CN112417122A

CN112417122A - Multi-turn question and answer method and device, computer readable storage medium and electronic equipment

Info

Publication number: CN112417122A
Application number: CN202011311513.6A
Authority: CN
Inventors: 关春生; 张晓清; 纪思阳
Original assignee: Taikang Insurance Group Co Ltd
Current assignee: Taikang Insurance Group Co Ltd
Priority date: 2020-11-20
Filing date: 2020-11-20
Publication date: 2021-02-26

Abstract

The disclosure relates to the field of intelligent question answering, and provides a multi-turn question answering method, a multi-turn question answering device, a computer readable storage medium and electronic equipment, wherein the method comprises the following steps: receiving session information of a user, reading a first instruction identifier corresponding to the session information from a text conversation instruction list, and configuring first question-answer information according to the first instruction identifier; receiving key information of a user aiming at the first question-answering information, configuring matched object information in each object information corresponding to the key information and the first instruction identification as target object information, and configuring second question-answering information according to the target object information and the second instruction identification, wherein the second instruction identification is executed after the first instruction identification; and repeating the steps until a result instruction identifier in the text conversation instruction list is obtained, and configuring target answer information according to the result instruction identifier. The method and the system can process multiple rounds of question answering processes and decision process branches by using the text dialogue instruction list, and improve the question answering efficiency of the question answering robot.

Description

Multi-turn question and answer method and device, computer readable storage medium and electronic equipment

Technical Field

The present disclosure relates to the field of intelligent question answering technologies, and in particular, to a multi-turn question answering method, a multi-turn question answering device, a computer-readable storage medium, and an electronic device.

Background

With the wide application of intelligent robots in various fields, a multi-round interaction method between a user and the intelligent robot has received more and more attention.

The existing intelligent question-answering task flow configuration depends on the recognition of skill word slots, the types and the number of the skill word slots are limited by the system construction condition, and if the skill word slots in the user question sentence cannot be matched with enough service parameters, the service flow cannot be completed.

In view of the above, there is a need in the art to develop a new multi-turn question-answering method and apparatus.

It is to be noted that the information disclosed in the above background section is only for enhancement of understanding of the background of the present disclosure, and thus may include information that does not constitute prior art known to those of ordinary skill in the art.

Disclosure of Invention

The present disclosure is directed to a multi-turn question and answer method, a multi-turn question and answer apparatus, a computer readable storage medium, and an electronic device, so as to improve the question and answer efficiency of the multi-turn question and answer at least to some extent.

Additional features and advantages of the disclosure will be set forth in the detailed description which follows, or in part will be obvious from the description, or may be learned by practice of the disclosure.

According to one aspect of the present disclosure, there is provided a multi-turn question answering method, the method comprising: receiving session information of a user, reading a first instruction identification corresponding to the session information in a text conversation instruction list, and configuring first question-answer information according to the first instruction identification, wherein the text conversation instruction list is configured with a plurality of instruction identifications with execution sequences and one or more object information corresponding to each instruction identification in a text file form; receiving key information of the user aiming at the first question and answer information, configuring matched object information in the key information and each object information corresponding to the first instruction identification as target object information, and configuring second question and answer information according to the target object information and a second instruction identification, wherein the second instruction identification is executed after the first instruction identification; and repeating the steps until a result instruction identification in the text conversation instruction list is obtained, and configuring target answer information according to the result instruction identification.

In some exemplary embodiments of the present disclosure, reading a first instruction identifier corresponding to session information in a text dialog instruction table includes: obtaining the context information of the session information, determining step information according to the session information and the context information, and reading the first instruction identification corresponding to the step information in the text dialogue instruction list according to the step information.

In some exemplary embodiments of the present disclosure, configuring first question and answer information according to the first instruction identification includes: and acquiring the first question-answer information corresponding to the first instruction identification in a question-answer data table according to the first instruction identification, and sending the first question-answer information to a terminal corresponding to the user.

In some exemplary embodiments of the present disclosure, the question-answer data table includes a question-answer data table and a base data table; configuring second question-answering information according to the target object information and a second instruction identifier, wherein the second question-answering information comprises: and obtaining question-answering information corresponding to the second instruction identification in the question-answering data table, obtaining one or more preset answer information corresponding to the target object information in the basic data table, and configuring the second question-answering information according to the question-answering information and the one or more preset answer information.

In some exemplary embodiments of the present disclosure, identifying configuration target answer information according to the result instruction includes: and acquiring key information corresponding to each instruction identifier, and configuring the target answer information according to the key information corresponding to each instruction identifier and the result instruction identifier.

In some exemplary embodiments of the present disclosure, configuring, as target object information, object information that matches the key information with object information corresponding to the first instruction identifier, includes: matching the key information with one or more object information corresponding to the first instruction identification; and when the key information is matched with the object information corresponding to the first instruction identification, configuring the object information matched with the key information as the target object information.

In some exemplary embodiments of the present disclosure, the method further comprises: when the key information is not matched with the object information corresponding to the first instruction identification, matching the key information with an irrelevant object list of the first instruction identification in a question-answer data table; if yes, configuring target question-answering information into first preset answering information related to the first instruction identification; and if not, configuring the target question-answering information into second preset answering information related to the first instruction identification.

