CN114138958B - Information interaction method, device, equipment and storage medium - Google Patents

Information interaction method, device, equipment and storage medium

Info

Publication number
CN114138958B
CN114138958B CN202111438910.4A CN202111438910A CN114138958B CN 114138958 B CN114138958 B CN 114138958B CN 202111438910 A CN202111438910 A CN 202111438910A CN 114138958 B CN114138958 B CN 114138958B
Authority
CN
China
Prior art keywords
target
service
interactive
statement
terminal
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202111438910.4A
Other languages
Chinese (zh)
Other versions
CN114138958A (en
Inventor
李科研
李智勇
陈孝良
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing SoundAI Technology Co Ltd
Original Assignee
Beijing SoundAI Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing SoundAI Technology Co Ltd filed Critical Beijing SoundAI Technology Co Ltd
Priority to CN202111438910.4A priority Critical patent/CN114138958B/en
Publication of CN114138958A publication Critical patent/CN114138958A/en
Application granted granted Critical
Publication of CN114138958B publication Critical patent/CN114138958B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/332—Query formulation
    • G06F16/3329—Natural language query formulation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Theoretical Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Human Computer Interaction (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The application discloses an information interaction method, device, equipment and storage medium, and relates to the technical field of Internet. The method comprises the steps of obtaining interactive sentences of a terminal in an ith round of dialogue when providing target service for the terminal, determining target service nodes corresponding to the interactive sentences in a plurality of service nodes included in a service flow of the target service, wherein each service node of the service flow is used for obtaining elements required for realizing the target service, each service node is provided with a guide sentence used for triggering the next round of dialogue, and determining target interactive sentences based on the plurality of interactive sentences included in a history dialogue before the ith round of dialogue and the guide sentences corresponding to the target service nodes, and sending the target interactive sentences to the terminal. In the scheme, the target service node is rapidly positioned by means of the order of the service flow, and the target interactive statement is accurately determined by combining the history dialogue, so that the information interaction efficiency is improved.

