	US 12,231,514 B2
	Method and apparatus for generating prediction information, electronic device, and computer readable medium
Liangchao Wu, Beijing (CN); Lizhe Zhang, Beijing (CN); Junyuan Xie, Beijing (CN); Di Wu, Beijing (CN); Jun Zhang, Beijing (CN); Cheng Chen, Beijing (CN); Longyijia Li, Beijing (CN); Chenliaohui Fang, Beijing (CN); Kan Liu, Beijing (CN); Long Chang, Beijing (CN); Long Huang, Beijing (CN); Yixiang Chen, Beijing (CN); Xiang Wu, Beijing (CN); Peng Zhao, Beijing (CN); and Xiaobing Liu, Beijing (CN)
Assigned to BEIJING BYTEDANCE NETWORK TECHNOLOGY CO., LTD., Beijing (CN)
Filed by Beijing Bytedance Network Technology Co., Ltd., Beijing (CN)
Filed on Aug. 5, 2022, as Appl. No. 17/882,397.
Application 17/882,397 is a continuation of application No. PCT/CN2021/080634, filed on Mar. 12, 2021.
Claims priority of application No. 202010270721.X (CN), filed on Apr. 8, 2020.
Prior Publication US 2022/0385739 A1, Dec. 1, 2022
Int. Cl. H04L 67/55 (2022.01); G06F 21/62 (2013.01)

CPC H04L 67/55 (2022.05) [G06F 21/6254 (2013.01)]

13 Claims

1. A method for transferring information, applied to a first backend server for a first application, comprising:

generating, by a processor of the first backend server for the first application based on first user characteristic information of a target user, anonymous user information of the target user;

sending, by the processor of the first backend server for the first application, the anonymous user information to a second backend server for a second application to enable a processor of the second backend server for the second application to generate prediction information based on the anonymous user information and second user characteristic information of the target user; wherein the first user characteristic information comprises information obtained from historical browsing information of the target user and login information of the target user on the first application, and a region where the target user is located, the second user characteristic information comprises the information obtained from the historical browsing information of the target user and the login information of the target user on the second application, and the region where the target user is located;

receiving, by the processor of the first backend server for the first application from the processor of the second backend server for the second application, the prediction information;

generating, by the processor of the first backend server for the first application based on the prediction information, push information;

pushing, by the processor of the first backend server for the first application, the push information to a terminal where the first application and the second application are installed corresponding to the target user to enable the target user to view the push information on the terminal by a clicking operation;

receiving, by the processor of the first backend server for the first application, a user request of the target user;

receiving, by the processor of the first backend server for the first application, user behavior information; and

in response to receiving the user behavior information, associating and storing, by the processor of the first backend server for the first application, an identifier of the user request and the user behavior information, wherein the user behavior information is used to characterize a behavior executed by the user for a jumping page, the jumping page is a page that the user jumps to after executing an operation on the push information;

wherein the generating, based on the first user characteristic information of the target user, the anonymous user information of the target user comprises:

inputting the first user characteristic information into a first prediction sub-model to obtain a first user characteristic embedding vector;

splicing the first user characteristic embedding vector and a context vector to obtain a splicing vector;

multiplying the splicing vector and a parameter matrix of the first prediction sub-model to obtain the anonymous user information of the target user.