WO2022217708A1

WO2022217708A1 - Artificial intelligence-based question and answer method and apparatus, server, and storage medium

Info

Publication number: WO2022217708A1
Application number: PCT/CN2021/097022
Authority: WO
Inventors: 任龙伟
Original assignee: 平安科技（深圳）有限公司
Priority date: 2021-04-13
Filing date: 2021-05-29
Publication date: 2022-10-20
Also published as: CN112925898A; CN112925898B

Abstract

An artificial intelligence-based question and answer method and apparatus, a server, and a storage medium. The method is applied to the field of artificial intelligence, and the method comprises: acquiring a question and answer request submitted by a first terminal, the question and answer request carrying user information of a target user and question description information (S101); acquiring a service information set of the target user according to the user information (S102); identifying the question intention of the target user by using the question description information to obtain an intention recognition result (S103); converting the intention recognition result into a target question and answer instruction (S104); matching a target reply template for the target user according to the service information set and the target question and answer instruction (S105); and obtaining reply content for the target user according to the target reply template, and outputting the reply content by means of the first terminal (S106). The described method can improve the quality of a question and answer service. The method relates to blockchain technology, and can write user information, question description information, and reply content into a blockchain.

Description

Question answering method, device, server and storage medium based on artificial intelligence

This application claims the priority of the Chinese patent application filed on April 13, 2021 with the application number 202110395698.1 and the invention titled "An artificial intelligence-based question answering method, device, server and storage medium", the entire content of which is approved by Reference is incorporated in this application.

technical field

The present application relates to the field of computer technology, in particular to the field of artificial intelligence technology, and in particular to a question answering method, device, server and storage medium based on artificial intelligence.

Background technique

Today, with the extensive development of e-commerce, e-government, and enterprise informatization, various websites or platforms are faced with massive demands for customer consultation, customer service, and customer notification. The traditional manual service method is far from meeting the demand. In recent years, the industry has introduced intelligent customer service robots to liberate manpower and improve efficiency.

The inventors found that the traditional intelligent customer service robot uses word segmentation processing, part-of-speech tagging, and a constantly enriched semantic knowledge base to perform question-style retrieval after semantic understanding, and retrieves answers from the question-and-answer database to answer. More intelligently, if an accurate answer cannot be retrieved, it will match the customer's question, and select a relatively accurate answer from a number of candidate results according to the similarity calculation.

The inventor realized that in the face of rapid business development and changes, and when different businesses need to be graded or have personalized requirements for answering content, such robots are often unable to respond quickly and accurately, and the quality of Q&A services is poor.

technical problem

The embodiments of the present application provide a question and answer method, device, server and storage medium based on artificial intelligence, which can improve the quality of question and answer service.

technical solutions

In a first aspect, an embodiment of the present application provides a question and answer method based on artificial intelligence, including: acquiring a question and answer request submitted by a first terminal, where the question and answer request carries user information and question description information of a target user; Obtain the business information set of the target user; identify the questioning intention of the target user by using the problem description information, and obtain an intent identification result; convert the intent identification result into a target question and answer instruction; according to the business information set and the target question answering instruction matches a target reply template for the target user; obtains reply content for the target user according to the target reply template, and outputs the reply content through the first terminal.

In a second aspect, the present application provides an artificial intelligence-based question-and-answer device, comprising: an acquisition module configured to acquire a question-and-answer request submitted by a first terminal, where the question-and-answer request carries user information and question description information of a target user; the The acquisition module is further configured to acquire the business information set of the target user according to the user information; the identification module is configured to identify the questioning intention of the target user by using the problem description information to obtain an intention identification result; the conversion module , for converting the intent recognition result into a target question and answer instruction; a matching module, for matching out a target reply template for the target user according to the business information set and the target question and answer instruction; the acquisition module, also using obtaining the reply content for the target user according to the target reply template; the communication module is used for outputting the reply content through the first terminal.

In a third aspect, an embodiment of the present application provides a server, including a processor and a memory, where the processor and the memory are connected to each other, wherein the memory is used to store a computer program, and the computer program includes program instructions, The processor is configured to invoke the program instruction and execute the following method: acquiring a question and answer request submitted by the first terminal, where the question and answer request carries user information and question description information of the target user; acquiring the question and answer request according to the user information; business information collection of the target user; use the problem description information to identify the questioning intention of the target user to obtain an intention identification result; convert the intention identification result into a target question and answer instruction; according to the business information collection and the The target question answering instruction matches a target reply template for the target user; obtains reply content for the target user according to the target reply template, and outputs the reply content through the first terminal.