According to an aspect of the present disclosure, there is provided a multi-turn question-answering device including: the system comprises a first question-answer information configuration module, a first question-answer information configuration module and a first question-answer information configuration module, wherein the first question-answer information configuration module is used for receiving conversation information of a user, reading a first instruction identification corresponding to the conversation information in a text conversation instruction list, and configuring first question-answer information according to the first instruction identification, and the text conversation instruction list is configured with a plurality of instruction identifications with execution sequences and one or more object information corresponding to each instruction identification in a text file form; a second question and answer information configuration module, configured to receive key information of the user for the first question and answer information, configure object information that matches the key information with object information corresponding to the first instruction identifier as target object information, and configure second question and answer information according to the target object information and a second instruction identifier, where the second instruction identifier is executed after the first instruction identifier; and the target response information configuration module is used for repeating the steps until a result instruction identifier in the text conversation instruction list is obtained, and configuring target response information according to the result instruction identifier.

According to an aspect of the present disclosure, there is provided a computer readable medium, on which a computer program is stored, which when executed by a processor, implements the multi-round question-answering method as described in the above embodiments.

According to an aspect of the present disclosure, there is provided an electronic device including: one or more processors; a storage device for storing one or more programs that, when executed by the one or more processors, cause the one or more processors to implement a multi-round question-answering method as described in the above embodiments.

According to the technical scheme, the multi-turn question answering method and device, the computer readable storage medium and the electronic device in the exemplary embodiment of the disclosure have at least the following advantages and positive effects:

the multi-turn question answering method comprises the steps of firstly receiving session information of a user, reading a first instruction identification corresponding to the session information in a text conversation instruction list, and configuring the first question answering information according to the first instruction identification, wherein the text conversation instruction list is configured with a plurality of instruction identifications with execution sequences and one or more object information corresponding to each instruction identification in a text file form; secondly, receiving key information of a user aiming at the first question and answer information, configuring matched object information in each object information corresponding to the key information and the first instruction identification as target object information, and configuring second question and answer information according to the target object information and the second instruction identification, wherein the second instruction identification is executed after the first instruction identification; and repeating the steps until a result instruction identifier in the text conversation instruction list is obtained, and configuring target answer information according to the result instruction identifier. On one hand, the multi-turn question-answering method disclosed by the disclosure configures a text conversation instruction list in a text file form, and a word slot filling method is not needed, so that the operability of the multi-turn question-answering flow configuration is enhanced, a robot trainer can quickly master the configuration mode of multi-turn interaction of a task flow, and the configuration efficiency is improved; on the other hand, the text dialogue instruction list is used for processing multiple rounds of question-answering processes and decision process branches, so that the working efficiency of the question-answering robot is improved; on the other hand, complex task flow multi-turn interaction logic can be built through the basic logic unit in the text conversation instruction list, rapid configuration and verification of complex business scenes are achieved, and business innovation landing efficiency is improved.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and together with the description, serve to explain the principles of the disclosure. It is to be understood that the drawings in the following description are merely exemplary of the disclosure, and that other drawings may be derived from those drawings by one of ordinary skill in the art without the exercise of inventive faculty.

FIG. 1 schematically illustrates a flow diagram of a multi-round question answering method according to an embodiment of the present disclosure;

FIG. 2 schematically illustrates a structural schematic of a multi-turn question-answering system according to an embodiment of the present disclosure;

FIG. 3 schematically illustrates a process flow diagram of a multi-round question answering method according to an embodiment of the present disclosure;

FIG. 4 schematically illustrates a flow diagram of a multi-round question answering method, in accordance with a specific embodiment of an embodiment of the present disclosure;

FIG. 5 is a diagram schematically illustrating a structure of a text dialog instruction table corresponding to a query hesitation period according to an embodiment of the present disclosure;

FIG. 6 schematically illustrates a structural diagram of an exclusion data table corresponding to query hesitation periods, according to an embodiment of the present disclosure;

FIG. 7 is a schematic diagram illustrating a structure of a question-back data table corresponding to query hesitation periods according to an embodiment of the present disclosure;

FIG. 8 schematically illustrates an interface diagram of multiple rounds of quiz answers corresponding to query hesitation periods, according to an embodiment of the present disclosure;

FIG. 9 is a schematic structural diagram of a text dialog instruction table corresponding to a universal supplementary scene according to an embodiment of the disclosure;

FIG. 10 is a schematic diagram illustrating the structure of an exclusion data table corresponding to a universal append scene in accordance with an embodiment of the present disclosure;

FIG. 11 is a schematic diagram illustrating a structure of a question-backing data table corresponding to a universal supplementary scenario according to an embodiment of the present disclosure;

FIG. 12 is a schematic diagram illustrating a structure of a base data table corresponding to a universal append scene in accordance with an embodiment of the present disclosure;

FIG. 13 schematically illustrates an interface diagram of multiple rounds of question answering in a omnipotent append scenario, according to one embodiment of the present disclosure;

FIG. 14 schematically illustrates a block diagram of a multi-turn question answering device, in accordance with an embodiment of the present disclosure;

FIG. 15 schematically shows a block schematic diagram of an electronic device according to an embodiment of the present disclosure;

FIG. 16 schematically shows a program product schematic according to an embodiment of the disclosure.

Detailed Description

Example embodiments will now be described more fully with reference to the accompanying drawings. Example embodiments may, however, be embodied in many different forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art.

Furthermore, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the disclosure. One skilled in the relevant art will recognize, however, that the subject matter of the present disclosure can be practiced without one or more of the specific details, or with other methods, components, devices, steps, and so forth. In other instances, well-known methods, devices, implementations, or operations have not been shown or described in detail to avoid obscuring aspects of the disclosure.

