Question and answer interaction method, device and equipment for intelligent household appliance after-sales service and medium

Abstract

The invention belongs to the technical field of intelligent household appliances, and particularly relates to an intelligent household appliance after-sales service question and answer interaction method and device, equipment and a medium. According to the method, when a question message of a user is detected, a guide reply and a solution set corresponding to the question message are determined; sending a guide reply to the user, and receiving a first interaction message fed back by the user; based on the first interaction message, determining a target solution from a solution set, and feeding back the target solution to the user; according to the method, a guided solution flow can be provided for the user, and the interactive message is fed back based on the guided reply, so that the determination range of the solution can be narrowed, and the accuracy of the reply is improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of intelligent household appliances, and particularly relates to a question and answer interaction method and device for intelligent household appliance after-sales service, equipment and a medium. BACKGROUND

[0002] At present, a user can interact with an after-sales service platform through an application program on a mobile terminal. For example, if a user has a question during use of an intelligent household appliance, the user can ask the after-sales service platform a question, and the after-sales service platform gives a corresponding reply based on the question asked by the user.

[0003] In the prior art, the after-sales service platform usually first analyzes and processes a user question by using natural language processing technology, and determines a matching reply / solution from a question and answer library based on an analysis result, and then feeds back the reply / solution to the user.

[0004] Since the replies / solutions in the existing question and answer library are fixed replies / solutions, the reply fed back to the user may be inaccurate, and the existing solution cannot guide the user to make a more accurate description, so that the user needs to ask a question multiple times to obtain a corresponding reply, thereby affecting the user experience. SUMMARY

[0005] The application provides a question and answer interaction method and device for intelligent household appliance after-sales service, equipment and a medium, to solve the defect that the user cannot be guided to make a more accurate description in the existing solution, so that the user needs to ask a question multiple times to obtain a corresponding reply.

[0006] In a first aspect, the application provides a question and answer interaction method for intelligent household appliance after-sales service, which comprises the following steps:

[0007] In response to a question message sent by a user, a reply strategy corresponding to the question message is determined, the reply strategy comprising a guide reply and a corresponding solution set;

[0008] The guide reply is sent to the user, and a first interaction message fed back by the user is received;

[0009] A target solution is determined from the solution set based on the first interaction message, and the target solution is fed back to the user.

[0010] In a possible implementation manner, the number of guide replies is multiple, and there is a dependency relationship between at least two guide replies. The target solution is determined from the solution set based on the first interaction message, comprising the following steps:

[0011] Based on the guide reply and the first interaction message, a new guide reply is matched, and the new guide reply is sent to the user, so that the user feeds back a new first interaction message until a new guide reply cannot be matched, wherein there is a dependency relationship between the new guide reply and the old guide reply.

[0012] Based on the new guide reply and the new first interaction message, a target solution is determined from the solution set.

[0013] In a possible implementation, the solution set includes at least one guide reply corresponding solution subset, and the determining of the target solution from the solution set based on the new guide reply and the new first interaction message includes:

[0014] Determining whether there is a solution subset corresponding to the new guide reply in the solution set;

[0015] In the case that there is a solution subset corresponding to the new guide reply in the solution set, the target solution is determined from the solution subset based on the new first interaction message.

[0016] In a possible implementation, the determining of the reply strategy corresponding to the problem message includes:

[0017] Retrieving a first result associated with the problem message from a knowledge base;

[0018] Inputting the problem message into a question and answer model to obtain a second result output by the reply model, wherein the question and answer model is trained based on historical common question interaction data;

[0019] Determining the reply strategy based on the first result and the second result.

[0020] In a possible implementation, the determining of the reply strategy based on the first result and the second result includes:

[0021] Determining at least one guide reply based on the first result and the second result;

[0022] In the case that the number of guide replies is one, the guide reply and the corresponding solution are taken as the reply strategy;

[0023] In the case that the number of guide replies is more than one, a dependency relationship between the guide replies is determined, and the guide replies, the dependency relationship, and the solution corresponding to each guide reply are taken as the reply strategy.

[0024] In a possible implementation, before the inputting the question message into the question and answer model to obtain the second result output by the reply model, the method further includes:

[0025] obtaining historical common question interaction data, the historical common question interaction data including a plurality of historical common questions and corresponding historical solutions;

[0026] taking the plurality of historical common questions as input items and the corresponding historical solutions as output items to train the large language model to obtain the trained reply model.