In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the following method: obtaining a question and answer submitted by a first terminal The question and answer request carries the user information and question description information of the target user; obtains the business information set of the target user according to the user information; uses the question description information to identify the questioning intention of the target user, and obtains Intention recognition result; convert the intention recognition result into a target question answering instruction; match a target reply template for the target user according to the business information set and the target question answering instruction; obtain a target answer template according to the target answer template reply content of the user, and output the reply content through the first terminal.

beneficial effect

In the face of rapid business development and changes, and the need for hierarchical processing of different businesses or individual requirements for response content, this application can obtain a collection of business information in real time, combined with the target question and answer obtained through technologies such as intent recognition Instructions are used to match templates, so as to generate answer content, the question and answer process is faster, the answer content is more accurate, and the quality of question and answer service is effectively improved.

Description of drawings

FIG. 1 is a schematic flowchart of an artificial intelligence-based question answering method provided by an embodiment of the present application.

FIG. 2a is a schematic diagram of a relationship between a question and answer instruction and a reply template provided by an embodiment of the present application.

FIG. 2b is a schematic diagram of a target page provided by an embodiment of the present application.

FIG. 2c is a schematic diagram of another target page provided by an embodiment of the present application.

FIG. 2d is a schematic diagram of another target page provided by an embodiment of the present application.

FIG. 2e is a schematic diagram of another target page provided by an embodiment of the present application.

FIG. 2f is a schematic diagram of a question and answer process provided by an embodiment of the present application.

FIG. 3 is a schematic flowchart of another question answering method based on artificial intelligence provided by an embodiment of the present application.

FIG. 4 is a schematic structural diagram of an artificial intelligence-based question answering device provided by an embodiment of the present application.

FIG. 5 is a schematic structural diagram of a server provided by an embodiment of the present application.

Embodiments of the present invention

The technical solutions in the embodiments of the present application will be described below with reference to the accompanying drawings in the embodiments of the present application.

The technical solution of the present application may relate to the field of artificial intelligence technology, for example, it can be applied to scenarios such as smart question and answer, so as to improve the quality of question and answer service, thereby promoting the construction of a smart city. Optionally, the data involved in this application, such as user information, reply templates and/or reply content, can be stored in a database, or can be stored in a blockchain, such as distributed storage through a blockchain, which is not limited in this application. .

Please refer to FIG. 1 , which is a schematic flowchart of an artificial intelligence-based question answering method according to an embodiment of the present application. This method can be applied to the server. The server can be a server or a cluster of servers. Specifically, the method may include the following steps.

S101. Obtain a question and answer request submitted by a first terminal, where the question and answer request carries user information and question description information of a target user.

The first terminal may be a smart phone, a tablet computer, a notebook computer, a self-service terminal, a robot, such as an intelligent question answering robot installed in a shopping mall. The target user may be a user corresponding to the first terminal.

The user information may include user identification information. User identification information is the information used to uniquely identify the target user, including but not limited to any of the following: user nickname, user account, user mobile phone number,

The problem description information includes a problem description sentence. In one embodiment, the question description information may include question description sentences and question context information, and the question context information may include an ID of the question and answer request (used to uniquely identify a question and answer request) and question and answer rounds (such as round 1, round 2) ).

In one embodiment, the question and answer request submitted by the first terminal may be forwarded to the server by the first terminal through another server, or may also be sent to the server by the first terminal.

In an application scenario, the question-and-answer request may be generated by the first terminal after detecting the click operation of the send button by the target user, obtaining a question description sentence input by the target user, and generating the question description sentence. Or, the question and answer request may also be generated according to the target reply content after the first terminal detects the click operation of the target user on the send button and obtains the target reply content selected by the target user from the existing reply content. The content may be a certain reply content that has been displayed on the chat interface.

S102. Acquire the service information set of the target user according to the user information.

In this embodiment of the present application, the server may query the target user's service information set according to the user information. For example, the server may query the target user's service information set from a specified database or target service system according to the user information. The link address for accessing the target business system may be configured based on the first plug-in template. Wherein, according to the different services used or consulted by the user, the acquired service information sets are also different. For example, for loan business, the business information set may include loan amount, loan time, repayment period, repayment amount and other business information; for customer service business, the business information set may include service date, user appeal, user attitude and other business information .

S103: Identify the questioning intent of the target user by using the question description information to obtain an intent identification result.

S104. Convert the intent recognition result into a target question answering instruction.

Wherein, the intent recognition result may include at least one keyword indicating the questioning intent of the target user. The target question-and-answer instruction refers to the question-and-answer instruction converted according to the intent recognition result. The target question answering instruction can be a string. In one embodiment, the target question answering instruction may be a character string indicating a business field corresponding to the question description information, and the character string is composed of numbers and/or letters.

In the embodiment of the present application, the server uses the problem description information to identify the questioning intention of the target user, and the method of obtaining the intention recognition result may be: the server invokes the intention recognition model to identify the questioning intention of the target user according to the problem description information, and obtains the intention identification result.

In one embodiment, the server converts the intent recognition result into the target question answering instruction may be: the server determines the target question answering instruction corresponding to the intent identification result according to the correspondence between the keyword and the question answering instruction.