The block diagrams shown in the figures are functional entities only and do not necessarily correspond to physically separate entities. I.e. these functional entities may be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different networks and/or processor means and/or microcontroller means.

The flow charts shown in the drawings are merely illustrative and do not necessarily include all of the contents and operations/steps, nor do they necessarily have to be performed in the order described. For example, some operations/steps may be decomposed, and some operations/steps may be combined or partially combined, so that the actual execution sequence may be changed according to the actual situation.

In the related art in the field, the question and answer flow is mainly implemented by using a word slot filling method and a software programming method. Specifically, the word slot filling method is a method for matching a skill word slot with a service parameter, the skill word slot is identified from a question, the service parameter in the question is obtained through a service parameter matching relationship, corresponding target information is called from a target service system through the service parameter, and a target sentence corresponding to the question is generated. However, the task flow configuration of the method depends on the recognition of the skill word slot, the type and the number of the skill word slot are restricted by the system construction condition, the number of the general word slot such as time, place, number, certificate number and the like is limited, and the requirement of complex business flow can not be met; the special word slots, such as product names, business terms and the like, are customized and developed, and can not meet the requirement that all the special word slots can be effectively identified. The software programming method needs more code development work when the task flow is configured, updated or modified, needs a robot trainer and a software engineer to complete the code development work in a matching way, is low in development efficiency of a new flow service scene, needs a certain debugging time for testing and deploying the code processing task flow, and influences deployment progress of the task flow due to defects of the code and processing logic.

Based on the problems in the related art, in an embodiment of the present disclosure, a multi-round question answering method is provided, and fig. 1 shows a flow diagram of the multi-round question answering method, as shown in fig. 1, the multi-round question answering method at least includes the following steps:

step S110: receiving session information of a user, reading a first instruction identification corresponding to the session information in a text conversation instruction list, and configuring first question and answer information according to the first instruction identification, wherein the text conversation instruction list is configured with a plurality of instruction identifications with execution sequences and one or more object information corresponding to each instruction identification in a text file form;

step S120: receiving key information of a user aiming at the first question-answering information, configuring matched object information in each object information corresponding to the key information and the first instruction identification as target object information, and configuring second question-answering information according to the target object information and the second instruction identification, wherein the second instruction identification is executed after the first instruction identification;

step S130: and repeating the steps until a result instruction identifier in the text conversation instruction list is obtained, and configuring target answer information according to the result instruction identifier.

According to the multi-turn question-answering method in the embodiment of the disclosure, on one hand, a text conversation instruction list is configured in a text file form, a word slot filling method is not needed, so that the operability of the multi-turn question-answering flow configuration is enhanced, a robot trainer can quickly master the configuration mode of multi-turn interaction of a task flow, and the configuration efficiency is improved; on the other hand, the text dialogue instruction list is used for processing multiple rounds of question-answering processes and decision process branches, so that the working efficiency of the question-answering robot is improved; on the other hand, complex task flow multi-turn interaction logic can be built through the basic logic unit in the text conversation instruction list, rapid configuration and verification of complex business scenes are achieved, and business innovation landing efficiency is improved.

It should be noted that the multi-turn question-answering method of the exemplary embodiment of the present disclosure is applied to a multi-turn question-answering system, fig. 2 shows a schematic structural diagram of the multi-turn question-answering system in the exemplary embodiment, as shown in fig. 2, the multi-turn question-answering system includes a user terminal, a robot trainer, a question-answering robot, and an external system, and a user sends session information and reply information to the question-answering robot through the user terminal and receives the question-answering information and reply information sent by the robot; the question-answering robot configures question-answering information and answering information based on a text dialogue instruction table and a question-answering data table by analyzing session information and answering information sent by a user terminal; the robot trainer configures the text dialogue instruction table and the question-answer data table in a text file form; when the question-answering robot needs to use the user information and/or the product information, a data interface of an external system needs to be called, and the user information and/or the product information are obtained through the data interface, so that the configuration of the question-answering information and the answering information is completed.

In order to make the technical scheme of the present disclosure clearer, each step of the multi-round question answering method is explained next.

In step S110, session information of a user is received, a first instruction identifier corresponding to the session information is read from a text dialog instruction table, and first question and answer information is configured according to the first instruction identifier, where the text dialog instruction table is configured with a plurality of instruction identifiers having an execution sequence and one or more object information corresponding to each instruction identifier in a text file form.

In an exemplary embodiment of the present disclosure, the session information includes a question text transmitted by the user on the terminal and user identification information including a user ID and/or a session ID.

In an exemplary embodiment of the present disclosure, the text dialog instruction table may be configured by the robot trainer in a text file manner, and in particular, may be configured by an EXCEL or WPS table. The text dialog instruction table includes a plurality of instruction identifiers having execution sequences and one or more pieces of object information corresponding to each instruction identifier, and the present disclosure does not specifically limit the number of instruction identifiers and the pieces of object information corresponding to each instruction identifier in the text dialog instruction table.

For example, table 1 shows an example table of a text dialog instruction table of an embodiment, as shown in table 1, the text dialog instruction table includes four instruction identifiers having an execution sequence, and object information corresponding to each instruction identifier, where the instruction identifier includes an instruction name and a parameter, the instruction name may be a logic processing type of the current round, the parameter may be an object of the logic processing, and the object information may be used by a user to determine a subsequent branch flow.