[0027] In a possible implementation, the determining the reply strategy corresponding to the question message includes:

[0028] In response to a second interaction message sent by a user, performing semantic analysis on the second interaction message to obtain a message type of the second interaction message, wherein the message type includes a question type and a reply type.

[0029] In a case where the message type is the question type, determining a reply strategy corresponding to the second interaction message.

[0030] In a second aspect, the present application provides a question and answer interaction device for intelligent household appliance after-sales service, the device including:

[0031] a determining module configured to determine, in response to a question message sent by a user, a reply strategy corresponding to the question message, the reply strategy including a guided reply and a corresponding solution set.

[0032] a transceiving module configured to send the guided reply to the user and receive a first interaction message fed back by the user.

[0033] The determining module is further configured to determine, based on the first interaction message, a target solution from the solution set.

[0034] The transceiving module is further configured to feed back the target solution to the user.

[0035] In a possible implementation, the number of guided replies is a plurality, and there is a dependency relationship between at least two guided replies. The determining module is configured to match a new guided reply based on the guided reply and the first interaction message.

[0036] The transceiving module is further configured to send the new guided reply to the user, so that the user feeds back a new first interaction message until a new guided reply cannot be matched. There is a dependency relationship between the new guided reply and an old guided reply.

[0037] The determining module is configured to determine a target solution from the solution set based on the new guidance reply and the new first interaction message.

[0038] In a possible implementation, the solution set includes at least one solution subset corresponding to a new guidance reply, and the determining module is configured to determine whether there is a solution subset corresponding to the new guidance reply in the solution set, and determine a target solution from the solution subset based on the new first interaction message in the case where there is a solution subset corresponding to the new guidance reply in the solution set.

[0039] In a possible implementation, the determining module is configured to retrieve a first result associated with the question message from a knowledge base, input the question message into a question and answer model to obtain a second result output by the answer model, and determine the reply strategy based on the first result and the second result, where the question and answer model is trained based on historical common question and answer data.

[0040] In a possible implementation, the determining module is configured to determine at least one guidance reply based on the first result and the second result, determine the guidance reply and a corresponding solution as the reply strategy in the case where the number of guidance replies is one, determine a dependency relationship among a plurality of guidance replies in the case where the number of guidance replies is a plurality, and determine the plurality of guidance replies, the dependency relationship, and a solution corresponding to each guidance reply as the reply strategy.

[0041] In a possible implementation, the apparatus further includes a training module, and the transceiver is further configured to obtain historical common question and answer data, where the historical common question and answer data includes a plurality of historical common questions and corresponding historical solutions.

[0042] The training module is configured to take the plurality of historical common questions as input items and the corresponding historical solutions as output items to train the large language model to obtain the trained answer model.

[0043] In a possible implementation, the determining module is configured to perform semantic analysis on a second interaction message sent by a user to obtain a message type of the second interaction message in response to the second interaction message, where the message type includes a question type and an answer type, and determine a reply strategy corresponding to the second interaction message in the case where the message type is the question type.

[0044] In a third aspect, the present application provides a question and answer interaction device for intelligent household appliance after-sales service, comprising: a processor, and a memory in communication connection with the processor;

[0045] The memory stores computer execution instructions.

[0046] The processor executes the computer execution instructions stored in the memory to implement the question and answer interaction method for intelligent household appliance after-sales service as described in the first aspect and various possible implementation manners of the first aspect.

[0047] In a fourth aspect, the present application provides a computer storage medium having computer execution instructions stored thereon, the computer execution instructions being executed by a processor to implement the question and answer interaction method for intelligent household appliance after-sales service as described in the first aspect and various possible implementation manners of the first aspect.

[0048] In a fifth aspect, the present application further provides a computer program product comprising a computer program, the computer program being executed by a processor to implement the steps of the method according to any one of the first aspect.

[0049] The question and answer interaction method, device, equipment and medium for intelligent household appliance after-sales service provided by the present application can determine the guiding reply and the solution set corresponding to the question message when detecting the question message of the user, send the guiding reply to the user, receive the first interaction message fed back by the user, determine the target solution from the solution set based on the first interaction message, and feed back the target solution to the user. The method can provide a guided solution process for the user, and since the interaction message is fed back based on the guiding reply, the determination range of the solution can be narrowed, thereby improving the accuracy of the reply. BRIEF DESCRIPTION OF DRAWINGS

[0050] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0051] Figure 1 A scene schematic diagram of a question and answer interaction method for intelligent household appliance after-sales service provided by an embodiment of the present application;

[0052] Figure 2 A flow schematic diagram of a question and answer interaction method for intelligent household appliance after-sales service provided by an embodiment of the present application Figure 1 ;

[0053] Figure 3 A flow schematic diagram of a question and answer interaction method for intelligent household appliance after-sales service provided by an embodiment of the present application Figure 2 ;

[0054] Figure 4is a structural schematic diagram of a question and answer interactive device for intelligent household appliance after-sales service provided by an embodiment of the present application.