In one embodiment, the server may also send question description information to a specified device (such as a specified robot), triggering the specified device to use the question description information to identify the questioning intent of the target user, obtain an intent identification result, and use the intent identification result Convert it to a target question and answer command, and receive the target question and answer command returned by the specified device.

S105. Match a target reply template for the target user according to the service information set and the target question answering instruction.

The target reply template refers to a reply template matched for the target user.

In one embodiment, the server may determine, according to the correspondence between the question and answer instruction and the answer template, the answer template corresponding to the target question and answer instruction as the target answer template.

In one embodiment, the server may determine a use condition corresponding to the target question and answer instruction, and obtain the target reply template when the use condition is satisfied. The usage condition can be used to judge whether the target reply template can be used, or can also be used to determine which reply template is specifically used as the target reply template.

In one embodiment, when the usage condition is a usage condition set based on a conditional expression, the server may acquire the conditional expression corresponding to the target question and answer instruction, and acquire the target reply template according to the service information set and the conditional expression.

In one embodiment, the server may obtain the conditional expression corresponding to the target question and answer instruction, and substitute each service information in the service information set into the conditional expression for operation to obtain the operation result; the server determines the reply corresponding to the operation result template, and determine the response template corresponding to the operation result as the target response template. Different operation results correspond to different reply templates. For example, referring to FIG. 2a, one operation instruction may correspond to 4 reply templates. Here, the reply template corresponding to the obtained operation result can be determined as the target reply template. The symbol "*" in the drawings of the embodiments of the present application hides the corresponding characters (such as numbers or letters), and in the actual application process, the corresponding characters (such as numbers or letters) are displayed at the location of the symbol.

In one embodiment, the server may output a target page through the second terminal, where the target page is a page capable of configuring question-and-answer instructions, usage conditions, and reply templates. Based on the target page, the second terminal may configure the target question-and-answer instruction, as well as the usage condition and reply template corresponding to the target question-and-answer instruction. After the second terminal completes the configuration of the target question and answer instruction, the use condition and the reply template corresponding to the target question and answer instruction, the server can obtain the target question and answer instruction, the use condition and the reply template corresponding to the target question and answer instruction, and save the target question and answer instruction, and the target question and answer instruction. The usage conditions and response templates corresponding to the question and answer command. In one embodiment, the second terminal may be a terminal corresponding to a developer who can use the server, or may also be a terminal corresponding to an administrator corresponding to the target business system, and so on. Wherein, the second terminal may be a desktop computer, a notebook computer, a laptop computer, and so on.

In one embodiment, the target page may be a page as shown in FIG. 2b to FIG. 2c, the second terminal may display the page, and the user corresponding to the second terminal may configure the target question and answer instruction based on the page, and the corresponding target question and answer instruction Use conditions and reply templates. After the configuration is completed, the user corresponding to the second terminal can click the OK button on the page to send the target question and answer instruction, the usage conditions and reply template corresponding to the target question and answer instruction to the server, and the server can receive the target question and answer. The instruction, as well as the use conditions and response templates corresponding to the target question and answer instructions, and save the target question and answer instructions, as well as the use conditions and answer templates corresponding to the target question and answer instructions. The target page includes a question and answer instruction input area, a use condition input area, and a reply template input area. The question and answer instruction input area is used to input question and answer instructions, such as target question and answer instructions. Use the conditional input area to enter conditional expressions. The reply template input area is used to enter reply templates. In one embodiment, when the conditional input area is used to set the conditional expression, it can be used to input the operator and the comparison value on both sides of the operator. In one embodiment, the target page may further include a status selection option, including a status valid option and a status invalid option, so that when it is determined that the setting is valid, it is determined whether the usage condition is satisfied.

In one embodiment, the server may output a first page through the second terminal, where the first page includes multiple question and answer instructions, and the multiple question and answer instructions include target question and answer instructions. After detecting the selection instruction for the target question and answer instruction sent by the second terminal, the server outputs a second page through the second terminal, so that the second terminal configures the conditional expression and reply corresponding to the target question and answer instruction based on the second page template. The server acquires the conditional expression and the reply template corresponding to the target question and answer instruction, and saves the conditional expression and the reply template corresponding to the target question and answer instruction. The second page includes a question and answer instruction display area, a conditional expression input area, and a reply template input area. On the second page here, the target question and answer instructions are displayed in the question and answer instruction display area. In one embodiment, the question and answer instruction display area may not respond to the question and answer instruction input operation, that is, the user corresponding to the second terminal will not be able to perform the question and answer instruction input operation based on the question and answer instruction display area. In one embodiment, when the conditional input area is used to set the conditional expression, it can be used to input the operator and the comparison value on both sides of the operator. In one embodiment, the second page may further include a status selection option, including a status valid option and a status invalid option, so that when it is determined that the setting is valid, it is determined whether the usage condition is satisfied. Wherein, examples of the second page can also refer to FIG. 2b-FIG. 2c. Although the style of the second page and the target page may be consistent, there are differences in the functions of the two in the question and answer instruction display area and the question and answer instruction input area.

