US 11,734,373 B2
Method, apparatus, device and computer storage medium for determining search result
Jizhou Huang, Beijing (CN); Haifeng Wang, Beijing (CN); and Wei Zhang, Beijing (CN)
Assigned to BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) CO., LTD., Beijing (CN)
Appl. No. 17/602,304
Filed by BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) CO., LTD., Beijing (CN)
PCT Filed May 28, 2020, PCT No. PCT/CN2020/092742
§ 371(c)(1), (2) Date Mar. 24, 2022,
PCT Pub. No. WO2021/128729, PCT Pub. Date Jul. 1, 2021.
Claims priority of application No. 201911373544.1 (CN), filed on Dec. 27, 2019.
Prior Publication US 2022/0237251 A1, Jul. 28, 2022
Int. Cl. G06F 16/9535 (2019.01); G06F 16/9538 (2019.01)
CPC G06F 16/9538 (2019.01) [G06F 16/9535 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A method of determining search results, wherein the method comprises:
obtaining a current query of a user, search history information of the user in a first time period, search history information of the user in a second time period and candidate search results for the current query, and inputting the current query of the user, the search history information of the user in the first time period, the search history information of the user in the second time period and the candidate search results for the current query into a search result ranking model, and determining search results corresponding to the current query according to scores of the candidate search results presented by the search result ranking model, the second time period being greater than the first time period;
wherein the scores of the candidate results presented by the search result ranking model are determined according to first similarity and a second similarity, the first similarity is a similarity between an integration of a vector representation of the current query and a vector representation of the search history information of the user in the first time period, and vector representations of the candidate search results, and the second similarity is a similarity between an integration of the vector representation of the current query and a vector representation of the search history information of the user in the second time period, and vector representations of the candidate search results.