Securing Android App Markets via Modeling and Predicting Malware Spread Between Markets

摘要: The Android ecosystem has recently dominated mobile devices. app markets, including official Google Play and other third party are becoming hotbeds, where malware originates spreads. been observed to both propagate within markets spread between markets. If the of can be predicted, market administrators take appropriate measures prevent outbreak minimize damages caused by malware. In this paper, we make first attempt protect modeling predicting To end, study social behaviors that affect malware, model these with multiple epidemic models, predict infection time order among for well-known families. achieve an accurate prediction spread, in following fashion: 1) a single market, within-market growth considering creation removal malware; 2) determine relevance calculating mutual information them; 3) based on previous two steps, simulate susceptible infected stochastically inference is performed using publicly available well-labeled dataset AndRadar. conduct extensive experiments evaluate our approach, collected large number (334,782) samples from 25 around world. experimental results show approach depict scale, 0.89 0.66 precision, respectively.