US 12,462,102 B2
Deep learning for multimedia classification
Sunil Bharitkar, Stevenson Ranch, CA (US)
Assigned to Samsung Electronics Co., Ltd., Suwon-si (KR)
Filed by Samsung Electronics Co., Ltd., Suwon-si (KR)
Filed on Oct. 3, 2023, as Appl. No. 18/480,166.
Claims priority of provisional application 63/412,844, filed on Oct. 3, 2022.
Prior Publication US 2024/0126990 A1, Apr. 18, 2024
Int. Cl. G06F 40/279 (2020.01); G06F 18/24 (2023.01); G06F 40/284 (2020.01); G06N 3/0442 (2023.01); G06N 3/0985 (2023.01)
CPC G06F 40/284 (2020.01) [G06N 3/0442 (2023.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
utilizing text information obtained from a title of a media content item and a trainable model for improving accuracy for classification of the media content item;
utilizing the trainable model using a sequence of text to numeric-vector embeddings for classification of the media content item; and
jointly optimizing at least one of a word embedding model parameter or a latent semantic analysis dimension, using the text information, and a classifier model for maximizing accuracy of the classification of the media content item.