US 12,327,202 B2
Entity tag association prediction method, device, and computer readable storage medium
Lizhi Liu, Shanghai (CN); Yu Wang, Shenzhen (CN); Haijian Jiang, Shanghai (CN); and Qing Min, Shanghai (CN)
Assigned to CHINA UNIONPAY CO., LTD., Shanghai (CN)
Appl. No. 18/713,631
Filed by CHINA UNIONPAY CO., LTD., Shanghai (CN)
PCT Filed Sep. 7, 2022, PCT No. PCT/CN2022/117421
§ 371(c)(1), (2) Date May 24, 2024,
PCT Pub. No. WO2023/093205, PCT Pub. Date Jun. 1, 2023.
Claims priority of application No. 202111424257.6 (CN), filed on Nov. 26, 2021.
Prior Publication US 2024/0419942 A1, Dec. 19, 2024
Int. Cl. G06N 7/01 (2023.01); G06F 17/16 (2006.01); G06N 3/045 (2023.01)
CPC G06N 7/01 (2023.01) [G06F 17/16 (2013.01); G06N 3/045 (2023.01)] 19 Claims
OG exemplary drawing
 
1. An entity tag association prediction method, comprising:
determining an entity relationship network, a tag relationship network and an entity tag association network, the entity tag association network comprising an unknown entity tag association relationship, wherein determining the entity relationship network, the tag relationship network and the entity tag association network comprises:
determining the entity relationship network Gu(custom character, εu) with multiple entities in an entity set custom character={u1, u2, . . . , un} as nodes, wherein n is a total number of entities, and εu ⊆custom character×custom character indicates a relationship between various entities in the entity set;
determining the tag relationship network Gh (custom character, εh) with multiple tags in a tag set custom character={h1, h2, . . . , hm} as nodes, wherein m is a total number of tags, and εh⊆custom character×custom character indicates a relationship between various tags in the tag set custom character={h1, h2, . . . , hm}; and
determining the entity tag association network custom character⊆{0,1}n×m according to existing annotation information, wherein association custom characterij between each entity ui (i=1, 2, . . . , n) and each tag hj (j=1, 2, . . . , m) is 1 or 0;
constructing an entity similarity graph according to the entity relationship network, constructing a tag similarity graph according to the tag relationship network and the entity tag association network, and constructing an entity tag association bipartite graph according to the entity tag association network;
extracting an entity feature, and constructing a tag feature according to the tag similarity graph;
integrating the entity similarity graph, the tag similarity graph, and the entity tag association bipartite graph into a graph convolutional network to construct a prediction model; and
inputting the entity feature and the tag feature into the prediction model for training until the prediction model converges, and outputting a prediction result of the prediction model, the prediction result comprising an association relationship between each entity and each tag.