[0055] Figure 5 is a structural schematic diagram of a question and answer interactive device for intelligent household appliance after-sales service provided by an embodiment of the present application.

[0056] The specific embodiments of the present application have been shown through the above drawings, and will be described in more detail hereinafter. These drawings and the written description are not intended to restrict the scope of the present application in any way, but to illustrate the concept of the present application for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0057] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described clearly and completely below in combination with the drawings in the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without any creative work fall within the scope of protection of the present application.

[0058] The terms "first", "second", "third", "fourth" and the like (if any) in the description, claims and drawings of the present application are used to distinguish similar objects, and do not necessarily indicate a specific order or a chronological sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0059] In the embodiments of the present application, the words "exemplary" or "for example" are used to mean example, illustration or description. Any embodiment or design solution described as "exemplary" or "for example" in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design solutions. Rather, the use of "exemplary" or "for example" is intended to present the relevant concept in a specific manner.

[0060] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of the relevant data need to comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for the user to choose authorization or refusal.

[0061] At present, a user can interact with an after-sales service platform through an application on a mobile terminal. For example, if a user has a question during use of a smart home appliance, the user can ask the after-sales service platform, and the after-sales service platform gives a corresponding reply based on the question asked by the user.

[0062] In the prior art, the after-sales service platform usually first analyzes and processes a user question by using natural language processing technology, and determines a matching reply / solution from a question and answer library based on an analysis result, and then feeds back the reply / solution to the user.

[0063] However, the reply / solution in the existing question and answer library is usually a fixed reply / solution. For example, if the question is "what should be done if there is too much foam in a washing machine?", the existing reply is "if there is too much foam in the washing machine, the washing machine can be run at a cleaning level, and after cleaning, the problem of too much foam should be well solved, and if the foam still exists, the clothes can be manually washed and then washed again. This may cause the reply fed back to the user to be inaccurate, and the existing technology cannot guide the user to make a more accurate description, which causes the user to make multiple questions to obtain a corresponding reply, thereby affecting the user experience.

[0064] Therefore, how to improve the accuracy of the reply and provide a guided solution process for the user is a problem to be solved at present.

[0065] To solve the above problems, the present application provides a question and answer interaction method for after-sales service of a smart home appliance. When a question message of a user is detected, at least one guided reply corresponding to the question message and a solution set corresponding to the question message are determined, and the at least one guided reply is sent to the user to guide the user to feed back a corresponding interaction message, and then a corresponding solution is determined from the solution set based on the interaction message fed back by the user. The method can provide a guided solution process for the user, and since the interaction message is fed back based on the guided reply, the determination range of the solution can be narrowed, thereby improving the accuracy of the reply.

[0066] The technical solutions of the present application and how the technical solutions of the present application solve the above technical problems will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes can not be described again in some embodiments. The embodiments of the present application will be described below with reference to the accompanying drawings.

[0067] Figure 1 A scene diagram of the question and answer interaction method for after-sales service of a smart home appliance provided by the embodiments of the present application.

[0068] As Figure 1As shown, the server 2 can be deployed with an after-sales service platform or a user service platform, for example, and a user can send a service message to the server through the terminal device 3, which can be used to query relevant data of the smart home appliance 1, or can be used to consult possible abnormal conditions of the smart home appliance 1, etc., which is not limited in the present application. After the server 2 receives the service message sent by the user through the terminal device 3, it feeds back the corresponding reply scheme based on the service message.

[0069] Figure 2 Flowchart of the question and answer interaction method for smart home appliance after-sales service provided by the embodiment of the present application Figure 1 The execution subject of the embodiment can be, for example, Figure 2 The server 2 shown in the embodiment can also be a platform deployed on the server 2, etc. As Figure 3 As shown, the question and answer interaction method for smart home appliance after-sales service shown in the embodiment includes:

[0070] S201, in response to the question message sent by the user, determining the reply strategy corresponding to the question message, the reply strategy including a guide reply and a corresponding solution set.