Description

Information interaction method, device, equipment and storage medium
Technical Field
The present application relates to the field of internet technologies, and in particular, to an information interaction method, apparatus, device, and storage medium.
Background
With the development of internet technology, how to provide services (such as telecommunication services, legal services, health services, etc.) to users in an online information interaction manner is a problem to be solved.
In the related art, online services are generally provided with online customer service, a user inputs a question, the online customer service searches for an answer corresponding to the question, and then returns the answer for the user to view. Wherein the online customer service determines an answer based on a quantity of information stored in the knowledge base that is related to the service.
In the above technology, because the information related to the service is more, and the questions presented by different users in the information interaction process are more different, the questions are more scattered, in this case, the online customer service needs to determine answers related to the questions one by one from a large amount of information stored in the knowledge base, so the information interaction efficiency is lower.
Disclosure of Invention
The embodiment of the application provides an information interaction method, an information interaction device, information interaction equipment and a storage medium, which can improve the efficiency of information interaction. The technical proposal is as follows:
in one aspect, an information interaction method is provided, and the method includes:
when providing a target service for a terminal, acquiring an interactive sentence of the terminal in an ith round of dialogue, wherein i is an integer greater than 1;
Determining a target service node corresponding to the interactive statement in a plurality of service nodes included in a service flow of the target service, wherein each service node of the service flow is used for acquiring elements required for realizing the target service, and each service node is provided with a guide statement for triggering a next round of dialogue;
And determining a target interactive sentence based on a plurality of interactive sentences included in the history dialogue before the ith round of dialogue and the guide sentences corresponding to the target service nodes, and sending the target interactive sentence to the terminal.
In one possible implementation manner, the target service is a health service, the health service is used for providing consultation of health information for users, and determining the target service node corresponding to the interactive statement in a plurality of service nodes included in a service flow of the target service comprises the following steps:
acquiring target elements of the health service from the interactive sentences;
and determining the target service node corresponding to the target element in a plurality of service nodes included in the service flow of the health service.
In a possible implementation manner, the determining, based on a plurality of interactive sentences included in the history dialogue before the ith round of dialogue and the guiding sentences corresponding to the target service node, the target interactive sentence includes:
Determining the target interactive statement from the guide statement corresponding to the target business node based on the target element, the acquired history element and a plurality of elements required for realizing the health service;
The history element is an element obtained from a plurality of interactive sentences included in the history dialogue, and the target interactive sentence is used for obtaining elements except the target element and the history element in the plurality of elements.
In one possible implementation, the method further includes:
performing intention recognition on the interactive sentences to obtain intention recognition results;
and under the condition that the intention recognition result is not matched with the target service, the type of the target service is redetermined based on the intention recognition result, and information interaction is performed based on the redetermined type of the target service.
In a possible implementation manner, after the acquiring the interactive statement of the terminal in the ith round of dialogue, the method further includes:
Under the condition that the sentence type of the interactive sentence is a question type, acquiring an answer corresponding to the interactive sentence, and sending the answer to a terminal;
And executing the step of determining the target service node corresponding to the interactive statement in a plurality of service nodes included in the service flow of the target service under the condition that the statement type of the interactive statement is not the question type.
In one possible implementation, the method further includes:
And executing the step of determining the target service node corresponding to the interactive statement in the plurality of service nodes included in the service flow of the target service under the condition that the answer corresponding to the interactive statement is not acquired.
In one possible implementation, the method further includes:
Acquiring historical interaction data, wherein the historical interaction data comprises a plurality of interaction sentences obtained in a historical information interaction process;
extracting questions and answers from the historical interaction data to obtain a plurality of questions and answers corresponding to each question;
and storing the corresponding relation between each question and the answer corresponding to the question in a knowledge base.
In one possible implementation manner, the storing, in a knowledge base, a correspondence between each of the questions and an answer corresponding to the question includes:
Determining target problems corresponding to each problem based on a correlation determination model, wherein the target problems are problems stored in the knowledge base and have the largest correlation parameters with the problems, and the correlation determination model is trained based on correlation parameters among a plurality of sample problems;
And updating the answer corresponding to the target question based on the answer corresponding to the question in the knowledge base.
In one possible implementation, the method further includes:
And under the condition that the answer audit is passed, executing the step of updating the answer corresponding to the target question based on the answer corresponding to the question in the knowledge base.
In one possible implementation, the method further includes:
Acquiring portrait information corresponding to the terminal, wherein the portrait information is used for describing a user to which the terminal belongs;
The determining, based on a plurality of interactive sentences included in the history dialogue before the ith round of dialogue and the guiding sentences corresponding to the target service nodes, the target interactive sentences includes:
Determining a first guide statement from at least one guide statement corresponding to the target service node based on a plurality of interactive statements included in a history dialogue before the ith round of dialogue;
determining a second guide statement from at least one guide statement corresponding to a next-stage service node associated with the first guide statement based on the portrait information;
And adjusting the second guide statement based on the portrait information to obtain the target interactive statement.
In one possible implementation, the method further includes:
extracting portrait information of the user to which the terminal belongs from historical interaction data, wherein the historical interaction data comprises a plurality of interaction sentences obtained in the process of interacting with the historical information of the terminal.
In another aspect, there is provided an information interaction apparatus, the apparatus including:
The sentence acquisition module is used for acquiring an interactive sentence of the terminal in an ith round of dialogue when providing a target service for the terminal, wherein i is an integer greater than 1;
The node determining module is used for determining a target service node corresponding to the interactive statement in a plurality of service nodes included in the service flow of the target service, wherein each service node of the service flow is used for acquiring elements required for realizing the target service, and each service node is provided with a guide statement for triggering the next dialog;
And the statement determining module is used for determining a target interactive statement based on a plurality of interactive statements included in the history dialog before the ith round of dialog and the guide statement corresponding to the target service node, and sending the target interactive statement to the terminal.
In one possible implementation manner, the target service is a health service, the health service is used for providing consultation of health information for a user, the node determining module is used for acquiring target elements of the health service from the interactive statement, and the target service node corresponding to the target elements is determined in a plurality of service nodes included in a service flow of the health service.
In one possible implementation manner, the statement determining module is configured to determine the target interaction statement from guide statements corresponding to the target service node based on the target element, the acquired history element, and a plurality of elements required for implementing the health service;
The history element is an element obtained from a plurality of interactive sentences included in the history dialogue, and the target interactive sentence is used for obtaining elements except the target element and the history element in the plurality of elements.
In one possible implementation, the apparatus further includes:
The intention recognition module is used for carrying out intention recognition on the interactive sentences to obtain an intention recognition result;
And the type determining module is used for re-determining the type of the target service based on the intention recognition result and performing information interaction based on the re-determined type of the target service under the condition that the intention recognition result is not matched with the target service.
In one possible implementation, the apparatus further includes:
The answer acquisition module is used for acquiring an answer corresponding to the interactive sentence and sending the answer to a terminal under the condition that the sentence type of the interactive sentence is a question type;
The node determining module is further configured to determine, in a case where the statement type of the interactive statement is not a question type, a target service node corresponding to the interactive statement from a plurality of service nodes included in a service flow of the target service.
In one possible implementation, the apparatus further includes:
the node determining module is further configured to determine, when an answer corresponding to the interactive sentence is not obtained, a target service node corresponding to the interactive sentence from a plurality of service nodes included in the service flow of the target service.
In one possible implementation, the apparatus further includes:
The data acquisition module is used for acquiring historical interaction data, wherein the historical interaction data comprises a plurality of interaction sentences obtained in a historical information interaction process;
the data extraction module is used for extracting questions and answers from the historical interaction data to obtain a plurality of questions and answers corresponding to each question;
and the data storage module is used for storing the corresponding relation between each question and the answer corresponding to the question in a knowledge base.
In one possible implementation manner, the data storage module is configured to determine, based on a relevance determination model, a target problem corresponding to each problem, where the target problem is a problem stored in the knowledge base and has a maximum relevance parameter with respect to the problem, the relevance determination model is trained based on relevance parameters between a plurality of sample problems, and in the knowledge base, based on an answer corresponding to the problem, update the answer corresponding to the target problem.
In one possible implementation manner, the data storage module is configured to update, in the knowledge base, an answer corresponding to the target question based on the answer corresponding to the question if the answer passes.
In one possible implementation, the apparatus further includes:
The portrait acquisition module is used for acquiring portrait information corresponding to the terminal, and the portrait information is used for describing a user to which the terminal belongs;