Table 1 text dialog instruction sheet example

The four instruction identifications in the text dialog instruction table in table 1 are instruction 1, instruction 2, instruction 3, and answer instructions, respectively. The instruction 1 is an ask { product name } instruction, and object information corresponding to the instruction comprises a product 1, a product 2 and other information; the instruction 2 is a cross { sales channel } instruction, the object information corresponding to the instruction and the product 1 is a channel 1 and a channel 2, and the object information corresponding to the product 2 is a channel 1 and a channel 3; the instruction 3 is an ask { city } instruction, the object information corresponding to each channel in the instruction is the same as that in the previous instruction, and the instruction comprises Beijing, Shanghai and the like; the last instruction identifier is a result instruction identifier, and corresponds to the answer instruction in fig. 2, and the answer instruction includes specific answer information. The ask { product name } instruction in the text dialog instruction list further includes a find { exclusion list } instruction for determining whether the product name replied by the user belongs to a product in the exclusion list.

It should be noted that, changes in the number of command identifiers and object information, or changes in the number of times of multiple rounds of question answering processes due to changes in the number of command identifiers and object information, all belong to the protection scope of the present disclosure.

Three instruction names are defined in the text dialog instruction table in table 1, and the three instruction names respectively represent the following types of logical processing, as shown in table 2:

TABLE 2 instruction name exemplary description

The number, type and actual meaning of the instruction names may be set according to actual conditions, which is not specifically limited in this disclosure.

The multi-turn question-answering method can be applied to a plurality of task scenes, each task scene can correspond to one task number, for example, the task can be a hesitation period query scene in the insurance field, also can be a premium query scene and the like, each task scene corresponds to one task number, and the method is not particularly limited in the disclosure.

In the exemplary embodiment of the disclosure, after receiving the session information of the user, the task number corresponding to the session information is determined according to the session information of the user, and the text dialogue instruction table corresponding to the task number is obtained according to the task number. That is, each task corresponds to one text conversation instruction table, if the task types are the same, a plurality of tasks may correspond to one text conversation instruction table, or one task may correspond to a plurality of text conversation instruction tables, which may be specifically set according to actual situations, and this disclosure does not specifically limit this.

In an exemplary embodiment of the present disclosure, the multi-turn question and answer method is applied to a distributed multi-turn question and answer system, which includes a plurality of servers among which a plurality of multi-turn question and answer tasks are distributed. And storing a plurality of interactive information of the robot and the user in a plurality of multi-turn question and answer tasks on a common server. Therefore, after receiving the session message of the user, the context information of the session information is obtained first, the step information is determined according to the session information and the context information, and the first instruction identification corresponding to the step information is read in the text dialog instruction table according to the step information. The context information may include robot question and answer information before the session information or information sent by the user, and may be based on the task number and/or user identification information.

Specifically, the instruction identifiers in the text dialogue instruction table have an execution sequence, and each instruction identifier in the text dialogue instruction table corresponds to one step information. After receiving the session message of the user, it needs to first determine which step the current multiple-round interaction of the user goes to according to the session information and the context information, and read the corresponding instruction identifier in the text dialog instruction table according to the step information.

In an exemplary embodiment of the present disclosure, the text dialog instruction table is configured with a question and answer data table, which may also be configured according to a text file form. The question-answer data table comprises question-answer operation corresponding to each instruction identification, the question-answer data table can comprise question-answer data tables, and the question-answer data table can comprise question-answer operation corresponding to each instruction identification in all text conversation instruction tables. The system can also comprise a plurality of question-answering data tables, and each question-answering data table comprises a question-answering operation corresponding to each instruction identification in each text conversation instruction table. The instructions in the unified text dialogue instruction list may also be included to represent corresponding question-reversing operation, which is not specifically limited by this disclosure. For example, the text dialogue instruction table in table 1 is configured with a question-backing data table shown in table 3, as shown in table 3:

table 3 example question-back data table

Instruction identification	Anti-interrogation technique
		ask { product name }	Asking you which product to purchase?
chose { sales channel }	Please select the sales channel of the product you buy:
		ask { City }	Ask you about which city you are buyingDo things?

In addition, the question answering data table can also comprise a basic data table, namely object information corresponding to each instruction identification is included, and the corresponding relation between the object information corresponding to each instruction identification is included. For example, the text dialog command table in table 1 is configured with a basic data table as shown in table 4, as shown in table 4:

TABLE 4 basic data sheet

Product name	Sale channel
		Product
1	Channel 1, channel 2
		Product 2	Channel 1, channel 3

Also, the question-answer data table may further include an exclusion data table corresponding to the exclusion list, that is, object information other than the object information included in the instruction identification in the text conversation instruction table. For example, the exclusion data table corresponding to the exclusion list in the text dialog instruction table in table 1 is shown in table 5:

TABLE 5 Elimination data Table

In an exemplary embodiment of the disclosure, first question and answer information corresponding to the first instruction identifier may be acquired in a question and answer data table according to the first instruction identifier, and the first question and answer information may be sent to a terminal corresponding to a user. Specifically, a question-answering operation corresponding to the first instruction identification is inquired in a question-answering data table according to the first instruction identification, and the question-answering operation corresponding to the first instruction identification is configured into first question-answering information. In addition, the second question-answering information may also be configured according to the method, that is, a question-answering operation corresponding to the second instruction identifier is queried in the question-answering data table according to the second instruction identifier, and the question-answering operation corresponding to the second instruction identifier is configured as the second question-answering information.