[0071] The question message sent by the user can be, for example, a question about the current abnormality of the smart home appliance, such as: What should I do if the washing machine has too much foam? Or the washing machine makes too much noise during dehydration, etc.

[0072] The guide reply can be pre-set and stored in the database by the maintenance personnel or the after-sales personnel, or can be obtained based on semantic analysis of the question message.

[0073] Continuing the above example, if the question message is "the washing machine makes too much noise during dehydration", the corresponding guide reply can include, for example: "Is the noise source the inner drum?" Or "Is the noise source the outer drum?" Etc.

[0074] It can be understood that the guide reply is not a solution to the question message, but it can be an intermediate reply associated with the question message, or a guided reply.

[0075] The solution set can include, for example, at least one solution corresponding to the guide reply, or can include the solution of the question message, or can include some general reply schemes.

[0076] The number of solutions corresponding to one guiding reply can be one or multiple, which is not limited in the present application. The conventional reply solution is mainly used to feed back the reply solution to the user when an explicit solution cannot be given based on the user's feedback or the problem message. For example, it can be "please describe the problem again" or "an order has been placed for you, and the service engineer will contact you later".

[0077] In this step, if a problem message sent by the user is detected, at least one guiding reply and a solution set corresponding to the problem message can be determined based on the problem message.

[0078] In a possible implementation, the specific steps of determining the reply strategy corresponding to the problem message can be as follows:

[0079] Retrieving from the knowledge base whether there is a first result associated with the problem message, and inputting the problem message into the question and answer model to obtain a second result output by the question and answer model, and then determining the reply strategy based on the first result and the second result.

[0080] The knowledge base may, for example, be a database containing information of intelligent household appliances, which may, for example, be constructed based on the specifications of intelligent household appliances and other related materials. The question and answer model is used to understand and analyze the input question and answer message to output a reply result corresponding to the question and answer message.

[0081] The first result may, for example, include the specification content corresponding to the problem message, the solution of the problem, or the guiding reply associated with the problem message.

[0082] It can be understood that, in the case where the first result includes the guiding reply, the knowledge base stores a plurality of preset common problems and the guiding reply and solution corresponding to each common problem.

[0083] The second result may, for example, include at least one guiding reply associated with the problem message and at least one solution corresponding to each guiding reply.

[0084] In a possible implementation, the specific implementation of determining the reply strategy based on the first result and the second result may, for example, be as follows:

[0085] Determining at least one guiding reply based on the first result and the second result.

[0086] The first result and the second result may each include at least one guidance reply. When determining the guidance reply, the first result and the second result may be aggregated first, and then at least one guidance reply and a corresponding solution are selected from the aggregated result. It can be understood that the priority of the second result is higher than the priority of the first result.

[0087] If the number of selected guidance replies is one, the guidance reply and the corresponding solution are taken as the reply strategy.

[0088] If the number of guidance replies is more than one, a dependency relationship between the guidance replies is determined, and the guidance replies, the dependency relationship, and a corresponding solution of each guidance reply are taken as the reply strategy.

[0089] It can be understood that the dependency relationship between the guidance replies may be pre-set by the after-sales personnel, or may be obtained by big data analysis based on actual interaction data. The present application does not limit this.

[0090] For example, the guidance replies include “Is the part that makes noise the inner cylinder?” and “Is the part that makes noise the outer cylinder?” There is a dependency relationship between the two. Only when the user replies “No” to “Is the part that makes noise the inner cylinder?” will the user be asked “Is the part that makes noise the outer cylinder?”

[0091] In a possible implementation, the question and answer model may be obtained by training based on historical common question interaction data. The following explains a possible training process of the question and answer model:

[0092] Obtain historical common question interaction data;

[0093] The historical common question interaction data may be, for example, frequently asked questions and corresponding historical replies in historical question and answer data, or may be interaction data determined based on big data. It may include, for example, a plurality of historical common questions and corresponding historical solutions.

[0094] The plurality of historical common questions are taken as input items, and the corresponding historical solutions are taken as output items. Then, the big language model is trained based on the input items and the output items, so that the big language model understands the association between the input items and the output items, and sorts the association between the plurality of historical common questions, thereby outputting the corresponding solutions.

