US 11,954,441 B2
Device and method for generating article markup information
Yi-Chun Lin, New Taipei (TW); Yueh-Yarng Tsai, New Taipei (TW); Pin-Cyuan Lin, New Taipei (TW); Ke-Han Pan, New Taipei (TW); and Sheng-Wei Chu, New Taipei (TW)
Assigned to Acer Incorporated, New Taipei (TW)
Filed by Acer Incorporated, New Taipei (TW)
Filed on Jan. 4, 2022, as Appl. No. 17/568,687.
Claims priority of application No. 110142549 (TW), filed on Nov. 16, 2021.
Prior Publication US 2023/0153535 A1, May 18, 2023
Int. Cl. G06F 40/295 (2020.01); G06N 3/08 (2023.01)
CPC G06F 40/295 (2020.01) [G06N 3/08 (2013.01)] 11 Claims
OG exemplary drawing
 
1. A device for generating article markup information, comprising:
a transceiver; and
a processor, coupled to the transceiver, used to:
perform segmentation processing on an article to generate a segmentation result;
perform named entity recognition on the segmentation result according to a named entity recognition model to generate a first recognition result;
according to an expansion list, determine whether the segmentation result comprises any word among a plurality of words in the expansion list;
when the segmentation result comprises any one of the words in the expansion list, perform expanded entity classification conversion on the first recognition result according to the expansion list and the segmentation result to generate a second recognition result; and
use the second recognition result and the segmentation result as markup information, and output the markup information.
 
10. A method for generating article markup information, comprising:
performing segmentation processing on an article to generate a segmentation result;
performing named entity recognition on the segmentation result according to a named entity recognition model to generate a first recognition result;
according to an expansion list, determining whether the segmentation result comprises any word among a plurality of words in the expansion list;
when the segmentation result comprises any one of the words in the expansion list, performing expanded entity classification conversion on the first recognition result according to the expansion list and the segmentation result to generate a second recognition result; and
using the second recognition result and the segmentation result as markup information, and outputting the markup information.