作者: Mahmudur Rahman , Mizanur Rahman , Bogdan Carbunar , Duen Horng Chau
DOI: 10.1109/TKDE.2017.2667658
关键词:
摘要: Fraudulent behaviors in Google Play, the most popular Android app market, fuel search rank abuse and malware proliferation. To identify malware, previous work has focused on executable permission analysis. In this paper, we introduce FairPlay, a novel system that discovers leverages traces left behind by fraudsters, to detect both apps subjected fraud. FairPlay correlates review activities uniquely combines detected relations with linguistic behavioral signals gleaned from Play data (87 K apps, 2.9 M reviews, 2.4M reviewers, collected over half year), order suspicious apps. achieves 95 percent accuracy classifying gold standard datasets of fraudulent legitimate We show 75 identified engage hundreds currently evade Bouncer's detection technology. also helped discovery more than 1,000 reported for 193 reveal new type “coercive” campaign: users are harassed into writing positive install other