[0095] In a possible implementation, before determining the reply strategy, it is also necessary to determine whether the message sent by the user is a question message. The specific process is as follows:

[0096] receive a second interaction message sent by the user, and perform semantic analysis on the second interaction message to obtain a message type of the second interaction message.

[0097] The message type includes a question type and a reply type.

[0098] In a case where the message type is the question type, it is determined that the second interaction message is a question message, and then step S201 in the above embodiment is performed.

[0099] S202, send the guided reply to the user, and receive a first interaction message fed back by the user.

[0100] It can be understood that when using the smart home appliance, the user may encounter various problems. When a problem occurs, the user usually cannot ask an accurate question because the user is not clear about the main cause of the problem, and thus the user gives a relatively broad question. Therefore, the step can send a guided reply corresponding to the question message to the user, so that the user feeds back a corresponding first interaction message.

[0101] Since the first interaction message is fed back based on the guided reply, the first interaction message can narrow down the problem range.

[0102] For example, the guided reply is "Is the noise-emitting part the inner drum?", and the first interaction message fed back by the user is "Yes". At this time, the question message "The washing machine makes too much noise during dehydration" can be narrowed down to "The inner drum of the washing machine makes too much noise during dehydration". The purpose of this is to feed back a more accurate reply scheme to the user based on the narrowed-down question message, so as to facilitate the user to solve the problem based on the reply scheme.

[0103] S203, determine a target solution from the solution set based on the first interaction message, and feed back the target solution to the user.

[0104] After obtaining the first interaction message, whether there is a corresponding target solution in the solution set can be determined based on the first interaction message, and the target solution is fed back to the user in a case where the target solution exists.

[0105] The question and answer interaction method for the intelligent household appliance after-sales service provided in the embodiment can provide a guided solution process for the user, and since the interaction message is fed back based on the guided reply, the determination range of the solution can be narrowed, thereby improving the accuracy of the reply.

[0106] Figure 2 The flowchart of the question and answer interaction method for the intelligent household appliance after-sales service provided in the embodiment of the application Figure 2 In the embodiment, the number of guided replies is multiple, and there is a dependency relationship between at least two guided replies. The embodiment is based on the embodiment Figure 4 The embodiment provides a possible implementation of determining the target solution from the solution set based on the first interaction message. The question and answer interaction method for the intelligent household appliance after-sales service provided in the embodiment includes the following steps.

[0107] S301, matching a new guided reply based on the guided reply and the first interaction message.

[0108] The new guided reply has a dependency relationship with the old guided reply. In the case where there are multiple guided replies, it can be determined based on the guided reply whether there is a new guided reply matching the first interaction message in the other guided replies.

[0109] Continue to illustrate by taking the above embodiment as an example: the position of the noise of the washing machine may be in the inner drum or in the outer drum. When the user asks, "The noise of the washing machine is too loud during dehydration", the guided reply "Do you ask which part of the noise is the inner drum?" can be fed back to the user.

[0110] When the user replies to "Do you ask which part of the noise is the inner drum?" as "No", it indicates that the position of the noise is not the inner drum. At this time, "Do you ask which part of the noise is the outer drum?" which has a dependency relationship with "Do you ask which part of the noise is the inner drum?" can be determined as a new guided reply based on "Do you ask which part of the noise is the inner drum?".

[0111] S302, sending the new guided reply to the user and receiving the new first interaction message fed back by the user.

[0112] S303, determining whether there is a guided reply matching the new guided reply and the new first interaction message; if yes, re-executing step S302, and if no, executing step S304.

[0113] Wherein, after obtaining the new first interaction message of the user feedback, it is necessary to determine whether a new guide reply with a dependency relationship is matched based on the new first interaction message and the corresponding guide reply, if yes, re-execute step S302 until a new guide reply cannot be matched.

[0114] It can be understood that each time a new guide reply is matched, the range of solutions is narrowed.

[0115] If a new guide reply cannot be matched, it means that the current guide reply is the optimal guide reply or no other guide reply depends on the current guide reply. At this time, the target solution can be determined from the solution set based on the current guide reply and the first interaction message corresponding to the current guide reply.

[0116] S304, determining the target solution from the solution set based on the new guide reply and the new first interaction message.

[0117] In one possible implementation, the solution set includes at least one solution subset corresponding to a guide reply; when determining the target solution, it can be determined whether there is a solution subset corresponding to the new guide reply in the solution set.

[0118] In the case that there is a solution subset corresponding to the new guide reply in the solution set, the target solution is determined from the solution subset based on the new first interaction message.

