Data mining of encrypted network traffic for adult content and gambling Android applications

Abstract

With the advancement of mobile computing technology, there has been a rampant increase in mobile-based pornography, gambling, and scams in the recent past, leading to deception and loss of personal property, resulting in serious societal and legal issues. Therefore, we decided to investigate adult and gambling applications, discover, and analyze their encrypted traffic characteristics. We crawled a massive database of malicious domains and automatically downloaded data, including 687 gambling applications and 334 adult Android applications This paper presents an empirical research study based on the real traffic of mobile pornography and gambling applications, delving into data mining and analyzing their traffic characteristics at the packet and session flow levels. As these applications involve the dissemination of adult content and scams, running them on an emulator and collecting traffic files does not constitute a violation of privacy. Through quantitative analysis of the encrypted traffic features of such applications, this research aims to provide assistance in the identification and classification of adult and gambling applications and related tasks.