In one embodiment, the manner in which the server outputs the second page through the second terminal is specifically that the server searches for a second plug-in template, where the second plug-in template includes the page code of the second page. The server sends the second plug-in template to the second terminal, so that the second terminal generates the second page according to the second plug-in template, and outputs the second page. Here, the second plug-in template can be embedded in the second terminal in the form of a page plug-in.

In one embodiment, when the use condition is based on a preset value, such as a use condition set by a general answer, the server may acquire a reply template corresponding to the target question-answer instruction as the target reply template. Compared with the usage condition set based on the conditional expression, the usage condition set based on the preset value does not need to perform conditional judgment. For example, the target page or the second page may also refer to the pages shown in FIG. 2d to FIG. 2e, and the target page and the second page have the above differences. From the pages shown in Figures 2d-2e, it can be seen that the conditions of use are generic responses.

S106: Obtain reply content for the target user according to the target reply template, and output the reply content through the first terminal.

In one embodiment, the server may determine the target reply template as the reply content to the target user.

In one embodiment, the server may identify the type of the target reply template, and when the target reply template is a general template, determine the target reply template as the reply content to the target user.

In one embodiment, when the target reply template is a replacement template, the server may use each service information in the service information set to replace the service variable corresponding to the service information in the target reply template, and obtain a replaced reply template, which will replace the The subsequent reply template is determined as the reply content to the target user.

In one embodiment, when the target reply template is a circular template, the server may use each service information in the service information set to replace the service variable corresponding to the service information in the target reply template multiple times to obtain multiple replaced replies template, and merge multiple replaced reply templates to obtain a merged reply template, and determine the merged reply template as the reply content to the target user.

For example, referring to Figure 2f, suppose the target user has 8 loan data, and the corresponding configured answer template is: application time: [application time], product: [product name], loan amount: [loan amount]. When the target answer template is a loop template, the server can process each piece of loan data (8 pieces in this example) of the target user in a loop, and use each piece of loan data to replace the business variables in the answer template to obtain 8 filled answer templates (The Nth loan data corresponds to the serial number N), and finally the 8 filled answer templates are combined into a complete answer template as the answer content to the target user.

It can be seen that in the embodiment of FIG. 1, the server can obtain the question and answer request submitted by the first terminal, and the question and answer request carries the user information and question description information of the target user; the server can obtain the target user's business information set according to the user information, and use the question description information. Obtain the intent recognition result about the target user; the server can convert the intent recognition result into a target question and answer instruction, and match a target reply template for the target user according to the business information set and the target question answer instruction, so as to obtain a reply for the target user according to the target reply template content, and output the reply content through the first terminal, this process can effectively improve the quality of the question and answer service.

Please refer to FIG. 3 , which is a schematic flowchart of another question-answering method based on artificial intelligence provided by an embodiment of the present application. This method can be applied to the server. The server can be a server or a cluster of servers. Specifically, the method may include the following steps.

S301. Obtain a question and answer request submitted by a first terminal, where the question and answer request carries user information and question description information of a target user, and the user information includes user identification information and service system indication information.

Wherein, the user identification information is information used to uniquely identify the target user, including but not limited to any of the following: user nickname, user account, user mobile phone number, service system indication information is the indication information that can be used to determine the target service system, For example, the service system indication information may be identification information of the target service system or an application account registered by a service provider (eg, an enterprise). Wherein, the identification information of the target service system is information used for a unique target service system, for example, the identification information of the target service system may be xx service system or xx service. The application account registered by the service provider is not the identification information of the service system. For example, the application account registered by the service provider may be the public account registered by the service provider on the public account platform.

S302. Acquire the service information set of the target user from the target service system according to the user identification information, where the target service system is the service system indicated by the service system indication information.

In one embodiment, the server obtains the service information set of the target user from the target service system according to the user identification information, and can query the link address for accessing the target service system for the server according to the service system indication information, and perform a query on the target service system for the server. The user identification information and the link address are spliced to obtain a spliced link address, so as to obtain the target user's service information set from the target service system according to the spliced link address. In one embodiment, the link address may be a link address for accessing the business information set of each user of the target business system.

In one embodiment, the server may look up a first plug-in template, where the first plug-in template includes page code of a link address configuration page. The first plug-in template may be embedded in the target business system in the form of a page plug-in. The server sends the first plug-in template to the target service system, so that the target service system configures a link address for accessing the target service system based on the first plug-in template. The server receives the link address for accessing the target business system returned by the target business system, and stores the link address for accessing the target business system. In an application scenario, the server can search for the first plug-in template, and then send the first plug-in template to the target business system. The administrator of the target business system can access the target business system and view the link address configuration generated based on the first plug-in template. page, and then configure the link address of the target business system based on the link address configuration page. Specifically, you can configure the link address for accessing the business information collection of each user of the target business system, and then return the configured link address to the target business system. The business system can forward the link address to the server.

S303: Identify the questioning intent of the target user by using the question description information, and obtain an intent identification result.