The statement determination module is further configured to determine a first guide statement from at least one guide statement corresponding to the target service node based on a plurality of interactive statements included in a history dialog before the ith dialog, determine a second guide statement from at least one guide statement corresponding to a next-stage service node associated with the first guide statement based on the portrait information, and adjust the second guide statement based on the portrait information to obtain the target interactive statement.
In one possible implementation, the apparatus further includes:
And the portrait extraction module is used for extracting portrait information from the historical interaction data to obtain portrait information of the user to which the terminal belongs, wherein the historical interaction data comprises a plurality of interaction sentences obtained in the process of interacting with the historical information of the terminal.
In another aspect, an electronic device is provided that includes one or more processors and one or more memories having at least one program code stored therein, the at least one program code loaded and executed by the one or more processors to implement the information interaction method of the above aspect.
In another aspect, a computer readable storage medium is provided, in which at least one program code is stored, and the at least one program code is loaded and executed by a processor, to implement the information interaction method described in the above aspect.
In another aspect, a computer program product is provided, the computer program product comprising at least one program code, the at least one program code being loaded and executed by a processor to implement the information interaction method of the above aspect.
The technical scheme provided by the embodiment of the application has the beneficial effects that at least:
in the embodiment of the application, when the target service is provided for the terminal, the target service node corresponding to the interactive statement of the ith round of dialogue is rapidly positioned by means of the order of the service flow of the target service, and further, the target interactive statement returned to the terminal is accurately determined by combining the history dialogue and the guide statement corresponding to the target service node, so that the information interaction efficiency is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic illustration of an implementation environment provided by an embodiment of the present application;
FIG. 2 is a flowchart of an information interaction method according to an embodiment of the present application;
FIG. 3 is a flowchart of an information interaction method according to an embodiment of the present application;
FIG. 4 is a flowchart of an information interaction method according to an embodiment of the present application;
FIG. 5 is a flowchart of a knowledge base construction method according to an embodiment of the present application;
FIG. 6 is a flowchart of an information interaction method according to an embodiment of the present application;
FIG. 7 is a block diagram of an information interaction device according to an embodiment of the present application;
fig. 8 is a block diagram of a server according to an embodiment of the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present application more apparent, the embodiments of the present application will be described in further detail with reference to the accompanying drawings.
The terms "first," "second," "third," and "fourth" and the like in the description and in the claims and drawings are used for distinguishing between different objects and not necessarily for describing a particular sequential or chronological order. Furthermore, the terms "comprising," "including," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those listed steps or elements but may include other steps or elements not listed or inherent to such process, method, article, or apparatus.
The user data (including but not limited to user equipment information, user personal information, etc.) according to the present application is information authorized by the user or sufficiently authorized by each party.
Fig. 1 is a schematic diagram of an implementation environment provided by an embodiment of the present application, and as shown in fig. 1, the implementation environment includes a terminal 101 and a server 102, where the terminal 101 and the server 102 are connected through a wired or wireless network.
In some embodiments, terminal 101 is at least one of a cell phone, tablet, desktop, or laptop portable computer. Server 102 may be at least one of a server, a server cluster composed of a plurality of servers, a cloud server, a cloud computing platform, and a virtualization center, which is not limited in this embodiment of the present application.
In some embodiments, terminal 101 is provided with a target application for providing background services by server 102, which can provide an operating environment for online customer service. The terminal 101 can implement functions such as information interaction, information presentation, and the like through the target application, and the terminal 101 can implement information interaction between online customer service and a user who operates the terminal through the target application. In the embodiment of the present application, the terminal 101 is configured to obtain an interactive sentence input by a user, and send the interactive sentence to the server 102. The server 102 is configured to receive the interactive sentence of the terminal 101, determine a target interactive sentence based on the interactive sentence, and send the target interactive sentence to the terminal 101. Correspondingly, the terminal 101 is further configured to receive the target interaction sentence, and display the target interaction sentence.
In the embodiment of the present application, the technical solution provided in the embodiment of the present application may be implemented by a server or a terminal as an execution body, or the technical method provided in the present application may be implemented by interaction between the terminal and the server, which is not limited in the embodiment of the present application.
The information interaction method provided by the embodiment of the application can be applied to the information interaction scene of any target service, such as online services including telecommunication service, legal service or health service. The embodiment of the application is illustrated by taking the target service as the health service as an example.
For example, taking health service as nucleic acid detection service, a user inputs an interactive sentence of 'i want to reserve nucleic acid detection' in the terminal 101, the terminal 101 sends the interactive sentence to the server 102, and the server 102 implements information interaction with the terminal by executing the information interaction method provided by the embodiment of the application, so as to provide the nucleic acid detection service for the user.
It should be noted that, in the embodiment of the present application, the target service is only taken as the health service, and the health service is taken as the nucleic acid detection service as the example, which is not limited to the scenario of the information interaction method of the present application, for example, the health service may be a nucleic acid result query service or a vaccine reservation service.
Fig. 2 is a flowchart of an information interaction method provided in an embodiment of the present application. The embodiment of the application is executed by a server, and the method comprises the following steps:
Step 201, when providing a target service for a terminal, a server acquires an interactive sentence of the terminal in an ith round of dialogue, wherein i is an integer greater than 1.
In the embodiment of the application, when a user operating the terminal wants to acquire the target service, an interactive sentence is input in the terminal, so that information interaction between the terminal and the server is triggered, and the information interaction process exists in a dialogue manner, such as single-round or multi-round dialogue.
Step 202, the server determines a target service node corresponding to the interactive sentence in a plurality of service nodes included in the service flow of the target service, each service node of the service flow is used for obtaining elements required for realizing the target service, and each service node is provided with a guiding sentence for triggering the next dialog.
In the embodiment of the application, the server sets corresponding service flows for each type of target service in advance, so that when a certain type of target service is provided for the terminal, corresponding service nodes are triggered in sequence according to the execution sequence of the service flows corresponding to the type, and information interaction with the terminal is realized. The method comprises the steps of dividing service flows according to the types of target services, and therefore the field of the target services is subdivided.
The service flow has corresponding service flow description language, and the server is used as dialogue engine, i.e. service flow engine, to load the service flow description language of the provided target service, thereby realizing the triggering of the service node. Optionally, the server manages the traffic flow by managing the traffic flow configuration file, for example, the management manner of the traffic flow includes adding the traffic flow or modifying the traffic flow, etc. The service flow configuration file comprises configuration information of the service flow, wherein the configuration information comprises the type of corresponding target service, the number of service nodes, elements corresponding to each service node and the like.
It should be noted that, the present application does not specifically limit the setting manner of the service flow and the service node. For example, the server sets traffic flows for the nucleic acid subscription service and the vaccination service, respectively. It should be noted that, the guide statement corresponding to the service node is obtained by means of natural language generation.
Step 203, the server determines a target interactive sentence based on a plurality of interactive sentences included in the history dialogue before the ith round of dialogue and the guiding sentences corresponding to the target service node, and sends the target interactive sentence to the terminal.
In the embodiment of the application, when the target service is provided for the terminal, the target service node corresponding to the interactive statement of the ith round of dialogue is rapidly positioned by means of the order of the service flow of the target service, and further, the target interactive statement returned to the terminal is accurately determined by combining the history dialogue and the guide statement corresponding to the target service node, so that the information interaction efficiency is improved.
In the embodiment of the application, when a user wants to acquire a target service by means of a terminal, the user inputs an interactive sentence in the 1 st round of dialogue in the terminal, the terminal acquires the interactive sentence, the interactive sentence is sent to a server, and the server receives the interactive sentence. In some embodiments, after receiving the interactive statement, the server performs intent recognition on the interactive statement to obtain an intent recognition result, determines a type of target service matched with the intent recognition result, and determines a service flow corresponding to the type, so that information interaction with the terminal is realized according to the service flow.
For example, in the first round of dialogue, the interactive sentence input by the user is "i want to reserve vaccine", the intention recognition result of the interactive sentence is "vaccine reservation", the target service type corresponding to the intention recognition result is "vaccine reservation", and then the server determines the service flow corresponding to the "vaccine reservation" type, that is, the vaccine reservation service flow.
In the embodiment of the application, since a plurality of elements are often needed to provide a target service, and after the first round of dialogue is started, a user has difficulty in providing a complete plurality of elements in one round of dialogue, and then the server acquires the plurality of elements through a plurality of rounds of dialogue with the terminal, thereby providing the target service for the terminal. For example, for vaccine reservation service, the server needs to obtain multiple elements such as age, gender, address and past medical history of the user, so that the server obtains the multiple elements through the information interaction method provided by the embodiment of the application, so as to provide high-quality target service for the terminal.