[0119] Wherein, the solution subset corresponding to the guide reply includes at least two solutions.

[0120] If there is a solution subset corresponding to the current guide reply in the solution set, the target solution can be determined from the multiple solutions corresponding to the current guide reply based on the first interaction message.

[0121] For example, the guide reply is "Please check if there is a lot of foam in the washing machine cylinder or detergent box?", and the corresponding solutions include: 1. "Please use it with confidence. This is a protection prompt of the washing machine detecting too much detergent foam, which is not a fault." 2. "The detection device may be abnormal, please repair".

[0122] If the first interaction message is "no", it means that there is no too much foam phenomenon in the inner cylinder and the detergent box of the washing machine, and at this time the target solution can be determined as solution 1.

[0123] If the first interaction message is "yes", it means that there is a lot of foam in the inner cylinder and the detergent box of the washing machine, and at this time the target solution can be determined as solution 2.

[0124] The method for question and answer interaction of intelligent household appliance after-sales service provided by the embodiment can improve the clarity of the description of the problem message by the user by sending a plurality of guided replies having a dependency relationship to the user, and the user feeds back corresponding interactive messages based on the plurality of guided replies, and then determines a solution based on the new guided reply and the corresponding new interactive message. The method can provide a guided solution process for the user, and the new interactive message replied by the user each time can more clearly describe the problem of the intelligent household appliance, thereby achieving the effect of narrowing the determination range of the solution and improving the accuracy of the reply when determining the solution based on the new interactive message.

[0125] Figure 4 The structural schematic diagram of the question and answer interaction device for intelligent household appliance after-sales service provided by the present application is shown in FIG. 1. Figure 5 As shown in FIG. 1, the present application provides a question and answer interaction device for intelligent household appliance after-sales service, which comprises:

[0126] A determination module 401 is configured to determine a reply strategy corresponding to a problem message sent by a user in response to the problem message, wherein the reply strategy comprises guided replies and a solution set corresponding to the guided replies.

[0127] A transceiver module 402 is configured to send the guided replies to the user and receive a first interactive message fed back by the user.

[0128] The determination module 401 is further configured to determine a target solution from the solution set based on the first interactive message.

[0129] The transceiver module 402 is further configured to feed back the target solution to the user.

[0130] In a possible implementation, the number of guided replies is a plurality, and there is a dependency relationship between at least two guided replies. The determination module 401 is configured to match a new guided reply based on the guided replies and the first interactive message.

[0131] The transceiver module 402 is further configured to send the new guided reply to the user, so that the user feeds back a new first interactive message until a new guided reply cannot be matched, wherein there is a dependency relationship between the new guided reply and the old guided reply.

[0132] The determination module 401 is configured to determine a target solution from the solution set based on the new guided reply and the new first interactive message.

[0133] In a possible implementation, the solution set includes at least one guided reply corresponding solution subset, the determination module 401 is configured to determine whether there is a new guided reply corresponding solution subset in the solution set, and in the case where there is a new guided reply corresponding solution subset in the solution set, determine a target solution from the solution subset based on the first interaction message.

[0134] In a possible implementation, the determination module 401 is configured to retrieve a first result associated with the question message from a knowledge base, input the question message into a question and answer model to obtain a second result output by the answer model, and determine the reply strategy based on the first result and the second result, where the question and answer model is trained based on historical common question interaction data.

[0135] In a possible implementation, the determination module 401 is configured to determine at least one guided reply based on the first result and the second result, in the case where the number of guided replies is one, take the guided reply and a corresponding solution as the reply strategy, and in the case where the number of guided replies is more than one, determine a dependency relationship between the guided replies, and take the guided replies, the dependency relationship, and a solution corresponding to each guided reply as the reply strategy.

[0136] In a possible implementation, the apparatus further includes a training module 403, and the transceiver module 402 is further configured to obtain historical common question interaction data, the historical common question interaction data including a plurality of historical common questions and corresponding historical solutions.

[0137] The training module 403 is configured to take the plurality of historical common questions as input items and the corresponding historical solutions as output items to train the large language model to obtain the trained reply model.

[0138] In a possible implementation, the determination module 401 is configured to, in response to a second interaction message sent by a user, perform semantic analysis on the second interaction message to obtain a message type of the second interaction message, where the message type includes a question type and an answer type, and in the case where the message type is the question type, determine a reply strategy corresponding to the second interaction message.

