Interaction message processing method and apparatus, computer device, and storage medium

Patent metadata

Field	Value
Publication number	US-2022247708-A1
Application number	US-202217727642-A
Country	US
Kind code	A1
Filing date	Apr 22, 2022
Priority date	Mar 26, 2019
Publication date	Aug 4, 2022
Grant date	—

Abstract

A method is performed at a backend server. The server receives a to-be-processed interaction message from a user interface of a social network platform. The server performs intention classification on the to-be-processed interaction message, to obtain an intention classification result of the to-be-processed interaction message. The server performs semantic analysis on the to-be-processed interaction message when the intention classification result comprises a mini program intention, to determine a message keyword of the to-be-processed interaction message. The server determines a program identifier of a target mini program corresponding to the to-be-processed interaction message according to the message keyword. The server transmits a message processing result based on the program identifier of the target mini program.

First claim

What is claimed is:

1 . A method performed at a backend server, the method comprising:
receiving, by the server, a to-be-processed interaction message from a user interface of a social network platform; performing intention classification on the to-be-processed interaction message, to obtain an intention classification result of the to-be-processed interaction message; performing semantic analysis on the to-be-processed interaction message when the intention classification result comprises a mini program intention, to determine a message keyword of the to-be-processed interaction message; determining a program identifier of a target mini program corresponding to the to-be-processed interaction message according to the message keyword; and transmitting a message processing result based on the program identifier of the target mini program.

2 . The method according to claim 1 , wherein performing intention classification on the to-be-processed interaction message, to obtain an intention classification result of the to-be-processed interaction message comprises:
extracting a message feature of the to-be-processed interaction message; and performing preset intention classification mapping on the message feature using a neural network model to obtain the intention classification result of the to-be-processed interaction message.

3 . The method according to claim 2 , wherein the neural network model is a fully connected neural network model, and performing preset intention classification mapping on the message feature by using a neural network model, to obtain the intention classification result of the to-be-processed interaction message comprises:
performing convolution mapping on the message feature by using a convolutional layer of the fully connected neural network model to obtain a convolution result; performing pooling mapping on the convolution result by using a pooling layer of the fully connected neural network model to obtain a pooling result; and performing fully connected mapping on the pooling result by using a fully connected layer of the fully connected neural network model, to obtain the intention classification result of the to-be-processed interaction message.

4 . The method according to claim 2 , wherein extracting a message feature of the to-be-processed interaction message comprises:
performing speech recognition on the to-be-processed interaction message to obtain a text message; and extracting a message feature of the text message.

5 . The method according to claim 1 , wherein transmitting a message processing result based on the program identifier of the target mini program comprises:
obtaining program information of the target mini program according to the program identifier, the program information comprising program link information or invoked interface information of the target mini program; and transmitting the message processing result based on the program information.

6 . The method according to claim 5 , wherein the transmitting the message processing result based on the program information comprises:
requesting page data of the target mini program according to the program information; and transmitting the message processing result based on the page data and the program information after receiving the page data.

7 . The method according to claim 1 , wherein the determining a program identifier of a target mini program corresponding to the to-be-processed interaction message according to the message keyword comprises:
determining candidate mini programs corresponding to the to-be-processed interaction message according to the message keyword; extracting mini program features of the candidate mini programs; and sorting the candidate mini programs according to the mini program features, and determining a program identifier of a candidate mini program arranged at a preset position as the program identifier of the target mini program.

8 . The method according to claim 7 , wherein the extracting mini program features of the candidate mini programs comprises:
extracting text features of the candidate mini programs and non-text features of the candidate mini programs in a case that the candidate mini programs are text-type mini programs; or extracting non-text features of the candidate mini programs in a case that the candidate mini programs are non-text-type mini programs.

9 . A server, comprising:
one or more processors; and memory storing one or more programs, the one or more programs comprising instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising: receiving a to-be-processed interaction message from a user interface of a social network platform; performing intention classification on the to-be-processed interaction message, to obtain an intention classification result of the to-be-processed interaction message; performing semantic analysis on the to-be-processed interaction message when the intention classification result comprises a mini program intention, to determine a message keyword of the to-be-processed interaction message; determining a program identifier of a target mini program corresponding to the to-be-processed interaction message according to the message keyword; and transmitting a message processing result based on the program identifier of the target mini program.

10 . The server according to claim 9 , wherein performing intention classification on the to-be-processed interaction message, to obtain an intention classification result of the to-be-processed interaction message comprises:
extracting a message feature of the to-be-processed interaction message; and performing preset intention classification mapping on the message feature using a neural network model to obtain the intention classification result of the to-be-processed interaction message.

11 . The server according to claim 10 , wherein the neural network model is a fully connected neural network model, and performing preset intention classification mapping on the message feature by using a neural network model, to obtain the intention classification result of the to-be-processed interaction message comprises:
performing convolution mapping on the message feature by using a convolutional layer of the fully connected neural network model to obtain a convolution result; performing pooling mapping on the convolution result by using a pooling layer of the fully connected neural network model to obtain a pooling result; and

12 . The server according to claim 9 , wherein transmitting a message processing result based on the program identifier of the target mini program comprises:
obtaining program information of the target mini program according to the program identifier, the program information comprising program link information or invoked interface information of the target mini program; and transmitting the message processing result based on the program information.

13 . The server according to claim 12 , wherein transmitting the message processing result based on the program information comprises:
requesting page data of the target mini program according to the program information; and transmitting the message processing result based on the page data and the program information after receiving the page data.

14 . The server according to claim 9 , wherein determining a program identifier of a target mini program corresponding to the to-be-processed interaction message according to the message keyword comprises:
determining candidate mini programs corresponding to the to-be-processed in

Assignees

• Tencent Tech Shenzhen Co Ltd

Inventors

• Zhang Yan
• Liu Jian
• Liu Wei
• Zhong Randy
• Liu Qiuge
• Niu Cheng
• Zhou Jie

Classifications

• G06Q10/40

  Business processes related to social networking or social networking services · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G06F40/30

  Semantic analysis · CPC title

• G06N3/0499

  Feedforward networks · CPC title

• G06N3/09

  Supervised learning · CPC title