US 12,393,843 B2
Human-machine multi-turn conversation method and system for human-machine interaction, and intelligent apparatus
Zirong Huang, Guangdong (CN); Jutao Jia, Guangdong (CN); Wei Wu, Guangdong (CN); Yuhui Li, Guangdong (CN); and Lin Dai, Guangdong (CN)
Assigned to Gree Electric Appliances, Inc. of Zhuhai, Guangdong (CN); and Leayun Technology Co., Ltd. of Zhuhai, Zhuhai (CN)
Appl. No. 17/623,665
Filed by Gree Electric Appliances, Inc. of Zhuhai, Guangdong (CN); and Leayun Technology Co., Ltd. of Zhuhai, Guangdong (CN)
PCT Filed Jun. 30, 2020, PCT No. PCT/CN2020/099423
§ 371(c)(1), (2) Date Dec. 29, 2021,
PCT Pub. No. WO2021/027421, PCT Pub. Date Feb. 18, 2021.
Claims priority of application No. 201910740449.4 (CN), filed on Aug. 12, 2019.
Prior Publication US 2022/0253710 A1, Aug. 11, 2022
Int. Cl. G06N 3/082 (2023.01); G06F 16/3329 (2025.01); G06F 16/334 (2025.01); G06N 5/02 (2023.01)
CPC G06N 3/082 (2013.01) [G06F 16/3329 (2019.01); G06F 16/3344 (2019.01); G06N 5/02 (2013.01)] 12 Claims
OG exemplary drawing
 
1. A human-machine multi-turn conversation method for human-machine interaction, comprising:
S1, establishing a knowledge graph of user conversation behavior information, wherein the knowledge graph of user conversation behavior information is obtained by training on a basic graph template framework, and the basic graph template framework contains basic nodes and relationships;
S2, determining, according to information currently input by a user, a node corresponding to the information currently input and at least one child node of the node in the knowledge graph;
S3, calculating a support degree of the at least one child node relative to the node according to a number of times of querying the node and the number of times of querying both the at least one child node and the node in a historical query record of the knowledge graph; and
S4, determining whether to output semantic information of the at least one child node by determining a size relation between the support degree and a preset support degree threshold value,
wherein in S3, the support degree is obtained by calculating a ratio of the number of times of querying both the at least one child node and the node to the number of times of querying the node.