For example, if the session information of the user is "help me to search for product information", the first instruction identifier in the text dialog instruction table searched through the session information is ask { product name }, and the question-back operation corresponding to ask for which product you purchased in the question data table according to the first instruction identifier is "ask for which product? "will ask the converse" ask you which product you buy? And sending the key information to a client corresponding to the user so that the user replies the key information corresponding to the first question-answer information aiming at the question.

In step S120, key information of the user for the first question and answer information is received, object information that matches the key information with each object information corresponding to the first instruction identifier is configured as target object information, and second question and answer information is configured according to the target object information and the second instruction identifier, where the second instruction identifier is executed after the first instruction identifier.

In an exemplary embodiment of the present disclosure, the key information is key information in the first question and answer information bioreaction information of the user, and if the reply information of the user includes a plurality of pieces of information, the key information related to the first question and answer information in the plurality of pieces of information needs to be acquired. For example, if the user asks for what product you buy for the above? If the reply information of the first question answering information is "product 1", the acquired key information for the first question answering information is the product 1; if the reply information of the user is "product 1 of the product that I purchased", the key information needs to be extracted according to the reply information of the user.

In an exemplary embodiment of the present disclosure, the key information is matched with one or more object information corresponding to the first instruction identification, and when the key information is matched with the object information corresponding to the first instruction identification, the object information matched with the key information is configured as target object information.

In an exemplary embodiment of the present disclosure, the specific process of configuring the second question and answer information according to the target object information and the second instruction identifier may include: and obtaining question-answering information corresponding to the second instruction identification in the question-answering data table, obtaining one or more preset answer information corresponding to the target object information in the basic data table, and configuring the second question-answering information according to the question-answering information and the one or more preset answer information. The preset answer information comprises one or more object information corresponding to the key information in a second instruction identifier, and the second instruction identifier is a second instruction identifier behind the first instruction identifier in the text conversation instruction list. Of course, the first question and answer information may also be configured by the above method for configuring the second question and answer information, which is not specifically limited by the present disclosure.

For example, when the target object information is a product 1, first, a second instruction identifier chose { sales channel } is obtained from a text conversation instruction list; then, a question return operation corresponding to the second instruction identification is obtained in the question return data table according to the second instruction identification, namely that a selling channel of the product bought by you is selected: and acquiring preset answer information 'channel 1, channel 2' corresponding to the second instruction identifier from the basic data table according to the product 1; and finally, configuring second question and answer information 'please select a sale channel of the product bought by you' according to the question-answering operation corresponding to the second instruction identification and the preset answer information: 1. a channel 1; 2. channel 2 ".

In addition, after receiving the key information of the user for the first question and answer information, it is necessary to first determine which step the current multiple rounds of interaction of the user go to according to the key information, and read a corresponding instruction identifier in the text conversation instruction list according to the step information. When a conversation message or a reply message of the user is received, the steps can be carried out, and the continuity of multiple rounds of interaction is ensured.

In step S130, the above steps are repeated until the result instruction identifier in the text dialog instruction table is obtained, and the target answer information is configured according to the result instruction identifier.

In an exemplary embodiment of the present disclosure, key information corresponding to each instruction identifier is acquired, and target answer information is configured according to the key information corresponding to each instruction identifier and the result instruction identifier. Specifically, if the session information of the user is related to a first task, the robot completes acquisition of key information through multiple rounds of question answering between the question answering robot and the user by the acquired text dialogue instruction list corresponding to the first task, configures target answer information according to the acquired key information and an answer template corresponding to the result instruction identification, and sends the target answer information to the user terminal to complete multiple rounds of question answering.

In an exemplary embodiment of the present disclosure, when the key information does not match the object information corresponding to the first instruction identifier, matching the key information with an exclusion list of the first instruction identifier in an exclusion data table; if the key information is matched with the exclusion list, configuring the target answer information into first preset answer information related to the first instruction identification; and if the key information does not match the exclusion list, configuring the target answer information into second preset answer information which is first related to the first instruction identification. The first preset answer information and the second preset answer information may be set according to an actual task scenario, which is not specifically limited by the present disclosure. For example, if the task scenario is query renewal time, and the first instruction is identified as ask { product name }, the first preset answer information may be configured as "hello, and the product purchased by you is not in the free renewal range. "you are good, and the product purchased by you meets the free renewal condition, and can be renewed for one year free. ".

In the exemplary embodiment of the present disclosure, if the key information does not match the exclusion list, the new task number and the text dialogue instruction list corresponding to the new task number may be determined again according to the key information, and the multi-round question-answering method in the above embodiment may be repeated.

In multiple rounds of question-answer interaction between the question-answer robot and the user, the question-answer robot configures question-answer information and sends the question-answer information to the user terminal, and the user gives key information according to the question-answer information, namely, one question-answer interaction is completed. The whole question-answer interaction process is controlled by the question-answer robot, if the question-answer robot sends question-answer information to the user terminal, and the reply information of the user is irrelevant to the question-answer information, the question-answer robot can repeatedly send the question-answer information until the reply information of the user contains key information.

In addition, if the reply information of the user is not related to the question and answer information, the task number can be matched again according to the current reply information, the text conversation instruction list can be determined again according to the current task number, and a plurality of new turn of question and answer tasks can be completed according to the text conversation instruction list.