S304. Convert the intent recognition result into a target question and answer instruction.

S305. Match out a target reply template for the target user according to the service information set and the target question answering instruction.

S306: Obtain reply content for the target user according to the target reply template, and output the reply content through the first terminal.

Wherein, step S303-step S306 may refer to step S103-step S106, which will not be repeated here.

In one embodiment, if the aforementioned problem description information does not include problem context information, the server may initialize a problem context information, and send the problem context information to the first terminal. If the question description information includes question context information, update the saved question context information, and send the reply content and the updated question context information (including the ID of the new question and answer request and the updated question and answer round) to the first terminal, so that the user terminal submits a new question and answer request according to the updated question context information.

It can be seen that in the embodiment shown in FIG. 3, the server can obtain the question and answer request submitted by the first terminal, and the question and answer request carries the user information and question description information of the target user, and the user information includes user identification information and business system indication information; the server Obtain the business information set of the target user from the target business system according to the user identification information, and the target business system is the business system indicated by the business system indication information; the server uses the problem description information to identify the questioning intent of the target user, obtains the intent identification result, and Converting the intent recognition result into a target question and answer instruction, so as to match a target reply template for the target user according to the business information set and the target question answer instruction, and obtain the reply content for the target user according to the target reply template, and output the reply content through the first terminal, This process can effectively connect the corresponding business system according to the actual question and answer business in the question and answer process to obtain the user's business information set. quality.

This application involves blockchain technology. For example, the question and answer request may be sent by the first terminal to the server through the blockchain network, and correspondingly, the content of the reply may also be sent to the first terminal through the blockchain network. Or, the server can write user information, question description information, and reply content into the blockchain. These data written to the blockchain can be used for subsequent question-and-answer service quality determination, or subsequent troubleshooting.

Please refer to FIG. 4 , which is a schematic structural diagram of an artificial intelligence-based question answering device according to an embodiment of the present application. The question answering device can be applied to the server. Specifically, the question and answer apparatus may include: an acquisition module 401, configured to acquire a question and answer request submitted by a first terminal, where the question and answer request carries user information and question description information of a target user. The obtaining module 401 is further configured to obtain the service information set of the target user according to the user information. The identification module 402 is configured to identify the questioning intention of the target user by using the question description information, and obtain an intention identification result. The conversion module 403 is configured to convert the intent recognition result into a target question and answer instruction. The matching module 404 is configured to match a target reply template for the target user according to the service information set and the target question answering instruction. The obtaining module 401 is further configured to obtain the reply content for the target user according to the target reply template. The communication module 405 is configured to output the reply content through the first terminal.

In an optional implementation manner, the user information includes user identification information and service system indication information, and the obtaining module 401 obtains the service information set of the target user according to the user information, specifically according to the user identification information The service information set of the target user is acquired from the target service system, where the target service system is the service system indicated by the service system indication information.

In an optional implementation manner, the obtaining module 401 obtains the service information set of the target user from the target service system according to the user identification information, specifically querying the service information set for accessing the target user according to the service system indication information The link address of the target business system; perform splicing processing on the user identification information and the link address to obtain a spliced link address; obtain the target user's business information from the target business system according to the spliced link address gather.

In an optional implementation manner, the question and answer apparatus further includes a search module 406 and a storage module 407 .

In an optional implementation manner, the search module 406 is configured to search for a first plug-in template, where the first plug-in template includes page code of a link address configuration page.

In an optional implementation manner, the communication module 405 is further configured to send the first plug-in template to the target service system, so that the target service system is configured to access the target service system based on the first plug-in template. Describe the link address of the target business system.

In an optional implementation manner, the communication module 405 is further configured to receive a link address returned by the target service system for accessing the target service system.

In an optional implementation manner, the storage module 407 is configured to store the link address for accessing the target service system.

In an optional implementation manner, the matching module 404 matches a target reply template for the target user according to the business information set and the target question and answer instruction, and specifically acquires a conditional expression corresponding to the target question and answer instruction; Substitute each service information in the service information set into the conditional expression for operation to obtain an operation result; determine a reply template corresponding to the operation result, and determine the reply template corresponding to the operation result as a target reply template.

In an optional implementation manner, the communication module 405 is further configured to output a first page through the second terminal, where the first page includes multiple question and answer instructions, and the multiple question and answer instructions include the target question and answer instruction, And after detecting the selection instruction for the target question and answer instruction sent by the second terminal, output a second page through the second terminal, so that the second terminal configures the target question and answer instruction corresponding to the second page based on the second page. conditional expressions and reply templates.

In an optional implementation manner, the obtaining module 401 is further configured to obtain a conditional expression and a reply template corresponding to the target question and answer instruction.

In an optional implementation manner, the storage module 407 is further configured to save a conditional expression and a reply template configured by the second terminal for the target question answering instruction.

