DOIONLINE

DOIONLINE NO - IJACEN-IRAJ-DOIONLINE-20805

Publish In
International Journal of Advance Computational Engineering and Networking (IJACEN)-IJACEN
Journal Home
Volume Issue
Issue
Volume-12,Issue-5  ( May, 2024 )
Paper Title
Enhancing Android Malware Detection through Adversarial Deep Learning Strategies
Author Name
Narendran G, Kaarthika N, Krishnakumar R, Nivedha S G
Affilition
Pages
17-20
Abstract
The escalation in mobile technology, specifically the utilization of smart phones and Android applications, highlights the complexity involved in retrieving essential data from applications, a procedure commonly referred to as data mining. The amalgamation of fraud identification and data mining in the Android market introduces a multifaceted situation. Given the multitude of applications vying for users' attention, with 2.8 million available on Google Play and 2.2 million on the Google App Store as of March 2017, alongside a substantial number of independent developers, the detection of deceitful behaviors within top-ranked applications emerges as a critical task. The objective of this study is to expose fraudulent practices in mobile applications, thereby establishing credibility and precision in rankings. Strategies such as human-operated water armies and bot farms manipulate rankings through the generation of counterfeit downloads, ratings, and feedback. The extraction of data, particularly user reviews, plays a crucial role in the algorithmic identification and alleviation of fraudulent actions that impact application rankings. Through this comprehensive strategy, the research endeavors to tackle the issue of upholding transparency and authenticity in the mobile application environment. Keywords - Fraud Detection System, Recurrent Neural Network, Ensemble Learning, Principal Component Analysis, Gated Recurrent Unit, Real-Time Fraud Detection
  View Paper