RiskRanker

摘要: Smartphone sales have recently experienced explosive growth. Their popularity also encourages malware authors to penetrate various mobile marketplaces with malicious applications (or apps). These apps hide in the sheer number of other normal apps, which makes their detection challenging. Existing anti-virus software are inadequate reactive nature by relying on known samples for signature extraction. In this paper, we propose a proactive scheme spot zero-day Android malware. Without and signatures, our is motivated assess potential security risks posed these untrusted apps. Specifically, developed an automated system called RiskRanker scalably analyze whether particular app exhibits dangerous behavior (e.g., launching root exploit or sending background SMS messages). The output then used produce prioritized list reduced that merit further investigation. When applied examine 118,318 total collected from markets over September October 2011, takes less than four days process all them effectively reports 3281 risky Among reported successfully uncovered 718 (in 29 families) 322 11 families). results demonstrate efficacy scalability police stripes.