In an optional implementation manner, the communication module 405 outputting the second page through the second terminal includes: searching through the search module 406 for a second plug-in template, where the second plug-in template includes a page of the second page code; sending the second plug-in template to the second terminal, so that the second terminal generates the second page according to the second plug-in template, and outputs the second page.

It can be seen that in the embodiment shown in FIG. 4 , the question and answer device can obtain the question and answer request submitted by the first terminal, and the question and answer request carries the user information and question description information of the target user; the question and answer device can obtain the target user's business information set according to the user information, And use the question description information to obtain the intent recognition result about the target user; the question answering device can convert the intent recognition result into a target question and answer instruction, and match a target reply template for the target user according to the business information set and the target question and answer instruction, so that according to the target answer template Obtaining the reply content for the target user, and outputting the reply content through the first terminal, this process can effectively improve the quality of the question and answer service.

Please refer to FIG. 5 , which is a schematic structural diagram of a server according to an embodiment of the present application. The server described in this embodiment may include: one or more processors 1000 and a memory 2000 . The processor 1000 and the memory 2000 may be connected through a bus or the like.

The processor 1000 may be a central processing unit (Central Processing Unit, CPU), and the processor may also be other general-purpose processors, digital signal processors (Digital Signal Processors, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC) , Off-the-shelf programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The memory 2000 may be a high-speed RAM memory, or a non-volatile memory (non-volatile memory), such as a disk memory. The memory 2000 is used to store a set of program codes, and the processor 1000 can call the program codes stored in the memory 2000 . Specifically: the processor 1000 is configured to obtain a question and answer request submitted by the first terminal, where the question and answer request carries the user information and question description information of the target user; obtain the service information set of the target user according to the user information; The question description information is used to identify the questioning intention of the target user, and an intention identification result is obtained; the intention identification result is converted into a target question and answer instruction; according to the business information set and the target question answer instruction, the target user is matched outputting a target reply template; obtaining reply content for the target user according to the target reply template, and outputting the reply content through the first terminal.

In one embodiment, the server described in this application may also include a communication interface (not shown in FIG. 5 ). The communication interface can be a standard wired or wireless communication interface.

In one embodiment, the processor 1000 may send the reply content to the first terminal through the communication interface, so that the first terminal displays the reply content.

In one embodiment, the user information includes user identification information and service system indication information, and the processor 1000 obtains the target user's service information set according to the user information, specifically, according to the user identification information from the target service system The service information set of the target user is obtained in the target service system, and the target service system is the service system indicated by the service system indication information.

In one embodiment, the processor 1000 obtains the service information set of the target user from the target service system according to the user identification information, and specifically queries the service information set for accessing the target service system according to the service system indication information. link address; perform splicing processing on the user identification information and the link address to obtain a spliced link address; obtain the target user's service information set from the target service system according to the spliced link address.

In one embodiment, the processor 1000 is further configured to search for a first plug-in template, where the first plug-in template includes page code of a link address configuration page; send the first plug-in template to the target business system, so that The target business system configures a link address for accessing the target business system based on the first plug-in template; receives the link address returned by the target business system for accessing the target business system, and saves the link address for accessing the target business system. The link address for accessing the target business system. In one embodiment, the processor 1000 may send the first plug-in template to the target service system through a communication interface, and may also receive a message returned by the target service system for accessing the target service system through the communication interface. link address. In one embodiment, the processor 1000 may store the link address for accessing the target service system through the memory 2000 .

In one embodiment, the processor 1000 matches a target reply template for the target user according to the service information set and the target question and answer instruction, specifically acquiring a conditional expression corresponding to the target question and answer instruction; Substitute each service information in the information set into the conditional expression for operation to obtain an operation result; determine a reply template corresponding to the operation result, and determine the reply template corresponding to the operation result as a target reply template.

In one embodiment, the processor 1000 is further configured to output a first page through the second terminal, where the first page includes multiple question and answer instructions, and the multiple question and answer instructions include the target question and answer instruction; After the second terminal sends the selection instruction for the target question and answer instruction, the second page is output through the second terminal, so that the second terminal configures the conditional expression and the corresponding conditional expression of the target question and answer instruction based on the second page. Reply template; obtain the conditional expression and reply template corresponding to the target question and answer instruction, and save the conditional expression and reply template configured by the second terminal for the target question and answer instruction. In one embodiment, the processor 1000 can send the first page to the second terminal through the communication interface, so that the second terminal can display the first page, and can also send the second page to the second terminal through the communication interface, so that the second terminal can display the first page. The second terminal displays the second page. In one embodiment, the processor 1000 may store the conditional expression and the reply template of the target question-answer quality configuration through the memory 2000 .

In one embodiment, the processor 1000 outputs the second page through the second terminal, specifically searching for a second plug-in template, where the second plug-in template includes page code of the second page; The template is sent to the second terminal, so that the second terminal generates the second page according to the second plug-in template, and outputs the second page.

