US 12,265,794 B2
Methods and systems for generating problem description
Rami Cohen, Tel Aviv (IL); Noa Haas, Mountain View, CA (US); Oren Sar Shalom, Mountain View, CA (US); and Alexander Zhicharevich, Tel Aviv (IL)
Assigned to INTUIT INC., Mountain View, CA (US)
Filed by INTUIT INC., Mountain View, CA (US)
Filed on Oct. 6, 2023, as Appl. No. 18/482,783.
Application 18/482,783 is a continuation of application No. 18/180,089, filed on Mar. 7, 2023, granted, now 11,822,891.
Application 18/180,089 is a continuation of application No. 17/242,231, filed on Apr. 27, 2021, granted, now 11,625,541, issued on Apr. 11, 2023.
Prior Publication US 2024/0037342 A1, Feb. 1, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 40/30 (2020.01); G06N 20/20 (2019.01); G10L 15/26 (2006.01); H04M 3/51 (2006.01)
CPC G06F 40/30 (2020.01) [G06N 20/20 (2019.01); G10L 15/26 (2013.01); H04M 3/5175 (2013.01)] 15 Claims
OG exemplary drawing
 
1. A method performed by a computing system comprising:
extracting a problem description from text;
estimating a situation vector from the problem description;
identifying, by a machine learning model, a pre-existing situation vector that closely matches the estimated situation vector, the machine learning model being trained by:
generating a training data set comprising a plurality of standardized situation descriptions extracted from pre-existing case notes and filtering the standardized situation descriptions to include a subset of situation descriptions that start with a same set of characters, and
learning, by the machine learning model, to map each standardized situation description to a vector space, semantically similar situation descriptions being closely mapped in the vector space; and
retrieving a situation description that corresponds to the identified pre-existing situation vector.