US 12,292,909 B2
Topic-based document segmentation
Allison Rossetto, New York, NY (US); and Michael Misiewicz, Brooklyn, NY (US)
Assigned to Yext, Inc., New York, NY (US)
Filed by Yext, Inc., New York, NY (US)
Filed on Jun. 30, 2022, as Appl. No. 17/854,829.
Prior Publication US 2024/0004911 A1, Jan. 4, 2024
Int. Cl. G06F 16/33 (2019.01); G06F 16/334 (2025.01); G06F 18/20 (2023.01); G06F 40/205 (2020.01)
CPC G06F 16/3347 (2019.01) [G06F 18/295 (2023.01); G06F 40/205 (2020.01)] 16 Claims
OG exemplary drawing
 
1. A method comprising:
identifying a document comprising text relating to a merchant system;
segmenting the document into a set of sentences;
generating, using a first machine-learning model executed by a processing device, an initial topic segmentation corresponding to the set of sentences based on an initial set of probabilities corresponding to each sentence, wherein, for each sentence of the set of sentences, the first machine-learning model generates the initial set of probabilities comprising a first probability corresponding to a first state and a second probability corresponding to a second state, and wherein the first state indicates that the sentence is a new topic and the second state indicates that the sentence is not a new topic; and
generating, using a second machine-learning model applied to the initial topic segmentation, a final topic segmentation corresponding to the document.