US 12,131,394 B1
Distributed system for automated restaurant order acquisition
Rama Krishna Sandeep Pokkunuri, Redmond, WA (US); Roger Scott Jenke, Seattle, WA (US); Harshal Pimpalkhute, Redmond, WA (US); Yahor Pushkin, Redmond, WA (US); Swapandeep Singh, Sammamish, WA (US); Vasanth Philomin, Seattle, WA (US); and Ganesh Kumar Gella, Redmond, WA (US)
Assigned to Amazon Technologies, Inc., Seattle, WA (US)
Filed by Amazon Technologies, Inc., Seattle, WA (US)
Filed on Mar. 31, 2021, as Appl. No. 17/219,640.
Int. Cl. G06Q 50/12 (2012.01); G06N 20/00 (2019.01)
CPC G06Q 50/12 (2013.01) [G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. One or more non-transitory computer-accessible storage media storing program instructions that when executed on or across one or more processors cause the one or more processors to:
execute a first set of one or more machine learning models to process at least a portion of a first communication from a first user of a first restaurant at an order coordinator external to a provider network to generate a graph indicating a plurality of concepts associated with an order detected in the first communication and relationships among the concepts, wherein the execution of the first set of one or more machine learning models resolves co-references of the concepts in the first communication to reduce a number of concepts in the graph;
send, over one or more network connections, the graph of the concepts detected in the first communication to one or more resources of an order management service implemented at least in part at the provider network;
receive, over the one or more network connections, a response to the first communication from the order management service, wherein the response is to be presented to the first user by an interaction device, wherein at least a portion of the response is prepared by the order management service via execution of a second set of one or more machine learning models, wherein the response includes an item list of the order populated based at least in part on the first communication; and
cause, by the order coordinator, the order of the first user to be fulfilled based at least in part on analysis of a second communication of the first user, wherein the second communication is detected after the response is presented to the first user.