In specific implementation, the processor 1000 described in the embodiments of the present application may perform the implementation manners described in the embodiments of FIG. 1 and FIG. 3 , and may also perform the implementation manners described in the embodiments of the present application, which will not be repeated here. .

Each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically alone, or two or more modules may be integrated into one module. The above-mentioned integrated modules can be implemented in the form of sampling hardware or in the form of sampling software function modules.

Embodiments of the present application further provide a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the methods in the foregoing embodiments can be implemented, or the computer program is When executed by the processor, the functions of each module of the apparatus in the foregoing embodiment can be implemented, which will not be repeated here. Optionally, the storage medium involved in this application, such as a computer-readable storage medium, may be non-volatile or volatile.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. During execution, the processes of the embodiments of the above-mentioned methods may be included. Wherein, the computer-readable storage medium can be volatile or non-volatile. For example, the computer storage medium may be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM), or a random access memory (Random Access Memory, RAM), or the like. The computer-readable storage medium may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function, and the like; Use the created data, etc.

Among them, the blockchain referred to in this application is a new application mode of computer technology such as distributed data storage, point-to-point transmission, consensus mechanism, and encryption algorithm. Blockchain, essentially a decentralized database, is a series of data blocks associated with cryptographic methods. Each data block contains a batch of network transaction information to verify its Validity of information (anti-counterfeiting) and generation of the next block. The blockchain can include the underlying platform of the blockchain, the platform product service layer, and the application service layer.

What is disclosed above is only a preferred embodiment of the present application, and of course, it cannot limit the scope of the right of the present application. Those skilled in the art can understand that all or part of the process of implementing the above-mentioned embodiment can be realized according to the right of the present application. The equivalent changes required to be made still fall within the scope covered by this application.

Claims

An artificial intelligence-based question answering method including:

obtaining a question and answer request submitted by the first terminal, where the question and answer request carries the user information and question description information of the target user;

Acquire the service information set of the target user according to the user information;

Identify the questioning intent of the target user by using the question description information, and obtain an intent identification result;

converting the intent recognition result into a target question answering instruction;

Matching a target reply template for the target user according to the service information set and the target question answering instruction;

The reply content for the target user is obtained according to the target reply template, and the reply content is output through the first terminal.
The method according to claim 1, wherein the user information includes user identification information and service system indication information, and the acquiring the service information set of the target user according to the user information includes:

The service information set of the target user is acquired from the target service system according to the user identification information, where the target service system is the service system indicated by the service system indication information.
The method according to claim 2, wherein the acquiring the service information set of the target user from the target service system according to the user identification information comprises:

Query the link address for accessing the target business system according to the business system indication information;

Perform splicing processing on the user identification information and the link address to obtain a spliced link address;

Acquire the service information set of the target user from the target service system according to the spliced link address.
The method of claim 3, wherein the method further comprises:

Find a first plug-in template, where the first plug-in template includes the page code of the link address configuration page;

sending the first plug-in template to the target service system, so that the target service system configures a link address for accessing the target service system based on the first plug-in template;

The link address for accessing the target business system returned by the target business system is received, and the link address for accessing the target business system is saved.
The method according to claim 1, wherein the matching a target reply template for the target user according to the service information set and the target question answering instruction comprises:

obtaining the conditional expression corresponding to the target question and answer instruction;

Substitute each service information in the service information set into the conditional expression to perform an operation to obtain an operation result;

A reply template corresponding to the operation result is determined, and the reply template corresponding to the operation result is determined as a target reply template.
The method of claim 5, wherein the method further comprises:

Outputting a first page through the second terminal, where the first page includes multiple question-and-answer instructions, and the multiple question-and-answer instructions include the target question-and-answer instruction;

After detecting the selection instruction for the target question and answer instruction sent by the second terminal, output a second page through the second terminal, so that the second terminal configures the target question and answer instruction corresponding to the second page based on the second page. Conditional expressions and reply templates;

Acquire a conditional expression and a reply template corresponding to the target question-and-answer instruction, and save the conditional expression and reply template configured by the second terminal for the target question-and-answer instruction.
The method according to claim 6, wherein the outputting the second page through the second terminal comprises:

Find a second plug-in template, where the second plug-in template includes the page code of the second page;

Sending the second plug-in template to the second terminal, so that the second terminal generates the second page according to the second plug-in template, and outputs the second page.
An artificial intelligence-based question answering device, comprising:

an obtaining module, configured to obtain a question and answer request submitted by the first terminal, where the question and answer request carries the user information and question description information of the target user;

The obtaining module is further configured to obtain the service information set of the target user according to the user information;

an identification module, configured to identify the questioning intention of the target user by using the question description information to obtain an intention identification result;

a conversion module for converting the intent recognition result into a target question answering instruction;

a matching module, configured to match a target reply template for the target user according to the service information set and the target question and answer instruction;

The obtaining module is further configured to obtain the reply content for the target user according to the target reply template;

