| US 11,704,589 B1 | ||
| Automatically identifying dynamic applications | ||
| Saurabh Sohoney, Bangalore (IN); Vineet Shashikant Chaoji, Bangalore (IN); and Pranav Garg, Bangalore (IN) | ||
| Assigned to AMAZON TECHNOLOGIES, INC., Seattle, WA (US) | ||
| Filed by Amazon Technologies, Inc., Seattle, WA (US) | ||
| Filed on Mar. 20, 2017, as Appl. No. 15/463,439. | ||
| Int. Cl. G06N 20/00 (2019.01); G06F 21/57 (2013.01); G06Q 30/06 (2023.01); G06Q 30/0601 (2023.01) | ||
| CPC G06N 20/00 (2019.01) [G06F 21/577 (2013.01); G06Q 30/0601 (2013.01)] | 20 Claims |
|
1. A non-transitory computer-readable medium embodying a program executable in at least one computing device, wherein when executed the program causes the at least one computing device to at least:
generate a machine learning model through a static analysis of a first plurality of applications that are manually classified as static applications and a second plurality of applications that are manually classified as dynamic applications; receive an application from an inbound processing queue; identify at least one first feature using a static analysis of the application, the at least one first feature being indicative of a dependence of the application on network-provided data in a critical path of execution; identify at least one second feature using the static analysis of the application, the at least one second feature being a network call that is not indicative of the dependence of the application on network-provided data in a critical path of execution; and automatically classify the application as a dynamic application instead of a static application using the machine learning model based at least in part on the at least one first feature as opposed to the at least one second feature. |