In today's world of smartphone ubiquity, mobile malware is an increasingly prevalent (and difficult to mitigate) threat. One problem area for contemporary malware analysts is determining which apps legitimately need the permissions they request, and which have nefarious motivations. This presentation introduces a novel approach to mobile malware analysis at scale: synthetic sentiment analysis. Leveraging associative models of permissions, analysts can quickly determine which apps "feel" most suspicious--a huge time saver in a field with millions of apps to assess.