A communication module, configured to output the reply content through the first terminal.
A server includes a processor and a memory, the processor and the memory being interconnected, wherein the memory is used to store a computer program, the computer program including program instructions, the processor is configured to invoke the The following program instructions are executed:

obtaining a question and answer request submitted by the first terminal, where the question and answer request carries the user information and question description information of the target user;

Acquire the service information set of the target user according to the user information;

Identify the questioning intent of the target user by using the question description information, and obtain an intent identification result;

converting the intent recognition result into a target question answering instruction;

Matching a target reply template for the target user according to the service information set and the target question answering instruction;

The reply content for the target user is obtained according to the target reply template, and the reply content is output through the first terminal.
The server according to claim 9, wherein executing the user information includes user identification information and service system indication information, and acquiring the target user's service information set according to the user information includes:

The service information set of the target user is acquired from the target service system according to the user identification information, where the target service system is the service system indicated by the service system indication information.
The server according to claim 10, wherein performing the acquiring the service information set of the target user from the target service system according to the user identification information comprises:

Query the link address for accessing the target business system according to the business system indication information;

Perform splicing processing on the user identification information and the link address to obtain a spliced link address;

Acquire the service information set of the target user from the target service system according to the spliced link address.
The server of claim 11, wherein the processor is further configured to perform:

Find a first plug-in template, where the first plug-in template includes the page code of the link address configuration page;

sending the first plug-in template to the target service system, so that the target service system configures a link address for accessing the target service system based on the first plug-in template;

The link address for accessing the target business system returned by the target business system is received, and the link address for accessing the target business system is saved.
The server according to claim 9, wherein performing the matching of the target response template for the target user according to the service information set and the target question answering instruction comprises:

obtaining the conditional expression corresponding to the target question and answer instruction;

Substitute each service information in the service information set into the conditional expression to perform an operation to obtain an operation result;

A reply template corresponding to the operation result is determined, and the reply template corresponding to the operation result is determined as a target reply template.
The server of claim 13, wherein the processor is further configured to perform:

Outputting a first page through the second terminal, where the first page includes multiple question-and-answer instructions, and the multiple question-and-answer instructions include the target question-and-answer instruction;

After detecting the selection instruction for the target question and answer instruction sent by the second terminal, output a second page through the second terminal, so that the second terminal configures the target question and answer instruction corresponding to the second page based on the second page. Conditional expressions and reply templates;

Acquire a conditional expression and a reply template corresponding to the target question-and-answer instruction, and save the conditional expression and reply template configured by the second terminal for the target question-and-answer instruction.
A computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the following method:

obtaining a question and answer request submitted by the first terminal, where the question and answer request carries the user information and question description information of the target user;

Acquire the service information set of the target user according to the user information;

Identify the questioning intent of the target user by using the question description information, and obtain an intent identification result;

converting the intent recognition result into a target question answering instruction;

Matching a target reply template for the target user according to the service information set and the target question answering instruction;

The reply content for the target user is obtained according to the target reply template, and the reply content is output through the first terminal.
The computer-readable storage medium according to claim 15, wherein executing the user information includes user identification information and service system indication information, and acquiring the target user's service information set according to the user information includes:

The service information set of the target user is acquired from the target service system according to the user identification information, where the target service system is the service system indicated by the service system indication information.
The computer-readable storage medium according to claim 16, wherein performing the acquiring the target user's service information set from the target service system according to the user identification information comprises:

Query the link address for accessing the target business system according to the business system indication information;

Perform splicing processing on the user identification information and the link address to obtain a spliced link address;

Acquire the service information set of the target user from the target service system according to the spliced link address.
The computer-readable storage medium of claim 17, wherein the computer program, when executed by the processor, is further configured to:

Find a first plug-in template, where the first plug-in template includes the page code of the link address configuration page;

sending the first plug-in template to the target service system, so that the target service system configures a link address for accessing the target service system based on the first plug-in template;

The link address for accessing the target business system returned by the target business system is received, and the link address for accessing the target business system is saved.
The computer-readable storage medium according to claim 15, wherein performing the matching to the target user according to the business information set and the target question answering instruction to obtain a target reply template, comprising:

obtaining the conditional expression corresponding to the target question and answer instruction;

Substitute each service information in the service information set into the conditional expression to perform an operation to obtain an operation result;

A reply template corresponding to the operation result is determined, and the reply template corresponding to the operation result is determined as a target reply template.
The computer-readable storage medium of claim 19, wherein the computer program, when executed by the processor, is further configured to:

Outputting a first page through the second terminal, where the first page includes multiple question-and-answer instructions, and the multiple question-and-answer instructions include the target question-and-answer instruction;

After detecting the selection instruction for the target question and answer instruction sent by the second terminal, output a second page through the second terminal, so that the second terminal configures the target question and answer instruction corresponding to the second page based on the second page. Conditional expressions and reply templates;

Acquire a conditional expression and a reply template corresponding to the target question-and-answer instruction, and save the conditional expression and reply template configured by the second terminal for the target question-and-answer instruction.