In an exemplary embodiment of the present disclosure, if other information such as user information and product information is needed to be used for configuring question answering information or target answering information, the other information such as user information and product information may be acquired by calling a data interface of an external system, and the target answering information is configured according to the acquired external information and result instruction identifier.

In an exemplary embodiment of the present disclosure, fig. 3 shows a processing flow diagram of a multi-round question-answering method executed by a question-answering robot, including a user terminal, a text dialogue instruction list, a question-answering data list, and an external system. In the present embodiment, taking two rounds of question answering as an example, as shown in fig. 3, in step S301, session information sent by a user terminal is received; in step S302, a logic determination is performed in the text dialog instruction list according to the session information; in step S303, query question answering data in the question answering data table according to the logical judgment result; in step S304, question-answer data is returned; in step S305, configuring question and answer information according to the question and answer data, and sending the question and answer information to the user terminal; in step S306, receiving key information sent by the user terminal; in step S307, a logic judgment is performed in the text dialogue instruction list according to the key information; in step S308, the question answering data is queried in the question answering data table according to the logical judgment result; in step S309, question-answer data is returned; in step S310, a data interface of an external system is called to obtain external data; in step S311, the answer information is configured according to the external data and the question and answer data, and is transmitted to the user terminal.

The logic judgment comprises the step of acquiring a currently executed instruction identifier in a text conversation instruction list according to the session information or the key information, the question-answer data comprises the anti-question-answer operation and the preset answer information, and the external data comprises user information, product information or other information.

Referring to fig. 4, a multi-turn question answering method is described in detail by taking an example of querying product information, and fig. 4 is a schematic flow chart of the multi-turn question answering method according to an embodiment, where the flow at least includes steps S401 to S415, specifically,

in step S401, receiving "help me find product information" sent by the user terminal;

in step S402, it is determined that the instruction identification ask { product name };

in step S403, a question-back operation about ask { product name } is acquired;

in step S404, question and answer information "ask what kind of product you buy? ";

in step S405, "product 1" sent by the user terminal is received;

in step S406, it is determined that the instruction identification chose { sales channel };

in step S407, answer-to-question and preset answer information about the chose { sales channel } are acquired;

in step S408, question and answer information "please select a sales channel of a product you buy" is configured and transmitted: 1. a channel 1; 2. channel 2 ";

in step S409, receive "2" sent by the ue;

in step S410, it is determined that the instruction identification ask { city };

in step S411, a question reversal regarding ask { city } is acquired;

in step S412, question and answer information "ask you about which city the present product was purchased? ";

in step S413, "beijing" transmitted by the user terminal is received;

in step S414, an answer instruction identification is determined;

in step S415, the answer information of the product 1 in channel 2 of beijing is configured and transmitted.

Referring to fig. 5 to 8, the multi-turn question-answering method is described in detail by taking query hesitation as an example, fig. 5 shows a schematic structural diagram of a text conversation instruction table corresponding to the query hesitation, and as shown in fig. 5, the instruction identifier corresponding to the query hesitation includes an instruction 1, an instruction 2, an instruction 3, an instruction 4, and an answer instruction. Fig. 6 is a schematic diagram showing a structure of an exclusion data table corresponding to query hesitation periods, which includes a product table without hesitation periods as shown in fig. 6. Fig. 7 shows a schematic structural diagram of a question-answer data table corresponding to the query hesitation period, and as shown in fig. 7, the question-answer information can be configured at instruction 1, instruction 3 and instruction 5 through a question-answer technique in the question-answer data table. Fig. 8 is a schematic diagram showing an interface for querying a plurality of rounds of quiz answers corresponding to the hesitation periods, and as shown in fig. 8, the reply information of the user is about products in the non-hesitation period product table in the exclusion data table, and the answer information in the text conversation instruction table in fig. 5 is "{ product name } without hesitation period", and therefore, the configured answer information is "life care advance delivery appointment without hesitation period".

The multi-turn question and answer method is described in detail with reference to fig. 9 to 13 by taking a universal addition scene as an example, fig. 9 shows a structural schematic diagram of a text conversation instruction list corresponding to the universal addition scene, fig. 10 shows a structural schematic diagram of an exclusion data list corresponding to the universal addition scene, fig. 11 shows a structural schematic diagram of a question-answering data list corresponding to the universal addition scene, fig. 12 shows a structural schematic diagram of a basic data list corresponding to the universal addition scene, and fig. 13 shows an interface schematic diagram of the multi-turn question and answer method in the universal addition scene.

Those skilled in the art will appreciate that all or part of the steps implementing the above embodiments are implemented as computer programs executed by a CPU. The computer program, when executed by the CPU, performs the functions defined by the method provided by the present invention. The program of (a) may be stored in a computer readable storage medium, which may be a read-only memory, a magnetic or optical disk, or the like.

Furthermore, it should be noted that the above-mentioned figures are only schematic illustrations of the processes involved in the method according to exemplary embodiments of the invention, and are not intended to be limiting. It will be readily understood that the processes shown in the above figures are not intended to indicate or limit the chronological order of the processes. In addition, it is also readily understood that these processes may be performed synchronously or asynchronously, e.g., in multiple modules.

Embodiments of the apparatus of the present disclosure are described below, which may be used to perform the multi-round question-answering method of the present disclosure described above. For details not disclosed in the embodiments of the apparatus of the present disclosure, please refer to the embodiments of the multi-round question-answering method described above in the present disclosure.

