Dynamic Analysis Based Mobile Malware Classification Using Supervised Machine Learning Methods


ISBN 978-93-5547-441-4 (Print)
ISBN 978-93-5547-442-1 (eBook)
DOI: 10.9734/bpi/mono/978-93-5547-441-4

 

Android based mobile malware are of great threat to the mobile users as they compromise user’s credentials without the knowledge of the user. Detection of Mobile malware and stopping their spread will save the system from further damage and misuse. Though many automated tools are available, sometimes it is very challenging to predict as malware handling methods need intelligent mechanisms. Artificial Intelligence is the promising domain which may ensure better handling of threats. Machine Learning is an important branch of AI which is expending in handling threats in an intelligent way. Intelligent intrusion detection systems are the need of the hour and machine learning methods are the most sought after and explored methods. This book titled, “Dynamic Analysis based Mobile Malware Classification using Supervised Machine Learning Methods”, discusses most of the promising Supervised Machine learning models and how they are applied for classifying Mobile Malware. This book starts from the fundamentals and gradually deals with the methodology and implementation of popular Supervised Machine Learning methods for Mobile Malware Detection. The most ideal platform Python is used, and the step-by-step methodology is presented during every phase of the implementation process. Important metrics are used to validate the performance of the ML methods and the most desirable model is recommended. Finding the most suitable model is an important phase in automating the entire process of Mobile Malware Detection. It is an ideal material for students, researchers, security professional and others who aspire to take up a serious career for the backend processing. I hope the book will be useful to the readers and many will find interesting components to explore further. My best wishes and prayers for all the readers for a great career.


Media Promotion

Content

Dynamic Analysis Based Mobile Malware Classification Using Supervised Machine Learning Methods

Padmavathi Ganapathi, D. Shanmugapriya, A. Roshni

Dynamic Analysis Based Mobile Malware Classification Using Supervised Machine Learning Methods, 20 January 2022, Page 1-168
https://doi.org/10.9734/bpi/mono/978-93-5547-441-4

Due to the rapid growth of android applications and mobile users in this technological era, there is a large increase in cyber attacks through mobile phones. During Pandemic period, mobile malware attacks are one of the top most cyber attacks observed in android mobile users to steal the user personal credentials by intrusion of adware, spyware, banking malware, SMS malware, riskware, viruses, Trojan horse, worms, keylogger and many more. Machine learning methods are very useful and amicable to detect mobile malwares. Automation of the mobile malware detection is the need of the hour and it is imperative to identify the most suitable machine learning techniques. This book entitled “Dynamic Analysis based Mobile Malware Classification using Supervised Machine Learning Methods” investigates the evaluation of supervised machine learning algorithms that are applied to detect and classify the mobile malwares. A systematic method of evaluation of supervised machine learning model to detect the malware data points and to classify them into binary classification as malware or benign is essential. The purpose of evaluating the supervised machine learning algorithms is to identify the best supervised machine learning model for mobile malware detection with high efficacy rate.  All important performance measures like Precision, Recall, F1 score, R2 score, Mean Absolute Error, Mean Squared Error and Root Mean Squared Error are applied and the entire experiments are conducted using benchmark dataset taken from kaggle community. Nine Supervised Machine Learning methods such as Decision Tree, Random Forest, K-Nearest Neighbor, Support Vector Machine, Naïve Bayes, AdaBoost, Multi-layer Perceptron, Logistic Regression and Linear Discriminant Analysis are experimented and the results are discussed. +