US 12,169,688 B2
Automatic content labeling
Sandro Cavallari, Singapore (SG)
Assigned to PayPal, Inc., San Jose, CA (US)
Filed by PayPal, Inc., San Jose, CA (US)
Filed on Oct. 3, 2023, as Appl. No. 18/376,199.
Application 18/376,199 is a continuation of application No. 17/014,636, filed on Sep. 8, 2020, granted, now 11,822,883.
Prior Publication US 2024/0143917 A1, May 2, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 40/20 (2020.01); G06F 16/33 (2019.01); G06F 18/2413 (2023.01); G06F 40/40 (2020.01); G06F 40/47 (2020.01); G06F 40/58 (2020.01); G06N 20/00 (2019.01); G06F 40/30 (2020.01)
CPC G06F 40/20 (2020.01) [G06F 16/3347 (2019.01); G06F 18/24147 (2023.01); G06F 40/40 (2020.01); G06F 40/47 (2020.01); G06F 40/58 (2020.01); G06N 20/00 (2019.01); G06F 40/30 (2020.01)] 20 Claims
OG exemplary drawing
 
1. A system comprising:
one or more processors; and
one or more machine-readable storage media having instructions stored thereon that, in response to being executed by the one or more processors, cause the system to perform operations comprising:
receiving a user query that includes unlabeled content;
embedding, using a machine learning model, the unlabeled content, wherein the embedding generates an unlabeled vector;
determining, from a plurality of labeled vectors stored in a vector index, a first set of labeled vectors that match the unlabeled vector, wherein the first set of labeled vectors are generated from a set of labeled content stored in a database;
calculating, based on a confidence score for at least one labeled vector in the first set of labeled vectors, a propagation score for the unlabeled vector, wherein the propagation score indicates whether a new label should be assigned to the unlabeled content;
assigning, based on the propagation score, the new label to the unlabeled content, wherein the new label is selected from the first set of labeled vectors;
storing the newly labeled content in the database;
retrieving based on the first set of labeled vectors and mappings between the vector index and the database storing labeled content, a set of labeled content from the database; and
responding, using the set of labeled content, to the user query.