Checking app behavior against app descriptions

[WeiFoo]

paper link:http://dl.acm.org/ft_gateway.cfm?id=2568276&ftid=1467964&dwn=1&CFID=609851752&CFTOKEN=92241930 dataset: use the data in reference http://www.csc.ncsu.edu/faculty/jiang/pubs/OAKLAND12.pdf

[bamal]

Paper Title: Checking App Behavior Against App Descriptions Authors: -Alessandra Gorla -Ilaria Tavecchia∗ -Florian Gross -Andreas Zeller

• Emails: {gorla, tavecchia, fgross, zeller}@cs.uni-saarland.de
• Objective: Knowing when a program does what it claims to do.
• Contribution: Proposition of CHABADA prototype that clusters apps by description topics, and identifys outliers by API usage within each cluster.
• Findings: Our CHABADA approach effectively identifies applications whose behavior would be unexpected given their description. Experiments were applied on a set of 22,500+ Android applications, our CHABADA prototype identified several anomalies; additionally, it flagged 56% of novel malware as such, without requiring any known malware patterns. @WeiFoo we can access data directly using this link.
• Dataset: https://www.st.cs.uni-saarland.de/appmining/chabada/
• Data Description: we have prepared a 50 MB dataset with the exact data that goes into our CHABADA prototype.

[reesjones]

Added as chabada.