In some embodiments, the targeted service is a health service, the health service being used to provide a consultation of health information for the user, the type of health service including nucleic acid detection, vaccination, nucleic acid results query, or other health service type. Accordingly, fig. 3 is a flowchart of an information interaction method provided in an embodiment of the present application. The embodiment of the application takes the target service as the health service, and takes the ith round of dialogue between the terminal and the server as an example for explanation, and the method comprises the following steps:
Step 301, the terminal obtains an interactive sentence in the ith round of dialogue, and sends the interactive sentence to the server, wherein i is an integer greater than 1.
In the embodiment of the application, the terminal provides health service for the user by means of the server, and in the process of multiple conversations between the terminal and the server, the terminal sends the interactive sentence to the server every time the terminal acquires the interactive sentence, so that the server can determine the conversation strategy according to the interactive sentence in time.
Step 302, the server receives the interactive sentence.
In some embodiments, the server, after receiving the interactive sentence sent by the terminal, stores the interactive sentence for subsequent processing of the interactive sentence.
Step 303, the server acquires the target element of the health service from the interactive statement.
Since the interactive sentence transmitted by the terminal may contain elements required for realizing the health service, the server acquires the target elements from the interactive sentence after acquiring the interactive sentence. Optionally, the implementation manner of the step 303 includes that the server performs natural language processing on the interactive sentence to obtain semantic information, and the target element is extracted from the semantic information. For example, the interactive sentence is "i did not burn", the semantic information is "user did not burn", and the target element is "no burn".
Step 304, the server determines a target service node corresponding to the target element from a plurality of service nodes included in the service flow of the health service.
The server determines the service node corresponding to the ith-1 round of dialogue before the ith round of dialogue, and correspondingly, determines the target service node corresponding to the target element from the next-level service nodes of the service nodes. For example, the service node corresponding to the i-1 round of dialogue is used for determining whether the user is febrile, that is, for acquiring the factor of whether the user is febrile, two next-stage service nodes of the service node are respectively a yes node and a no node, and if the target factor is no febrile, the target service node is the no node.
In the embodiment of the application, for the ith round of dialogue, the target element is acquired from the interactive sentence in a natural language understanding mode, and the trigger of the target service node is driven by the target element, so that the element acquisition process in the process of providing health service is realized in the form of service flow, and the ordering of information interaction is improved.
Step 305, the server determines a target interactive sentence from the guide sentences corresponding to the target business nodes based on the target element, the acquired history element and a plurality of elements required for realizing the health service.
The history element is an element obtained from a plurality of interactive sentences included in a history dialogue before the ith dialogue, and the target interactive sentence is used for obtaining elements except the target element and the history element in the plurality of elements. In some embodiments, the implementation of step 305 includes the server determining elements other than the target element and the history element from the plurality of elements, and determining a target interaction statement for acquiring the element from the guide statement corresponding to the target service node.
For example, the plurality of elements are "sex, age, residence address", the target element is "woman", the history element is "26 years", then the remaining elements to be acquired are "residence address", and accordingly, the target interactive sentence is "please ask you where you now reside".
In some embodiments, in the case that the number of the remaining elements to be acquired is a plurality of, the server selects a guide sentence with the forefront execution sequence from a plurality of guide sentences corresponding to the target service node as the target interaction sentence, or the server randomly selects one guide sentence as the target interaction sentence. Optionally, the server sets an execution sequence of each guide statement corresponding to the service node in advance, so that the elements can be acquired according to the execution sequence. The execution sequence may be set as needed, which is not particularly limited in the embodiment of the present application.
In the embodiment of the application, the server decides the conversation strategy of the current round according to the current conversation and the historical conversation, namely the context information of the conversation. Wherein the determination of the dialog strategy includes querying the user for more detailed intent by sending a target interaction statement, or clarifying, confirming a portion or elements of the user's current intent, thereby providing a health service to the user based on the most current intent.
In some embodiments, in the process of providing health services for the terminal, the user operating the terminal may change the intention in the process of dialogue, and the server needs to provide health services for the terminal according to the intention of the user, accordingly, before step 304, the information interaction method provided by the embodiment of the application further includes the steps of performing intention recognition on the interactive statement by the server to obtain an intention recognition result, and re-determining the type of health services based on the intention recognition result and performing information interaction based on the re-determined type of health services if the intention recognition result does not match with the health services.
After each interactive sentence sent by the terminal is obtained, the server carries out intention recognition on the interactive sentence, and maintains the complete intention of the user in time. If the intent recognition result and the health service do not match, indicating that the user's intent has changed, then the implementation of the server to re-determine the type of health service based on the intent recognition result includes the server determining the type of health service that the intent recognition result matches. Accordingly, the implementation manner of the server for information interaction based on the redetermined type of the health service includes that the server determines a service flow matched with the type according to the redetermined type of the health service, and performs information interaction according to the service flow, that is, performs the operations of step 304-step 305 according to the redetermined service flow.
For example, the intention recognition result corresponding to the interactive sentence of the i-1 th round of dialogue is "nucleic acid detection reservation", the intention recognition result corresponding to the interactive sentence of the i-1 th round of dialogue is "vaccine reservation", the intention recognition result is changed, the type of the health service determined again is "vaccine reservation", and the server determines that the service flow corresponding to the "vaccine reservation" type is the vaccine reservation service flow.
In one possible implementation, in case the intended recognition result matches the health service, the server performs the operations of steps 304-305 based on the traffic flow of the health service.
In the embodiment of the application, the state of the multi-round dialogue is tracked by redetermining the intention recognition result for each interactive sentence, so that the intention of a user is maintained according to the latest state of the multi-round dialogue, the target interactive sentence is determined according to the latest intention of the user, and the accuracy of the target interactive sentence is further improved.
And 306, the server sends the target interactive statement to the terminal.
In the embodiment of the application, after obtaining the target interactive statement, the server sends the target interactive statement to the terminal.
Step 307, the terminal receives the target interactive statement and displays the target interactive statement.
In some embodiments, after receiving the target interactive sentence sent by the server, the terminal displays the target interactive sentence on the display interface, so that a user operating the terminal can view the target interactive sentence in time.
In the embodiment of the application, the dialogue state tracking and the dialogue strategy determination are carried out in a business flow mode so as to realize multi-round dialogue, thereby completing dialogue tasks with higher efficiency on the premise of meeting the intention of a user to the maximum extent.
In the embodiment of the application, when providing health service for the terminal, on one hand, the target service node corresponding to the interactive statement of the ith round of dialogue is selected from the service flow of the health service, the task is rapidly positioned by means of the order of the service flow, and on the other hand, the target interactive statement is determined by combining the history dialogue and the guide statement corresponding to the target service node, so that the information interaction efficiency is improved.
In some embodiments, the sentence type of the interactive sentence input by the user in the terminal is not a question type, for example, the interactive sentence is "i want to reserve nucleic acid detection", "query the result of nucleic acid detection", or "latest epidemic situation information", then the interactive sentence has no fixed answer, and therefore, the server can determine the target interactive sentence for replying to the interactive sentence in the form of a service flow, that is, the server performs the operations of steps 303-306 in the case that the sentence type of the interactive sentence is not a question type.
In other embodiments, the sentence type of the interactive sentence input by the user in the terminal is a question type, for example, why the interactive sentence is "what is vaccine", "what is attention to vaccination", or "what symptom is pneumonia", and then, for the interactive sentence of the question type, the server can obtain an answer corresponding to the interactive sentence, so as to return the answer to the terminal. Accordingly, fig. 4 is a flowchart of an information interaction method provided in an embodiment of the present application. The embodiment of the application takes the statement type of the interactive statement sent by the terminal as an example for describing the problem statement, and the method comprises the following steps:
step 401, the terminal obtains the interactive sentence in the ith round of dialogue, and sends the interactive sentence to the server, wherein i is an integer greater than 1.
Step 402, the server receives the interactive statement.
In some embodiments, the implementation manners of step 401 to step 402 are the same as the implementation manners of step 301 to step 302, and are not described herein.
Step 403, the server obtains the answer corresponding to the interactive sentence under the condition that the sentence type of the interactive sentence is the question type.
In some embodiments, the server builds a knowledge base in advance, which is used to store the correspondence between questions and answers. Wherein, the questions in the knowledge base are used as indexes (queries), and the answers are used as the values corresponding to the indexes. And when the server obtains the interactive statement of the question type, retrieving an answer corresponding to the interactive statement from the knowledge base. Correspondingly, the implementation mode of the server for obtaining the answer corresponding to the interactive sentence comprises the step that the server takes the interactive sentence as an index and searches the answer corresponding to the interactive sentence from a knowledge base.
In one possible implementation, the server performs the operation of step 404 if an answer corresponding to the interactive sentence is obtained. In another possible implementation manner, the answers corresponding to the interactive sentences may not be stored in the knowledge base, and the information interaction method provided by the embodiment of the application further includes the step that the server executes the operations of the steps 303-306 under the condition that the answers corresponding to the interactive sentences are not obtained.
