US 12,175,193 B2
System and method for lookup source segmentation scoring in a natural language understanding (NLU) framework
Omer Anil Turkkan, Santa Clara, CA (US); Edwin Sapugay, Foster City, CA (US); and Phani Bhushan Kumar Nivarthi, Fremont, CA (US)
Assigned to ServiceNow, Inc., Santa Clara, CA (US)
Filed by ServiceNow, Inc., Santa Clara, CA (US)
Filed on Jan. 19, 2022, as Appl. No. 17/579,063.
Claims priority of provisional application 63/140,031, filed on Jan. 21, 2021.
Prior Publication US 2022/0229990 A1, Jul. 21, 2022
Int. Cl. G06F 40/289 (2020.01); G06F 40/205 (2020.01); G06F 40/279 (2020.01); G06F 40/284 (2020.01); G06F 40/30 (2020.01); G06N 20/00 (2019.01)
CPC G06F 40/289 (2020.01) [G06F 40/279 (2020.01); G06F 40/284 (2020.01); G06F 40/30 (2020.01); G06N 20/00 (2019.01); G06F 40/205 (2020.01)] 20 Claims
OG exemplary drawing
 
1. A natural language understanding (NLU) framework, comprising:
at least one memory configured to store a lookup source system and a segmentation scoring subsystem, wherein the lookup source system has a plurality of lookup sources that each include a respective source data representation compiled from source data of a client database; and
at least one processor configured to execute stored instructions to cause the NLU framework to perform actions comprising:
performing, in parallel at the plurality of lookup sources, lookup source inference of a user utterance to determine a plurality of segmentations of the user utterance by:
dividing the user utterance into a plurality of sub-phrases;
providing the plurality of sub-phrases as input to each lookup source of the plurality of lookup sources of the lookup source system; and
performing, in parallel, lookup source inference of each sub-phrase of the plurality of sub-phrases at each lookup source of the plurality of lookup sources of the lookup source system, wherein each of the plurality of segmentations indicates how tokens of the user utterance have been matched to tokens of source data represented by the respective source data representation of one or more of the plurality of lookup sources;
performing, via the segmentation scoring subsystem, segmentation scoring of the plurality of segmentations of the user utterance to generate a corresponding plurality of segmentation scores for the plurality of segmentations; and
determining one or more intents of the user utterance, one or more entities of the user utterance, or any combination thereof, based at least in part on the plurality of segmentations and the plurality of corresponding segmentations scores.