[0139] The question and answer interaction apparatus for intelligent home appliance after-sales service provided in this embodiment can execute the question and answer interaction method for intelligent home appliance after-sales service provided in the method embodiments, and has similar implementation principles and technical effects, which will not be described here in detail.

[0140] Figure 5 A structural schematic diagram of an intelligent home appliance after-sales service question and answer interaction device provided by an embodiment of the present application is shown in FIG. 1. As shown in the figure, the intelligent home appliance after-sales service question and answer interaction device 500 can include at least one processor 501, at least one storage medium 502, and a communication interface 503. The at least one processor 501 is configured to implement the method provided by the above-described embodiments of the present application. Figure 5

[0141] The at least one storage medium 502 is configured to store computer execution instructions and / or data. The storage medium 502 is coupled to the processor 501. The coupling in the embodiment of the present application is indirect coupling or communication connection between devices, units, or modules, which can be electrical, mechanical, or other forms, and is used for information interaction between devices, units, or modules. The processor 501 can operate in cooperation with the storage medium 502. The processor 501 can execute the computer execution instructions stored in the storage medium 502. At least one of the at least one storage medium can be included in the processor.

[0142] The communication interface 503 is configured to communicate with other devices through a transmission medium, so that the device 500 can communicate with other devices. The communication interface 503 can be, for example, a transceiver, an interface, a bus, a circuit, or a device capable of realizing the transceiving function. The microprocessor 501 can utilize the communication interface 503 to transceive data and / or information, and is configured to implement the method provided by the above-described embodiments of the present application. For details, refer to the detailed description in the foregoing embodiments, which will not be repeated here.

[0143] The specific connection medium between the microprocessor 501, the storage medium 502, and the communication interface 503 is not limited in the embodiment of the present application. In the embodiment of the present application, the processor 501, the storage medium 502, and the communication interface 503 are connected through a bus 504. Figure 5 Figure 5 The connection mode between other components is only schematically illustrated, and is not limited. The bus can be divided into an address bus, a data bus, a control bus, etc. For convenience of representation, only one thick line is used to represent the bus in the figure, but it does not mean that there is only one bus or only one type of bus. ​

[0144] The embodiment of the present application further provides a computer storage medium having a computer program stored thereon. The computer program is executed by the processor to implement the technical solution of the above-described method embodiment, and has similar implementation principles and technical effects, which will not be repeated here.

[0145] The embodiment of the present application further provides a computer program product including a computer program. The computer program is executed by the processor to implement the technical solution of the above-described method embodiment, and has similar implementation principles and technical effects, which will not be repeated here.​​​

[0146] It should be noted that, for the foregoing method embodiments, the method is described as a series of acts for simplicity, but the application is not limited to the order of the acts for performing the application, as some steps can be performed in other order or simultaneously. Moreover, those skilled in the art will understand that the embodiments described in the specification are all optional embodiments of the application, and the acts and modules involved are not necessarily essential to the application.

[0147] It should be further noted that, although the steps in the flowchart are shown in order according to the arrows, the steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, the execution of the steps is not strictly limited in order, and the steps can be executed in other order. Moreover, at least some of the steps in the flowchart can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of the sub-steps or stages is not necessarily sequential, but can be performed alternately or alternately with at least part of other steps or sub-steps or stages of other steps.

[0148] It should be understood that the above-described device embodiments are only schematic, and the device of the application can also be implemented in other ways. For example, the division of units / modules in the above-described embodiments is only a logical function division, and actual implementation can have another division manner. For example, multiple units / modules or components can be combined, or can be integrated into another system, or some features can be omitted or not executed.

[0149] In addition, unless otherwise specified, each functional unit / module in each embodiment of the application can be integrated in one unit / module, or each unit / module can exist physically, or two or more units / modules can be integrated together. The integrated unit / module can be realized in the form of hardware or in the form of a software program module.

[0150] If the integrated units / modules are implemented in the form of hardware, the hardware can be a digital circuit, an analog circuit, etc. The physical implementation of the hardware structure includes, but is not limited to, transistors, memristors, etc. Unless otherwise specified, the processor can be any appropriate hardware processor, such as a CPU, a GPU, an FPGA, a DSP, an ASIC, etc. Unless otherwise specified, the storage unit can be any appropriate magnetic storage medium or magneto-optical storage medium, such as resistive random access memory (RRAM), dynamic random access memory (DRAM), static random access memory (SRAM), enhanced dynamic random access memory (EDRAM), high-bandwidth memory (HBM), hybrid memory cube (HMC), etc.

