US 12,353,574 B2
Page processing method, electronic apparatus and non-transitory computer-readable storage medium
Xiaoqiang Lei, Beijing (CN); and Yunfei Wang, Beijing (CN)
Assigned to Baidu Online Network Technology (Beijing) Co., Ltd., Beijing (CN)
Appl. No. 17/278,370
Filed by Baidu Online Network Technology (Beijing) Co., Ltd., Beijing (CN)
PCT Filed Jul. 14, 2020, PCT No. PCT/CN2020/101910
§ 371(c)(1), (2) Date Mar. 22, 2021,
PCT Pub. No. WO2021/098242, PCT Pub. Date May 27, 2021.
Claims priority of application No. 202010125624.1 (CN), filed on Feb. 27, 2020.
Prior Publication US 2022/0114269 A1, Apr. 14, 2022
Int. Cl. G06F 21/62 (2013.01); G06F 16/957 (2019.01); G06F 16/958 (2019.01)
CPC G06F 21/62 (2013.01) [G06F 16/9577 (2019.01); G06F 16/958 (2019.01)] 18 Claims
OG exemplary drawing
 
1. A page processing method, comprising:
determining a plurality of layout object nodes of a page, according to an obtained Hypertext Markup Language (HTML) file;
filtering the plurality of layout object nodes according to a preset recall rule to obtain a layout object node satisfying the recall rule, after laying out the plurality of layout object nodes of the page;
predicting whether the layout object node satisfying the recall rule is a designated target node; and
shielding the designated target node, and generating a shielded page based on remaining layout target nodes after the shielding,
wherein predicting whether the layout object node satisfying the recall rule is the designated target node comprises:
calculating a node characteristic of the layout object node satisfying the recall rule, according to attribute information of the layout object node satisfying the recall rule, wherein the node characteristic is a feature having specified number of dimensions extracted and calculated from the attribute information of the layout object node, the specified number of dimensions is greater than or equal to 10;
processing the node characteristic by using a preset node prediction model, to obtain a probability value of the layout object node satisfying the recall rule being the designated target node; and
determining whether the layout object node satisfying the recall rule is the designated target node according to the probability value.