US 12,147,763 B2
System and method for identifying sentiment in text strings
Gregor Stewart, San Mateo, CA (US); Tzu-Ting Kuo, San Mateo, CA (US); and Andrew Yeager, San Mateo, CA (US)
Assigned to Medallia, Inc., San Francisco, CA (US)
Filed by MEDALLIA, INC., San Francisco, CA (US)
Filed on Mar. 29, 2021, as Appl. No. 17/216,048.
Application 17/216,048 is a continuation of application No. 16/297,137, filed on Mar. 8, 2019, granted, now 10,963,639.
Prior Publication US 2021/0216708 A1, Jul. 15, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 40/30 (2020.01); G06F 16/904 (2019.01); G06F 16/906 (2019.01); G06F 40/216 (2020.01); G06F 40/253 (2020.01); G06N 20/00 (2019.01)
CPC G06F 40/253 (2020.01) [G06F 16/904 (2019.01); G06F 16/906 (2019.01); G06F 40/216 (2020.01); G06N 20/00 (2019.01)] 18 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
receiving an input;
evaluating the input with a first model to determine an initial sentiment and confidence thereof;
if the confidence is greater than a threshold, using the initial sentiment;
if the confidence is lower than the threshold, accessing a list of secondary sentiments and evaluating the input for each of the secondary sentiments;
if the confidence is lower than the threshold, generating a relevantly similar confidence (RSC) score associated with each of the secondary sentiments; and
if the confidence is lower than the threshold, determining whether to use the initial sentiment or one sentiment of the secondary sentiments based on the generated RSC scores.