Fig. 14 schematically illustrates a block diagram of a multi-turn question answering device according to one embodiment of the present disclosure.

Referring to fig. 14, a multi-turn question answering apparatus 1400 according to one embodiment of the present disclosure, the multi-turn question answering apparatus 1400 includes: a configure first question and answer information module 1401, a configure second question and answer information module 1402 and a configure target answer information module 1403. Specifically, the method comprises the following steps:

a first question-answering information configuring module 1401, configured to receive session information of a user, read a first instruction identifier corresponding to the session information in a text conversation instruction list, and configure the first question-answering information according to the first instruction identifier, where the text conversation instruction list is configured with a plurality of instruction identifiers having an execution sequence and one or more object information corresponding to each instruction identifier in a text file form;

a second question-answer information configuring module 1402, configured to receive key information of the user for the first question-answer information, configure object information, which is matched with the key information and object information corresponding to the first instruction identifier, as target object information, and configure second question-answer information according to the target object information and the second instruction identifier, where the second instruction identifier is executed after the first instruction identifier;

a configure target answer information module 1403, configured to repeat the above steps until a result instruction identifier in the text dialog instruction table is obtained, and configure target answer information according to the result instruction identifier.

In an exemplary embodiment of the present disclosure, the first question and answer information module 1401 may be further configured to obtain context information of the session information, determine step information according to the session information and the context information, and read a first instruction identifier corresponding to the step information in the text dialog instruction table according to the step information.

In an exemplary embodiment of the present disclosure, the first question and answer information module 1401 may be further configured to obtain, according to the first instruction identifier, first question and answer information corresponding to the first instruction identifier in the question and answer data table, and send the first question and answer information to a terminal corresponding to the user.

In an exemplary embodiment of the disclosure, the configure first question and answer information module 1402 may further be configured to obtain question information corresponding to the second instruction identifier in a question data table, obtain one or more preset answer information corresponding to the target object information in a basic data table, and configure the second question and answer information according to the question information and the one or more preset answer information. Wherein, the question-answer data table comprises a question-return data table and a basic data table.

In an exemplary embodiment of the present disclosure, the configuration target response information module 1403 obtains key information corresponding to each instruction identifier, and configures the target response information according to the key information corresponding to each instruction identifier and the result instruction identifier.

The specific details of each multi-turn question-answering device are described in detail in the corresponding multi-turn question-answering method, and therefore are not described herein again.

It should be noted that although in the above detailed description several modules or units of the apparatus for performing are mentioned, such a division is not mandatory. Indeed, the features and functionality of two or more modules or units described above may be embodied in one module or unit, according to embodiments of the present disclosure. Conversely, the features and functions of one module or unit described above may be further divided into embodiments by a plurality of modules or units.

In an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided.

As will be appreciated by one skilled in the art, aspects of the present invention may be embodied as a system, method or program product. Thus, various aspects of the invention may be embodied in the form of: an entirely hardware embodiment, an entirely software embodiment (including firmware, microcode, etc.) or an embodiment combining hardware and software aspects that may all generally be referred to herein as a "circuit," module "or" system.

An electronic device 1500 according to this embodiment of the invention is described below with reference to fig. 15. The electronic device 1500 shown in fig. 15 is only an example and should not bring any limitation to the functions and the scope of use of the embodiments of the present invention.

As shown in fig. 15, electronic device 1500 is in the form of a general purpose computing device. Components of electronic device 1500 may include, but are not limited to: the at least one processing unit 1510, the at least one storage unit 1520, a bus 1530 connecting different system components (including the storage unit 1520 and the processing unit 1510), and a display unit 1540.

Wherein the memory unit stores program code that is executable by the processing unit 1510 to cause the processing unit 1510 to perform steps according to various exemplary embodiments of the present invention as described in the above section "exemplary methods" of the present specification. For example, the processing unit 1510 may execute step S110 shown in fig. 1, receive session information of a user, read a first instruction identifier corresponding to the session information in a text dialog instruction table, and configure first question and answer information according to the first instruction identifier, where the text dialog instruction table is configured with a plurality of instruction identifiers having an execution order and one or more object information corresponding to each instruction identifier in a text file form; step S120, receiving key information of a user aiming at the first question and answer information, configuring matched object information in the key information and each object information corresponding to the first instruction identification as target object information, and configuring second question and answer information according to the target object information and the second instruction identification, wherein the second instruction identification is executed after the first instruction identification; and step S130, repeating the steps until a result instruction identifier in the text conversation instruction list is obtained, and configuring target answer information according to the result instruction identifier.

The storage unit 1520 may include readable media in the form of volatile storage units, such as a random access memory unit (RAM)15201 and/or a cache memory unit 15202, and may further include a read only memory unit (ROM) 15203.

Storage unit 1520 may also include a program/utility 15204 having a set (at least one) of program modules 15205, such program modules 15205 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may comprise an implementation of a network environment.

Bus 1530 may be any bus representing one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

The electronic device 1500 can also communicate with one or more external devices 1700 (e.g., keyboard, pointing device, bluetooth device, etc.), with one or more devices that enable a viewer to interact with the electronic device 1500, and/or with any devices (e.g., router, modem, etc.) that enable the electronic device 1500 to communicate with one or more other computing devices. Such communication may occur via input/output (I/O) interface 1550. Also, the electronic device 1500 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the internet) via the network adapter 1560. As shown, the network adapter 1560 communicates with the other modules of the electronic device 1500 over the bus 1530. It should be appreciated that although not shown, other hardware and/or software modules may be used in conjunction with the electronic device 1500, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.

Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein may be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solution according to the embodiments of the present disclosure may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (which may be a CD-ROM, a usb disk, a removable hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which may be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.

In an exemplary embodiment of the present disclosure, there is also provided a computer-readable storage medium having stored thereon a program product capable of implementing the above-described method of the present specification. In some possible embodiments, aspects of the invention may also be implemented in the form of a program product comprising program code means for causing a terminal device to carry out the steps according to various exemplary embodiments of the invention described in the above section "exemplary methods" of the present description, when said program product is run on the terminal device.

Referring to fig. 16, a program product 1600 for implementing the above method according to an embodiment of the present invention is described, which may employ a portable compact disc read only memory (CD-ROM) and include program code, and may be run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited in this regard and, in the present document, a readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

A computer readable signal medium may include a propagated data signal with readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A readable signal medium may also be any readable medium that is not a readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Program code embodied on a readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., through the internet using an internet service provider).

Furthermore, the above-described figures are merely schematic illustrations of processes involved in methods according to exemplary embodiments of the invention, and are not intended to be limiting. It will be readily understood that the processes shown in the above figures are not intended to indicate or limit the chronological order of the processes. In addition, it is also readily understood that these processes may be performed synchronously or asynchronously, e.g., in multiple modules.

Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is to be limited only by the terms of the appended claims.

Claims

1. A multi-turn question answering method is characterized by comprising the following steps:

receiving session information of a user, reading a first instruction identification corresponding to the session information in a text conversation instruction list, and configuring first question-answer information according to the first instruction identification, wherein the text conversation instruction list is configured with a plurality of instruction identifications with execution sequences and one or more object information corresponding to each instruction identification in a text file form;

receiving key information of the user aiming at the first question and answer information, configuring matched object information in the key information and each object information corresponding to the first instruction identification as target object information, and configuring second question and answer information according to the target object information and a second instruction identification, wherein the second instruction identification is executed after the first instruction identification;

and repeating the steps until a result instruction identification in the text conversation instruction list is obtained, and configuring target answer information according to the result instruction identification.

2. The multi-turn question answering method according to claim 1, wherein reading a first instruction identifier corresponding to the session information in a text session instruction list comprises:

obtaining the context information of the session information, determining step information according to the session information and the context information, and reading the first instruction identification corresponding to the step information in the text dialogue instruction list according to the step information.

3. The multi-round question answering method according to claim 1, wherein configuring first question answering information according to the first instruction identification comprises:

and acquiring the first question-answer information corresponding to the first instruction identification in a question-answer data table according to the first instruction identification, and sending the first question-answer information to a terminal corresponding to the user.

4. The multi-round question-answering method according to claim 3, wherein the question-answering data table includes a question-answering data table and a basic data table;

configuring second question-answering information according to the target object information and a second instruction identifier, wherein the second question-answering information comprises:

and obtaining question-answering information corresponding to the second instruction identification in the question-answering data table, obtaining one or more preset answer information corresponding to the target object information in the basic data table, and configuring the second question-answering information according to the question-answering information and the one or more preset answer information.

5. The multi-turn question answering method according to claim 1, wherein configuring target answer information according to the result instruction identification comprises:

and acquiring key information corresponding to each instruction identifier, and configuring the target answer information according to the key information corresponding to each instruction identifier and the result instruction identifier.

6. The multi-round question answering method according to claim 1, wherein configuring object information, which is matched with the key information and each object information corresponding to the first instruction identifier, as target object information comprises:

matching the key information with one or more object information corresponding to the first instruction identification;

and when the key information is matched with the object information corresponding to the first instruction identification, configuring the object information matched with the key information as the target object information.

7. The multi-round question answering method according to claim 6, characterized by comprising: the method further comprises the following steps:

when the key information is not matched with the object information corresponding to the first instruction identification, matching the key information with an irrelevant object list of the first instruction identification in a question-answer data table;

if yes, configuring target question-answering information into first preset answering information related to the first instruction identification;

and if not, configuring the target question-answering information into second preset answering information related to the first instruction identification.

8. A multi-turn question answering device, comprising:

the system comprises a first question-answer information configuration module, a first question-answer information configuration module and a first question-answer information configuration module, wherein the first question-answer information configuration module is used for receiving conversation information of a user, reading a first instruction identification corresponding to the conversation information in a text conversation instruction list, and configuring first question-answer information according to the first instruction identification, and the text conversation instruction list is configured with a plurality of instruction identifications with execution sequences and one or more object information corresponding to each instruction identification in a text file form;

a second question and answer information configuration module, configured to receive key information of the user for the first question and answer information, configure object information that matches the key information with object information corresponding to the first instruction identifier as target object information, and configure second question and answer information according to the target object information and a second instruction identifier, where the second instruction identifier is executed after the first instruction identifier;

and the target response information configuration module is used for repeating the steps until a result instruction identifier in the text conversation instruction list is obtained, and configuring target response information according to the result instruction identifier.

9. A computer-readable storage medium on which a computer program is stored, the program, when executed by a processor, implementing a multi-round question-answering method according to any one of claims 1 to 7.

10. An electronic device, comprising:

one or more processors;

a storage device to store one or more programs that, when executed by the one or more processors, cause the one or more processors to implement the multi-round question-answering method according to any one of claims 1 to 7.