the data collected by the network. We are able to discover potentially hazardous programmes and behaviours taking place on the network by making use of machine learning and doing an in-depth study of the data pertaining to the network. Because dangerous programmes generate the vast majority of the time benign network traffic, modelling the harmless traffic leads to a problem with imbalanced data. This is because modelling the innocuous traffic leads to an issue. Unbalanced classification strategies, including SMOTE + SVM. SVM costsensitive (SVMCS), and C4.5 cost-sensitive, are all available to our organization’s users. As soon as the rate of imbalance reaches a particular degree of severity, the performance of most classification algorithms begins to decline. This phenomenon is known as the “balance threshold.” This system operates on the principle of machine learning. Users are given the option to analyse the detection performance of the different ways by using this prototype system and applying a variety of classification algorithms to the same data collection. This gives users the ability to compare the effectiveness of the various approaches.
Keywords- Network traffic, Malicious applications, Malware detection, Artificial Intelligence, ML Techniques