In the embodiment of the application, because more information is involved in the target service and the obtained interactive sentences possibly have no corresponding answers, the information interaction with the terminal is realized by means of the service flow corresponding to the target service, so that even if no answer is obtained, the normal provision of the target service can be realized, the stability of the dialogue function is further improved, and the user experience is also improved.
Step 404, the server sends the answer to the terminal.
Step 405, the terminal receives the answer and displays the answer.
In some embodiments, the implementation of step 404-step 405 is the same as the implementation of step 306-step 307, and will not be described again here.
In the embodiment of the application, for the interactive sentences of the question types, the quick provision of the target service is realized by directly acquiring the answers corresponding to the interactive sentences, and the efficiency of information interaction is improved.
In the embodiment of the application, before providing the target service for the terminal, the server firstly constructs an initial knowledge base, and then maintains the knowledge base according to the interactive sentences obtained in the information interaction process based on the target service. Optionally, the initial knowledge base is constructed in a manner that a technician acquires a question and a corresponding answer from a large amount of information related to the target service, and the server stores the question and the corresponding answer in the knowledge base to obtain the initial knowledge base. Wherein, a large amount of information related to the target service can be acquired by means of books, websites, questionnaires and the like.
Accordingly, fig. 5 is a flowchart of a knowledge base construction method according to an embodiment of the present application. The embodiment of the application is executed by a server, and the method comprises the following steps:
step 501, a server acquires historical interaction data, wherein the historical interaction data comprises a plurality of interaction sentences obtained in a historical information interaction process.
The historical interaction data comprise historical interaction data generated in the historical information interaction process of the server and the plurality of terminals. In some embodiments, the implementation manner of storing the historical interaction data by the server comprises the steps that in the process of each information interaction, the server stores the interaction statement sent by the terminal and generated by the server, so that the historical interaction data are obtained. Or the server periodically collects the log generated in the information interaction process and extracts the historical interaction data from the log.
For example, taking a target service as an epidemic situation prevention and control consultation service as an example, a server acquires an online log of intelligent customer service (such as an anti-epidemic consultation robot) for performing the epidemic situation prevention and control consultation service, and acquires historical interaction data related to the epidemic situation prevention and control consultation service from the online log.
And 502, extracting questions and answers from the historical interaction data by the server to obtain a plurality of questions and answers corresponding to each question.
In the history information interaction process based on the target service, the user can ask a question to the server through the terminal, so that the terminal sends the question to the server, and the server solves the question to obtain an answer, so that the server can extract the question and the answer from the history interaction data.
Step 503, the server stores the corresponding relation between each question and the answer corresponding to the question in the knowledge base.
In some embodiments, the server stores the correspondence between the obtained questions and answers in a knowledge base. Therefore, as the extracted questions are original questions input by the user through the terminal, the questions stored in the knowledge base are more matched with questions possibly asked by the user in a real target service scene by storing the original questions and corresponding answers in the knowledge base, and further, the searching speed of the answers can be improved in the process of providing target services later.
In other embodiments, the questions sent by the terminal during the information interaction may be related to the questions stored in the knowledge base, and then the server can determine the relevance parameters of the questions extracted from the historical interaction data, and store the questions and the corresponding answers according to the relevance parameters. Accordingly, the implementation manner of step 503 includes that the server determines a target question corresponding to each question based on a relevance determination model, the target question is a stored question in the knowledge base and has the largest relevance parameter with the question, the relevance determination model is trained based on relevance parameters among a plurality of sample questions, and in the knowledge base, the answer corresponding to the target question is updated based on the answer corresponding to the question.
The relevance parameter is used for measuring the semantic relevance degree between two problems, the representation mode of the relevance parameter can be set according to requirements, and the application is not particularly limited to the description, for example, the relevance parameter is represented in the form of a score. In the embodiment of the application, the server trains the correlation determination model in advance. Optionally, the server acquires a plurality of sample questions in advance, marks the correlation parameters of the sample questions, takes the correlation parameters as supervision, and trains a correlation determination model according to the plurality of sample questions. It should be noted that the correlation determination model may be a machine learning model, which is not particularly limited in the present application.
The server inputs the extracted problems and the problems stored in the knowledge base into the relevance determination model as model input data, and the relevance determination model outputs target problems and relevance parameters between the target problems and the extracted problems.
The server can update the questions in the knowledge base in combination with the magnitude of the correlation parameter, and optionally, the server updates the answers corresponding to the target questions in the knowledge base based on the answers corresponding to the questions. And under the condition that the correlation parameter is not greater than the correlation threshold, the server supplements the corresponding relation between the questions and the corresponding answers to the knowledge base.
The magnitude of the correlation threshold may be set as required, which is not particularly limited in the present application. It should be noted that, the server may update and train the correlation determination model periodically to improve the accuracy of the model.
In the embodiment of the application, whether the questions to be supplemented to the knowledge base are repeated with the stored questions in the knowledge base is determined by means of the correlation determination model, so that the similar questions can be clustered, and the questions in the knowledge base and the corresponding answers are optimized.
In some embodiments, in the process of updating the stored questions and answers in the knowledge base, the extracted questions and answers are audited to ensure the accuracy of the questions and answers, and correspondingly, the information interaction method provided by the embodiment of the application further comprises the steps that the server executes the step of updating the answers corresponding to the target questions based on the answers corresponding to the questions in the knowledge base under the condition that the answers are audited.
In one possible implementation, in the event that the audit is not passed, the server discards the audit-failed questions and answers or the technician modifies the audit-failed questions and answers until the questions and answers pass.
It should be noted that, before the verification is passed, the original questions and the original answers are still stored in the knowledge base, so that the server still performs information interaction with the terminal by the original answers when performing information interaction based on the knowledge base. In the embodiment of the application, the accuracy of the questions and the answers in the knowledge base can be ensured by auditing the questions and the answers updated into the knowledge base.
In addition to historical interaction data, the server may also obtain other data to build a knowledge base. Optionally, the server acquires historical telephone record information of the customer service telephone based on the target service, extracts questions and answers from the historical telephone record information, and constructs a knowledge base according to the questions and the answers. Of course, the server may also maintain the knowledge base in other ways, which the present application is not particularly limited to.
In an embodiment of the present application, the server may periodically perform the operations of steps 501-503 to maintain a knowledge base to enrich and build the knowledge base, thereby providing a better target service by means of the knowledge base. For example, the problems in the knowledge base are classified regularly, and the same problems are combined, so that the knowledge base is cleaned. In the process of providing the target service based on the service flow, the service flow comprises a plurality of service nodes, each service node possibly corresponds to one or more problems, and then the server can update and maintain a large number of problems stored in the knowledge base according to the service nodes, so that the knowledge base in the subdivision field aiming at the target service is constructed, and the coverage rate of long-tail problems of users is improved. For example, the server performs important maintenance on service nodes with high frequency of use.
In the embodiment of the application, because the history interaction data can represent the target service required by the user in the real information interaction process, the knowledge base is constructed based on the history interaction data, so that the stored questions and answers in the knowledge base can be more close to the requirements of the user, and the accuracy of the knowledge base is improved.
In the embodiment of the application, the server can acquire the portrait information of the user in the process of providing the target service for the terminal, so that the server can store the portrait information, and the stored portrait information is used for determining target interactive sentences in the subsequent information interaction process, thereby improving the efficiency of information interaction and improving the experience of the user. Accordingly, fig. 6 is a flowchart of an information interaction method provided in an embodiment of the present application. The embodiment of the application is executed by a server, and the method comprises the following steps:
Step 601, the terminal acquires an interactive sentence in the ith round of dialogue, and sends the interactive sentence to the server, wherein i is an integer greater than 1.
Step 602, the server receives the interactive sentence.
In step 603, the server determines a target service node corresponding to the interactive sentence among a plurality of service nodes included in the service flow of the target service, where each service node of the service flow is used to obtain an element required for implementing the target service, and each service node has a guiding sentence for triggering a next round of dialogue.
In some embodiments, the implementation of step 601 to step 603 is the same as the implementation of step 301 to step 304, and will not be described here again.
Step 604, the server acquires portrait information corresponding to the terminal, wherein the portrait information is used for describing the user to which the terminal belongs.
The server stores the user identification and the portrait information correspondingly, so that the subsequent portrait information is convenient to acquire. The user identification is identification such as a user account number. Optionally, the information interaction method provided by the embodiment of the application further comprises the following steps that the server extracts portrait information from the historical interaction data to obtain portrait information of the user to which the terminal belongs, wherein the historical interaction data comprises a plurality of interaction sentences obtained in the process of interacting with the historical information of the terminal.