[0151] If the integrated units / modules are implemented in the form of software program modules and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the essential part or all or part of the technical solutions that make contributions to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the embodiments of the present application. The aforementioned storage medium includes a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0152] In the above embodiments, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments. The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described, but as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the present application.

[0153] Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. It is intended that the specification and examples be considered as exemplary only, with the true scope and spirit of the application being indicated by the following claims.

[0154] It is to be understood that the application is not limited to the precise construction herein disclosed and shown in the drawings, and that various changes in shape, size and arrangements of parts can be made without departing from the scope of the application. The scope of the application is limited only by the claims that follow.

Claims

1. A method for question and answer interaction of intelligent household appliance after-sales service, characterized in that, The method comprises: in response to a question message sent by a user, determining a reply strategy corresponding to the question message, the reply strategy comprising: a guide reply and a corresponding solution set; sending the guide reply to the user and receiving a first interaction message fed back by the user; determining a target solution from the solution set based on the first interaction message and feeding back the target solution to the user.

2. The method of claim 1, wherein, The number of guide replies is multiple, and there is a dependency relationship between at least two guide replies. Determining a target solution from the solution set based on the first interaction message comprises: based on the guide reply and the first interaction message, matching a new guide reply and sending the new guide reply to the user to make the user feed back a new first interaction message until a new guide reply cannot be matched, wherein there is a dependency relationship between the new guide reply and the old guide reply; determining a target solution from the solution set based on the new guide reply and the new first interaction message.

3. The method of claim 2, wherein, The solution set comprises: at least one guide reply corresponding to a solution subset. Determining a target solution from the solution set based on the new guide reply and the new first interaction message comprises: determining whether there is a solution subset corresponding to the new guide reply in the solution set; if there is a solution subset corresponding to the new guide reply in the solution set, determining a target solution from the solution subset based on the new first interaction message.

4. The method of claim 1, wherein, The determination of the reply strategy corresponding to the question message comprises: retrieving a first result associated with the question message from a knowledge base; inputting the question message into a question and answer model to obtain a second result output by the reply model, wherein the question and answer model is trained based on historical common question interaction data; determining the reply strategy based on the first result and the second result.

5. The method of claim 4, wherein, The determination of the reply strategy based on the first result and the second result comprises: determining at least one guide reply based on the first result and the second result; if the number of guide replies is one, the guide reply and the corresponding solution are taken as the reply strategy; if the number of guide replies is multiple, the dependency relationship between multiple guide replies is determined, and multiple guide replies, the dependency relationship and the solution corresponding to each guide reply are taken as the reply strategy.

6. The method of claim 4, wherein, Before inputting the question message into the question and answer model to obtain the second result output by the reply model, the method further comprises: obtaining historical common question interaction data, the historical common question interaction data comprising: multiple historical common questions and corresponding historical solutions; taking multiple historical common questions as input items and corresponding historical solutions as output items to train a large language model to obtain a trained reply model.

7. The method of claim 1, wherein, The determination of the reply strategy corresponding to the question message comprises: In response to a second interaction message sent by a user, performing semantic analysis on the second interaction message to obtain a message type of the second interaction message, wherein the message type comprises a question type and a reply type; In a case where the message type is the question type, determining a reply strategy corresponding to the second interaction message.

8. An intelligent home appliance after-sales service question and answer interaction device, characterized in that, The apparatus comprises: A determining module configured to determine, in response to a question message sent by a user, a reply strategy corresponding to the question message, the reply strategy comprising a guide reply and a corresponding solution set; A transceiving module configured to send the guide reply to the user and receive a first interaction message fed back by the user; The determining module is further configured to determine, based on the first interaction message, a target solution from the solution set; The transceiving module is further configured to feed back the target solution to the user.

9. An intelligent home appliance after-sales service question and answer interaction device, characterized in that, Comprise: A memory; A processor; The memory stores computer execution instructions; The processor executes the computer execution instructions stored in the memory to implement the question and answer interaction method for the smart home appliance after-sales service according to any one of claims 1-7.

10. A computer storage medium, characterized in that, The computer storage medium stores computer execution instructions, and the computer execution instructions are executed by the processor to implement the question and answer interaction method for the smart home appliance after-sales service according to any one of claims 1-7.

Images