The historical interaction data are corresponding to the same user identifier. The representation information includes at least one of gender, age, residence address, preference information, and the like. In the embodiment of the application, the portrait information of the user is extracted from the historical interaction data with the terminal, so that data support is provided for the subsequent target service, the time for information acquisition is saved, and the information interaction efficiency is improved.
Step 605, the server determines a first guiding sentence from at least one guiding sentence corresponding to the target service node based on a plurality of interactive sentences included in the history dialogue before the ith round of dialogue.
In some embodiments, the implementation of the server determining the first guide statement is the same as the implementation of step 305, and will not be described in detail herein.
Step 606, the server determines a second guide statement from at least one guide statement corresponding to a next-stage service node associated with the first guide statement based on the representation information.
For each guiding sentence corresponding to the target service node, the guiding sentence may be associated with a next-stage service node, and then the next-stage service node acquires other elements from the user according to the elements included in the portrait information, and because the server has acquired the portrait information, it is not necessary to send a first guiding sentence to the terminal to guide the user to input the corresponding elements.
Step 607, the server adjusts the second guiding sentence based on the image information to obtain the target interactive sentence.
The server may use the second guide sentence as a template sentence, and accordingly, the implementation manner of step 607 includes that the server fills corresponding elements in the portrait information into the second guide sentence to obtain the target interactive sentence. For example, the interactive statement sent by the terminal is "i want to reserve nucleic acid detection", the first guide statement is "please ask you for where you are now living", if the portrait information already includes the living address of the user, the determined second guide statement is "you want to reserve XX hospitals", and the adjusted target interactive statement is "you want to reserve XX hospitals of XX city XX", wherein XX hospitals of XX city XX are determined according to the portrait information, for example, a hospital near the living address is selected.
In the embodiment of the present application, the steps 605 to 607 are one implementation manner of determining, by the server, the target interactive sentence based on a plurality of interactive sentences included in the history dialogue before the ith round of dialogue and the guide sentences corresponding to the target service node.
Step 608, the server sends the target interactive statement to the terminal.
Step 609, the terminal receives the target interactive sentence and displays the target interactive sentence.
In some embodiments, the implementation of steps 608-609 is the same as the implementation of steps 306-307, and will not be described in detail herein.
In the embodiment of the application, the information interaction with the terminal is realized by combining the portrait information of the user, so that the acquisition process of some elements in the information interaction process is saved, the information interaction efficiency is greatly improved, and the user experience is improved.
Fig. 7 is a block diagram of an information interaction device provided in an embodiment of the present application, referring to fig. 7, the device includes:
A sentence acquisition module 701, configured to acquire, when a target service is provided for a terminal, an interactive sentence of the terminal in an ith round of dialogue, where i is an integer greater than 1;
A node determining module 702, configured to determine, among a plurality of service nodes included in a service flow of the target service, a target service node corresponding to the interaction sentence, where each service node of the service flow is configured to obtain an element required for implementing the target service, and each service node has a guide sentence for triggering a next round of dialogue;
statement determination module 703, configured to determine a target interaction statement based on a plurality of interaction statements included in a history dialog before the ith round of dialog and a guide statement corresponding to the target service node, and send the target interaction statement to the terminal.
In one possible implementation, the target service is a health service, the health service is used for providing consultation of health information for a user, the node determining module 702 is used for obtaining target elements of the health service from the interactive statement, and determining the target service node corresponding to the target elements in a plurality of service nodes included in a service flow of the health service.
In a possible implementation manner, the statement determining module 703 is configured to determine the target interaction statement from the guide statement corresponding to the target service node based on the target element, the acquired history element, and a plurality of elements required for implementing the health service;
The history element is an element obtained from a plurality of interactive sentences included in the history dialogue, and the target interactive sentence is used for obtaining elements except the target element and the history element in the plurality of elements.
In one possible implementation, the apparatus further includes:
The intention recognition module is used for carrying out intention recognition on the interactive statement to obtain an intention recognition result;
and the type determining module is used for re-determining the type of the target service based on the intention recognition result and performing information interaction based on the re-determined type of the target service under the condition that the intention recognition result is not matched with the target service.
In one possible implementation, the apparatus further includes:
the answer acquisition module is used for acquiring an answer corresponding to the interactive sentence and sending the answer to the terminal under the condition that the sentence type of the interactive sentence is the question type;
The node determining module 702 is further configured to determine, in a case where the statement type of the interactive statement is not a question type, a target service node corresponding to the interactive statement from a plurality of service nodes included in the service flow of the target service.
In one possible implementation, the apparatus further includes:
The node determining module 702 is further configured to determine, in a case where an answer corresponding to the interactive sentence is not obtained, a target service node corresponding to the interactive sentence from a plurality of service nodes included in the service flow of the target service.
In one possible implementation, the apparatus further includes:
The data acquisition module is used for acquiring historical interaction data, wherein the historical interaction data comprises a plurality of interaction sentences obtained in a historical information interaction process;
The data extraction module is used for extracting questions and answers from the historical interaction data to obtain a plurality of questions and answers corresponding to each question;
And the data storage module is used for storing the corresponding relation between each question and the answer corresponding to the question in the knowledge base.
In one possible implementation manner, the data storage module is configured to determine, based on a relevance determining model, a target question corresponding to each question, where the target question is a question stored in the knowledge base and has a maximum relevance parameter with respect to the question, the relevance determining model is trained based on relevance parameters between a plurality of sample questions, and in the knowledge base, based on an answer corresponding to the question, the answer corresponding to the target question is updated.
In one possible implementation, the data storage module is configured to update, in the knowledge base, an answer corresponding to the target question based on an answer corresponding to the question if the answer passes.
In one possible implementation, the apparatus further includes:
the portrait acquisition module is used for acquiring portrait information corresponding to the terminal, wherein the portrait information is used for describing a user to whom the terminal belongs;
The statement determination module 703 is further configured to determine, based on a plurality of interactive statements included in a history dialog preceding the ith dialog, a first guide statement from at least one guide statement corresponding to the target service node, determine, based on the representation information, a second guide statement from at least one guide statement corresponding to a next service node associated with the first guide statement, and adjust the second guide statement based on the representation information to obtain the target interactive statement.
In one possible implementation, the apparatus further includes:
And the portrait extraction module is used for extracting portrait information of the historical interaction data to obtain portrait information of the user of the terminal, wherein the historical interaction data comprises a plurality of interaction sentences obtained in the process of interacting with the historical information of the terminal.
In the embodiment of the application, when the target service is provided for the terminal, the target service node corresponding to the interactive statement of the ith round of dialogue is rapidly positioned by means of the order of the service flow of the target service, and further, the target interactive statement returned to the terminal is accurately determined by combining the history dialogue and the guide statement corresponding to the target service node, so that the information interaction efficiency is improved.
In the embodiment of the present application, an electronic device is provided as a server for example. Fig. 8 is a block diagram of a server 800 according to an embodiment of the present application. The server 800 may include one or more processors (Central Processing Units, CPU) 801 and one or more memories 802, where the one or more memories 802 are used to store at least one program code, and the at least one program code is loaded and executed by the one or more processors 801 to implement the information interaction method in the above embodiments. Of course, the server 800 may also have a wired or wireless network interface, a keyboard, an input/output interface, and other components for implementing the functions of the device, which are not described herein.
Those skilled in the art will appreciate that the architecture shown in fig. 8 is not limiting and that more or fewer components than shown may be included or certain components may be combined or a different arrangement of components may be employed.
The embodiment of the application also provides a computer readable storage medium, wherein at least one program code is stored in the computer readable storage medium, and the at least one program code is loaded and executed by a processor to realize the information interaction method in the embodiment.
Embodiments of the present application also provide a computer program product comprising at least one program code stored in a computer readable storage medium. The processor of the electronic device reads the program code from the computer-readable storage medium, and the processor executes the program code so that the electronic device performs the information interaction method in the above-described embodiment.
In some embodiments, a computer program according to an embodiment of the present application may be deployed to be executed on one electronic device, or on a plurality of electronic devices located at one site, or on a plurality of electronic devices distributed at a plurality of sites and interconnected by a communication network, where the plurality of electronic devices distributed at the plurality of sites and interconnected by the communication network may constitute a blockchain system.
The foregoing is illustrative of the present application and is not to be construed as limiting thereof, but rather as various modifications, equivalent arrangements, improvements, etc., which fall within the spirit and principles of the present application.

Claims (13)

1. An information interaction method, characterized in that the method comprises:
When a target service is provided for a terminal, acquiring an interactive sentence of the terminal in an ith round of dialogue, wherein i is an integer greater than 1, the target service is a health service, and the health service is used for providing consultation of health information for a user;
acquiring target elements of the health service from the interactive sentences;
Determining a target service node corresponding to the target element in a plurality of service nodes included in the service flow of the health service, wherein each service node of the service flow is used for acquiring elements required for realizing the target service, and each service node is provided with a guide statement for triggering the next round of dialogue;
And determining elements except the target element and the acquired history element from a plurality of elements required for realizing the health service, determining a target interaction statement for acquiring the rest elements to be acquired from guide statements corresponding to the target service node according to the execution sequence of each guide statement, and sending the target interaction statement to the terminal, wherein the history element is an element acquired from a plurality of interaction statements included in a history dialogue before the ith round of dialogue, and the target interaction statement is used for acquiring the elements except the target element and the history element from the plurality of elements.
2. The method according to claim 1, wherein the method further comprises:
performing intention recognition on the interactive sentences to obtain intention recognition results;
and under the condition that the intention recognition result is not matched with the target service, the type of the target service is redetermined based on the intention recognition result, and information interaction is performed based on the redetermined type of the target service.
3. The method of claim 1, wherein after the acquiring the interactive statement of the terminal in the ith round of dialogue, the method further comprises:
Under the condition that the sentence type of the interactive sentence is a question type, acquiring an answer corresponding to the interactive sentence, and sending the answer to a terminal;
And executing the step of determining a target service node corresponding to the target element acquired from the interactive sentence in a plurality of service nodes included in the service flow of the target service when the sentence type of the interactive sentence is not a question type.
4. A method according to claim 3, characterized in that the method further comprises:
and executing the step of determining the target service node corresponding to the target element acquired from the interactive sentence in a plurality of service nodes included in the service flow of the target service under the condition that the answer corresponding to the interactive sentence is not acquired.
5. The method according to claim 3 or 4, characterized in that the method further comprises:
Acquiring historical interaction data, wherein the historical interaction data comprises a plurality of interaction sentences obtained in a historical information interaction process;
extracting questions and answers from the historical interaction data to obtain a plurality of questions and answers corresponding to each question;
and storing the corresponding relation between each question and the answer corresponding to the question in a knowledge base.
6. The method of claim 5, wherein storing the correspondence between each of the questions and answers corresponding to the questions in a knowledge base comprises:
Determining target problems corresponding to each problem based on a correlation determination model, wherein the target problems are problems stored in the knowledge base and have the largest correlation parameters with the problems, and the correlation determination model is trained based on correlation parameters among a plurality of sample problems;
And updating the answer corresponding to the target question based on the answer corresponding to the question in the knowledge base.
7. The method of claim 6, wherein the method further comprises:
And under the condition that the answer audit is passed, executing the step of updating the answer corresponding to the target question based on the answer corresponding to the question in the knowledge base.
8. The method according to claim 1, wherein the method further comprises:
Acquiring portrait information corresponding to the terminal, wherein the portrait information is used for describing a user to which the terminal belongs;
after the target service node corresponding to the target element is determined, the method further comprises:
Determining elements except the target element and the acquired history elements from a plurality of elements for realizing the health service, wherein the history elements are elements acquired from a plurality of interactive sentences included in a history dialogue before the ith round of dialogue, and determining a first guide sentence for acquiring the rest elements to be acquired from at least one guide sentence corresponding to the target service node according to the execution sequence of each guide sentence;
determining a second guide statement from at least one guide statement corresponding to a next-stage service node associated with the first guide statement based on the portrait information;
And adjusting the second guide statement based on the portrait information to obtain the target interactive statement.
9. The method of claim 8, wherein the method further comprises:
extracting portrait information of the user to which the terminal belongs from historical interaction data, wherein the historical interaction data comprises a plurality of interaction sentences obtained in the process of interacting with the historical information of the terminal.
10. An information interaction device, the device comprising:
The sentence acquisition module is used for acquiring an interactive sentence of the terminal in an ith round of dialogue when providing a target service for the terminal, wherein i is an integer greater than 1, the target service is a health service, and the health service is used for providing consultation of health information for a user;
The node determining module is used for acquiring target elements of the health service from the interactive sentences, determining target service nodes corresponding to the target elements in a plurality of service nodes included in a service flow of the health service, wherein each service node of the service flow is used for acquiring elements required for realizing the target service, and each service node is provided with a guide sentence for triggering a next round of dialogue;
A statement determination module, configured to determine, from a plurality of elements required for implementing the health service, elements other than the target element and the acquired history element, determine, from guide statements corresponding to the target service node, a target interaction statement for acquiring remaining elements to be acquired according to an execution order of each guide statement, and send the target interaction statement to the terminal, where the history element is an element acquired from a plurality of interaction statements included in a history dialogue before the ith dialogue, and the target interaction statement is used to acquire elements other than the target element and the history element from the plurality of elements.
11. An electronic device comprising one or more processors and one or more memories, the one or more memories having stored therein at least one piece of program code that is loaded and executed by the one or more processors to implement the information interaction method of any of claims 1-9.
12. A computer readable storage medium having stored therein at least one program code, the at least one program code being loaded and executed by a processor to implement the information interaction method of any one of claims 1 to 9.
13. A computer program product, characterized in that the computer program product comprises at least one program code, which is loaded and executed by a processor to implement the information interaction method of any one of claims 1 to 9.
CN202111438910.4A 2021-11-25 2021-11-25 Information interaction method, device, equipment and storage medium Active CN114138958B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111438910.4A CN114138958B (en) 2021-11-25 2021-11-25 Information interaction method, device, equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111438910.4A CN114138958B (en) 2021-11-25 2021-11-25 Information interaction method, device, equipment and storage medium

Publications (2)

Publication Number Publication Date
CN114138958A CN114138958A (en) 2022-03-04
CN114138958B true CN114138958B (en) 2025-07-15

Family

ID=80389637

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111438910.4A Active CN114138958B (en) 2021-11-25 2021-11-25 Information interaction method, device, equipment and storage medium

Country Status (1)

Country Link
CN (1) CN114138958B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118897878B (en) * 2024-06-14 2025-12-02 上海涧意信息科技有限公司 An AI-based customer interaction service system
CN119648126B (en) * 2024-11-07 2025-10-10 百度在线网络技术（北京）有限公司 Delivery process control method, device and storage medium

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110704703A (en) * 2019-09-27 2020-01-17 北京百度网讯科技有限公司 Man-machine conversation method and device
CN111933134A (en) * 2020-07-23 2020-11-13 珠海大横琴科技发展有限公司 Man-machine interaction method and device, electronic equipment and storage medium

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103914548B (en) * 2014-04-10 2018-01-09 北京百度网讯科技有限公司 Information search method and device
CN110008322B (en) * 2019-03-25 2023-04-07 创新先进技术有限公司 Word recommendation method and device in multi-round dialogue scene
CN111428483B (en) * 2020-03-31 2022-05-24 华为技术有限公司 Voice interaction method and device and terminal equipment
CN113641807B (en) * 2021-07-28 2024-05-24 北京百度网讯科技有限公司 Training method, device, equipment and storage medium for dialogue recommendation model

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110704703A (en) * 2019-09-27 2020-01-17 北京百度网讯科技有限公司 Man-machine conversation method and device
CN111933134A (en) * 2020-07-23 2020-11-13 珠海大横琴科技发展有限公司 Man-machine interaction method and device, electronic equipment and storage medium

Also Published As

Publication number Publication date
CN114138958A (en) 2022-03-04

Similar Documents

Publication Publication Date Title
CN107741976B (en) Intelligent answering method, device, medium and electronic device
CN111767386B (en) Dialogue processing method, device, electronic equipment and computer readable storage medium
CN109002515A (en) A kind of method and apparatus of intelligent response
WO2021066903A1 (en) Providing explainable product recommendation in a session
CN112818689B (en) An entity recognition method, model training method and device
CN110597962A (en) Search result display method, device, medium and electronic equipment
CN114138958B (en) Information interaction method, device, equipment and storage medium
WO2025097987A1 (en) Model evaluation method, electronic device, and storage medium
CN112330400A (en) Product recommendation method and device, computer equipment and storage medium
CN118885592A (en) Agent-based information processing method, device, electronic device and storage medium
EP2613275B1 (en) Search device, search method, search program, and computer-readable memory medium for recording search program
CN113792129B (en) Intelligent session method, device, computer equipment and medium
CN115191002A (en) Matching system, matching method, and matching program
CN113486166B (en) Construction method, device and equipment of intelligent customer service robot and storage medium
CN118626717A (en) Fine-tuning method for large language model, resource recommendation method, device and equipment
CN118692450A (en) Question and answer processing method, device, equipment, storage medium and program product
CN109408679A (en) Method, apparatus, electronic equipment and the storage medium of intelligent management application program
CN108228950A (en) A kind of information processing method and device
CN110209804B (en) Target corpus determining method and device, storage medium and electronic device
US11399097B2 (en) Systems and methods for search based call routing
CN117473070B (en) Multi-channel application method of intelligent robot, intelligent robot and storage medium
CN110929014A (en) Information processing method, information processing device, electronic equipment and storage medium
CN118395227A (en) Social assistance method, social assistance device, electronic equipment and computer readable storage medium
US20160314701A1 (en) Method and system for managing a wordgraph
CN119474285A (en) Result generation method, device, equipment and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant