Multi-agent-based interaction method and multi-agent-based interaction system

Abstract

The embodiment of the invention discloses an interaction method based on multiple agents and an interaction system based on multiple agents. According to the embodiment of the invention, after a terminal sends inquiry information to a server, a health consultation agent at a server side determines a processing mode of a current consultation round according to the inquiry information; the drug recommendation agent obtains a drug use suggestion scheme of the user according to the inquiry content and inquiry information of the previous consultation round, and then the information integration agent obtains health consultation feedback information including drug recommendation instructions corresponding to the drug use suggestion scheme according to the drug use suggestion scheme and sends the health consultation feedback information to the terminal. And after receiving the health consultation feedback information, the terminal displays the health consultation feedback information. According to the embodiment of the invention, medicine recommendation can be performed on the user based on a multi-agent interaction mode, so that the medicine purchasing efficiency of the user can be improved when the user cannot judge the required medicine, and the medicine purchasing experience of the user is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular to an interaction method based on multi-agent and an interaction system based on multi-agent. BACKGROUND

[0002] With the continuous development of computer technology and Internet technology, users can purchase medicines through online channels or offline channels. When purchasing medicines through online channels, users usually need to determine which type of medicine to purchase according to their own symptoms, but if the user cannot clearly determine which type of medicine to purchase, they usually need to spend some time consulting a doctor about the medication plan before purchasing the appropriate medicine, which will reduce the user's shopping experience. SUMMARY

[0003] Therefore, the embodiments of the present application provide an interaction method based on multi-agent and an interaction system based on multi-agent to recommend medicines to users in a multi-agent interaction manner, thereby improving the user's medicine selection efficiency and shopping experience when the user cannot determine the required medicine.

[0004] In a first aspect, the embodiments of the present application provide an interaction method based on multi-agent, applicable to a server, comprising: In response to receiving current inquiry information of a target user in a current consultation round, a health consultation agent determines a processing mode of the current consultation round according to the current inquiry information, the processing mode including ending health consultation, outputting question information, or determining a medication suitability result, the medication suitability result being used to represent whether the target user is suitable for medication; In response to the processing mode including ending health consultation, a medicine recommendation agent obtains a medication recommendation scheme of the target user according to the inquiry content of the previous consultation round and the current inquiry information; In response to obtaining the medication recommendation scheme, an information integration agent obtains health consultation feedback information according to the medication recommendation scheme, the health consultation feedback information including medicine recommendation instructions corresponding to the medication recommendation scheme; Sending the health consultation feedback information to a terminal.

[0005] In a second aspect, the embodiments of the present application provide an interaction method based on multi-agent, applicable to a terminal, comprising: Sending current inquiry information; In response to receiving the health consultation feedback information of the current inquiry information, the health consultation feedback information is displayed, the health consultation feedback information is obtained by an information integration agent according to a drug use recommendation scheme for a target user, the drug use recommendation scheme is obtained by a drug recommendation agent according to inquiry content of a previous consultation round and the current inquiry information, the inquiry content is determined according to a processing mode of a current consultation round, the processing mode is determined by a health consultation agent according to the current inquiry information, the processing mode includes ending health consultation, outputting question information, or determining a drug use suitability result, the drug use suitability result is used to represent whether the target user is suitable for drug use, and the health consultation feedback information includes drug recommendation instructions corresponding to the drug use recommendation scheme.

[0006] In a third aspect, an embodiment of the present application provides an interaction method based on multiple agents, and the method comprises the following steps: sending current inquiry information; In response to receiving current inquiry information of a target user in a current consultation round, a health consultation agent determines a processing mode of the current consultation round according to the current inquiry information, the processing mode includes ending health consultation, outputting question information, or determining a drug use suitability result, the drug use suitability result is used to represent whether the target user is suitable for drug use; In response to the processing mode including ending health consultation, a drug recommendation agent obtains a drug use recommendation scheme for the target user according to inquiry content of a previous consultation round and the current inquiry information; In response to obtaining the drug use recommendation scheme, an information integration agent obtains health consultation feedback information according to the drug use recommendation scheme, the health consultation feedback information includes drug recommendation instructions corresponding to the drug use recommendation scheme; sending the health consultation feedback information to a terminal; In response to receiving the health consultation feedback information of the current inquiry information, the health consultation feedback information is displayed.

[0007] In a fourth aspect, an embodiment of the present application provides an interaction system based on multiple agents, and the system comprises the following steps: a server configured to, in response to receiving current inquiry information of a target user in a current consultation round, determine, by a health consultation intelligent agent, a processing mode of the current consultation round according to the current inquiry information, in response to the processing mode including ending health consultation, acquire, by a drug recommendation intelligent agent, a drug recommendation scheme of the target user according to inquiry content of a previous consultation round and the current inquiry information, in response to acquiring the drug recommendation scheme, acquire, by an information integration intelligent agent, health consultation feedback information according to the drug recommendation scheme, and send the health consultation feedback information to a terminal, the processing mode including ending health consultation, outputting question information or determining a drug suitability result, the drug suitability result used to represent whether the target user is suitable for drug use, and the health consultation feedback information including drug recommendation instructions corresponding to the drug recommendation scheme; and a terminal configured to send current inquiry information, and in response to receiving health consultation feedback information of the current inquiry information, display the health consultation feedback information.

[0008] In a fifth aspect, an embodiment of the present application provides an interactive device based on multiple intelligent agents, suitable for a server, the device comprising: a processing mode determination unit configured to, in response to receiving current inquiry information of a target user in a current consultation round, determine, by a health consultation intelligent agent, a processing mode of the current consultation round according to the current inquiry information, the processing mode including ending health consultation, outputting question information or determining a drug suitability result, the drug suitability result used to represent whether the target user is suitable for drug use; a scheme determination unit configured to, in response to the processing mode including ending health consultation, acquire, by a drug recommendation intelligent agent, a drug recommendation scheme of the target user according to inquiry content of a previous consultation round and the current inquiry information; a feedback information determination unit configured to, in response to acquiring the drug recommendation scheme, acquire, by an information integration intelligent agent, health consultation feedback information according to the drug recommendation scheme, the health consultation feedback information including drug recommendation instructions corresponding to the drug recommendation scheme; an information sending unit configured to send the health consultation feedback information to a terminal.

[0009] In a sixth aspect, an embodiment of the present application provides an interactive device based on multiple intelligent agents, suitable for a terminal, the device comprising: an information sending unit configured to send current inquiry information; The information receiving unit is configured to display health consultation feedback information in response to receiving the current inquiry information, the health consultation feedback information being obtained by an information integration agent according to a medication recommendation scheme for the target user, the medication recommendation scheme being obtained by a drug recommendation agent according to inquiry content of a previous consultation round and the current inquiry information, the inquiry content being determined according to a processing mode of the current consultation round, the processing mode being determined by the health consultation agent according to the current inquiry information, the processing mode including ending the health consultation, outputting question information, or determining a medication suitability result, the medication suitability result being used to represent whether the target user is suitable for medication, and the health consultation feedback information including drug recommendation instructions corresponding to the medication recommendation scheme.

[0010] In a seventh aspect, an electronic device is provided, including a memory and a processor, the memory being configured to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method according to any one of the first aspect to the third aspect.

[0011] In an eighth aspect, a computer readable storage medium is provided, the computer readable storage medium storing a computer program, the computer program being executed by a processor to implement the method according to any one of the first aspect to the third aspect.

[0012] In a ninth aspect, a computer program product is provided, the computer program product including a computer program / instruction, the computer program / instruction being executed by a processor to implement the method according to any one of the first aspect to the third aspect.

[0013] After the terminal sends the inquiry information to the server, the health consultation agent of the server side determines the processing mode of the current consultation round according to the inquiry information. If it is determined that the current processing mode is to end the health consultation, the drug recommendation agent obtains a medication recommendation scheme for the user according to the inquiry content of the previous consultation round and the inquiry information. Then, the information integration agent obtains health consultation feedback information including drug recommendation instructions corresponding to the medication recommendation scheme according to the medication recommendation scheme, and sends the health consultation feedback information to the terminal. After the terminal receives the health consultation feedback information, the health consultation feedback information is displayed. The embodiments of the present application can recommend drugs to the user based on the multi-agent interaction mode, so as to improve the drug selection efficiency of the user when the user cannot judge the required drug, and improve the drug purchase experience of the user. BRIEF DESCRIPTION OF DRAWINGS

[0014] The above and other objects, features and advantages of the present application will become more apparent from the following description when taken in conjunction with the accompanying drawings, in which: Figure 1 is a hardware architecture schematic diagram of a multi-agent based interaction system of an embodiment of the present application; Figure 2 is a structural block diagram of an intelligent interaction system of an embodiment of the present application; Figure 3 is a flowchart of a multi-agent based interaction method of an embodiment of the present application; Figure 4 is a flowchart of a multi-agent based interaction method of an embodiment of the present application; Figure 5 is a flowchart of a multi-agent based interaction method of an embodiment of the present application; Figure 6 is a flowchart of a multi-agent based interaction method of an embodiment of the present application; Figure 7 is a flowchart of a multi-agent based interaction method of an embodiment of the present application; Figure 8 is a flowchart of a multi-agent based interaction method of an embodiment of the present application; Figure 9 is a flowchart of a multi-agent based interaction method of an embodiment of the present application; Figure 10 is a flowchart of a multi-agent based interaction method of an embodiment of the present application; Figures 11-12 is a multi-agent interaction process schematic diagram of an embodiment of the present application; Figure 13 is a flowchart of a multi-agent based interaction method of an embodiment of the present application; Figure 14 is a flowchart of a multi-agent based interaction method of an embodiment of the present application; Figure 15 is a flowchart of a multi-agent based interaction method of an embodiment of the present application on the server side; Figure 16 is a flowchart of a multi-agent based interaction method of an embodiment of the present application on the terminal side; Figure 17 is a schematic diagram of a multi-agent based interaction device of an embodiment of the present application; Figure 18 is a schematic diagram of a multi-agent based interaction device of an embodiment of the present application; Figure 19 is a schematic diagram of an electronic device of an embodiment of the present application. DETAILED DESCRIPTION

[0015] The present application is described in the following based on examples, but the present application is not limited to these examples only. In the following detailed description of the present application, some specific details are described in detail. The present application can also be fully understood without the description of these details by those skilled in the art. In order to avoid obscuring the essence of the present application, well-known methods, processes, procedures, elements and circuits are not described in detail.

[0016] In addition, those of ordinary skill in the art will appreciate that the drawings provided herein are for illustrative purposes and are not necessarily drawn to scale.

[0017] Unless the context clearly requires otherwise, throughout the description, the words "comprise", "comprising", and the like are to be construed in an inclusive sense as opposed to an exclusive or exhaustive sense; that is to say, in the sense of "including, but not limited to".

[0018] In the description of the present application, it should be understood that the terms "first", "second" and the like are only for the purpose of description and cannot be understood as indicating or implying relative importance. In addition, in the description of the present application, unless otherwise stated, the meaning of "multiple" is two or more.

[0019] The technical scheme of the present application can be applied to the transaction and distribution service of instant e-commerce platforms, such as Taobao flash shopping, Taofxian, Eleme takeout and retail, etc. The technical scheme of the present application can combine user input and provide non-diagnostic medical knowledge consultation and medication safety auxiliary service to users based on multi-agent interaction. The content output by the intelligent interaction system should not be used as the basis for disease diagnosis, treatment or any legally binding written proof documents, and cannot replace the judgment of doctors.

[0020] Online drug purchase brings great convenience to users. In order to facilitate users to purchase drugs according to symptoms, some instant e-commerce platforms will provide online doctor consultation entrance. When purchasing drugs through online channels, if the user cannot clearly judge which type of drug should be purchased, the user can consult the doctor through the online doctor consultation entrance to consult the medication scheme. In this case, there may be a situation that a large number of users consult at the same time, and the user needs to wait for a period of time and complete the consultation before purchasing the symptomatic drug, which will reduce the user's drug purchase experience. Some instant e-commerce platforms do not provide online doctor consultation entrance, and if the user does not have basic medical knowledge, it is difficult to purchase the symptomatic drug. Therefore, the existing online drug purchase method will reduce the user's drug purchase experience.

[0021] In order to solve the above problems, the embodiment of the present application provides an interactive method based on multi-agent and an interactive system based on multi-agent, to recommend drugs to users in a multi-agent interaction manner, so as to improve the drug selection and purchase efficiency of users when they cannot judge the required drugs, and improve the drug purchase experience of users.

[0022] Figure 1 is a schematic diagram of the hardware architecture of the multi-agent based interaction system of the embodiments of the present application. As shown in Figure 1 the multi-agent based interaction system of the embodiments of the present application includes a server 11 and a terminal 12 on the side of an instant e-commerce platform.

[0023] The server 11 and the terminal 12 can establish a communication connection through a network or other communication means, so as to realize the interaction of information and data. It should be understood that, although Figure 1 only a certain number of servers 11 and terminals 12 are shown, this does not mean to limit the respective number, and multiple servers and multiple terminals can be included in the system.

[0024] The server 11 should be understood as a device providing data processing, database, communication facilities. For example, the server 11 can refer to a single physical server with related communication and data storage and database facilities, or can refer to a collection of networked or clustered processors, related networks and storage devices, and operate on software and one or more database systems and application software provided by the services of the supporting server. The server 11 can be a monolithic server or a distributed server across multiple computers or computer data centers, or can be various types of cloud servers. In some embodiments, each server can include hardware, software, or an embedded logic component or a combination of two or more such components for performing appropriate functions supported or implemented by the server.

[0025] The terminal 12 is a communication terminal capable of running computer programs. These communication terminals can be mobile phones, tablet computers, palmtop computers, wearable devices, vehicle integrated in-vehicle terminals, etc. The terminal 12 has a communication module capable of wired or wireless communication. In some embodiments, the terminal 12 includes at least one remote communication module, such as a communication circuit for WLAN, GPRS, 2G / 3G / 4G / 5G remote communication. The terminal 12 also has a display device and an input device, the display device can be a liquid crystal display, an LED display or a projection device. The input device may, for example, include a touch screen, a key, a pressure sensor, etc., and the terminal 12 receives the user's instructions through the input device and interacts with the user through the display device.

[0026] The embodiments of the present application are based on an intelligent interaction system to assist users in completing the drug purchase process. The intelligent interaction system 20 of the embodiments of the present application is deployed on the server 11 side, and the intelligent interaction system 20 includes a plurality of agents (agents), each agent can autonomously interact information based on a pre-set interaction sequence.

[0027] Figure 2 is a structural block diagram of the intelligent interaction system of the embodiments of the present application. As shown inFigure 2 As shown, the intelligent interaction system of the embodiment of the present application can include a health consultation intelligent agent 21, a drug recommendation intelligent agent 22 and an information integration intelligent agent 23. After receiving the current inquiry information of the user in the current consultation round, the health consultation intelligent agent 21 determines the processing mode of the current consultation round according to the current inquiry information, the processing mode including determining a health consultation judgment result, outputting question information or determining a drug suitability result, the health consultation judgment result being used to represent whether to end the health consultation, and the drug suitability result being used to represent whether the target user is suitable for drug taking. When the processing mode determined by the health consultation intelligent agent 21 includes determining the health consultation judgment result, and the health consultation judgment result is to end the health consultation, the drug recommendation intelligent agent 22 obtains a drug taking suggestion scheme of the user according to the inquiry content of the previous consultation round and the current inquiry information. After obtaining the drug taking suggestion scheme of the user, the information integration intelligent agent 23 obtains health consultation feedback information according to the drug taking suggestion scheme of the user, the health consultation feedback information including the drug recommendation instructions corresponding to the drug taking suggestion scheme. Therefore, the server 11 can send the health consultation feedback information to the terminal 12.

[0028] In an optional implementation manner of the embodiment of the present application, the intelligent interaction system 20 can further include an intention recognition intelligent agent (not shown in the figure) and an inquiry intelligent agent (not shown in the figure). The intention recognition intelligent agent can determine whether the target user initiates the inquiry for the first time, and if so, the user can choose to interact with the inquiry intelligent agent; if not, the user can choose to interact with the health consultation intelligent agent 21. The inquiry intelligent agent can determine the symptom summary information and the initial drug taking suggestion information of the user according to the current inquiry information. The health consultation intelligent agent 21 can determine the processing mode according to the current inquiry information and the inquiry content of the previous round, the inquiry content including the symptom summary information and the initial drug taking suggestion information of the user.

[0029] In an optional implementation manner of the embodiment of the present application, the intelligent interaction system 20 can further include a qualification audit intelligent agent (not shown in the figure). The qualification audit intelligent agent can obtain the drug taking suggestion scheme output by the drug recommendation intelligent agent 22, and determine the drug sale information corresponding to the drug taking suggestion scheme, if the drug sale information represents that the user has the purchase qualification, the qualification audit intelligent agent can provide the drug taking suggestion scheme output by the drug recommendation intelligent agent 22 to the information integration intelligent agent 23; if the drug sale information represents that the user does not have the purchase qualification, the qualification audit intelligent agent can feed back to the drug recommendation intelligent agent 22. The drug recommendation intelligent agent 22 can again obtain the drug taking suggestion scheme of the user according to the inquiry content of the previous consultation round and the current inquiry information.

[0030] In an optional implementation of the embodiment of the present application, the intelligent interaction system 20 can further comprise an artificial health consultation intelligent agent (not shown in the figure). The health consultation intelligent agent 21 can select to interact with the artificial health consultation intelligent agent when the processing manner comprises determining a health consultation judgment result and determining a drug suitability result, and the health consultation judgment result indicates that the health consultation is not ended, and the drug suitability result indicates that the user is not suitable for taking the drug. The artificial health consultation intelligent agent can obtain artificial consultation prompt information according to the current inquiry information. Therefore, the server 11 can send the artificial consultation prompt information to the terminal 12.

[0031] In an optional implementation of the embodiment of the present application, if the processing manner comprises determining a health consultation judgment result and determining a drug suitability result, and the health consultation judgment result indicates that the health consultation is not ended, and the drug suitability result indicates that the target user is suitable for taking the drug, the health consultation intelligent agent 21 can determine that the current consultation round is ended.

[0032] In an optional implementation of the embodiment of the present application, if the processing manner is to output the question information, the health consultation intelligent agent 21 can determine the current question information of the current round. Therefore, the server 11 can send the current question information to the terminal 12.

[0033] In an optional implementation of the embodiment of the present application, the intelligent interaction system 20 can create and update the current question information according to the information type of the current question information, by invoking a predetermined tool package.

[0034] Therefore, in the embodiment of the present application, the terminal 12 can display the health consultation feedback information after receiving the health consultation feedback information of the current round.

[0035] In an optional implementation of the embodiment of the present application, the terminal 12 can display the current question information after receiving the current question information of the current round.

[0036] In an optional implementation of the embodiment of the present application, the terminal 12 can display the artificial consultation prompt information after receiving the artificial consultation prompt information.

[0037] In an optional implementation of the embodiment of the present application, the health consultation feedback information can comprise a page jump control. The terminal 12 can display a drug purchase page corresponding to the health consultation feedback information in response to the page jump control being triggered, and the drug purchase page comprises an order submission control. The terminal 12 can further send an order submission request according to the drug purchase information in response to the order submission control being triggered.

[0038] In an optional implementation of the embodiment of the present application, the terminal 12 can send a prescription issuing request according to the symptom summary information of the target user and the drug identification of the prescription drug in response to the fact that the drug selection information includes the prescription drug. The terminal 12 can also send an order submission request according to the drug selection information and the target prescription in response to the fact that the target prescription is received, the target prescription being a prescription corresponding to the prescription drug.

[0039] The following is described by means of a method embodiment. Figure 3 is a flowchart of the multi-agent-based interaction method of the embodiment of the present application. As shown in Figure 3 the method of the embodiment includes the following steps: Step S301, sending current inquiry information.

[0040] In the embodiment, the instant e-commerce platform can provide the terminal with an access mode of the intelligent interaction system, such as a jump control, through a client application program, an applet, etc. Taking the client application program as an example, after the user triggers the jump control of the intelligent interaction system displayed in the client application program, the terminal can jump to the interaction page of the intelligent interaction system and input inquiry information through the interaction page of the intelligent interaction system. Therefore, in this step, the terminal can determine the inquiry information input by the user in the interaction page of the intelligent interaction system as the current inquiry information and send the current inquiry information to the server.

[0041] Step S302, in response to receiving the current inquiry information of the target user in the current consultation round, the health consultation agent determines the processing mode of the current consultation round according to the current inquiry information.

[0042] In the embodiment, the intelligent interaction system can implement the interaction process between at least part of the agents in the intelligent interaction system based on the inquiry information of the target user in each consultation round. The server can pass the current inquiry information to the intelligent interaction system through various existing communication modes, such as local inter-process communication, local Transmission Control Protocol (TCP), Remote Procedure Call (RPC), etc., after receiving the current inquiry information of the target user.

[0043] Therefore, in this step, the health consultation agent can determine the processing mode of the current consultation round according to the current inquiry information after receiving the current inquiry information of the target user in the current consultation round.

[0044] In this embodiment, the health consultation intelligent agent can be implemented based on a large language model. The large language model for implementing the health consultation intelligent agent can be fine-tuned based on a first training sample set. The first training sample set can be determined according to doctor-patient dialogue information. The first training sample set can include health inquiry text input by different users in each round of online health consultation, context information of the health inquiry text, and corresponding health question text and / or health consultation judgment label and / or drug use judgment label. If the health inquiry text corresponds to the health question text in the doctor-patient dialogue information, it can be determined that the processing manner includes outputting the question information, otherwise it is determined that the processing manner does not include outputting the question information. If the doctor makes a drug recommendation for the health inquiry text of the user in the doctor-patient dialogue information, it can be determined that the health consultation judgment label represents the end of the health consultation. If the doctor gives a medical advice of not using drugs for the health inquiry text of the user in the doctor-patient dialogue information, for example, the doctor advises the user to take further examination, it can be determined that the health consultation judgment label represents that the health consultation does not end, and the drug use judgment label represents that the drug use is not appropriate. In the process of fine-tuning the large language model for implementing the health consultation intelligent agent, the input of the large language model for implementing the health consultation intelligent agent can be determined according to each health inquiry text, or each health inquiry text and corresponding context information, and the training target of the large language model for implementing the health consultation intelligent agent can be determined according to the corresponding health question text / or health consultation judgment label and / or drug use judgment label, until the large language model for implementing the health consultation intelligent agent reaches the corresponding training termination condition, for example, the loss function converges, the accuracy reaches a preset threshold, etc. The present embodiment does not limit this.

[0045] The health consultation intelligent agent can determine the prompt of the large language model according to the current inquiry information of the target user in the current consultation round, and determine the processing manner of the current consultation round based on the large language model.

[0046] The processing manner includes at least one of determining a health consultation judgment result, outputting question information, and determining a drug use appropriateness result. The health consultation judgment result is used to represent whether the health consultation ends, and the drug use appropriateness result is used to represent whether the target user is appropriate for drug use.

[0047] Optionally, in order to ensure that the health consultation intelligent agent can fully understand the chief complaint (i.e., the main symptoms, signs and symptom duration of the target user) of the target user, the present embodiment can also score the large language model for implementing the health consultation intelligent agent, and fine-tune the large language model for implementing the health consultation intelligent agent based on the score. The scoring rule corresponding to the health consultation intelligent agent can be set according to actual needs, and the present embodiment does not limit this. In an optional implementation manner, the scoring rule corresponding to the health consultation intelligent agent can refer to the following table: Figure 4 is a flowchart of a multi-agent based interaction method according to an embodiment of the present application. As shown in the figure, in an optional implementation of the embodiment, step S302 can include the following steps: Figure 4 Step S401, in response to receiving the current inquiry information, the intent recognition agent determines the corresponding intent recognition result according to the current inquiry information.

[0048] Before determining the processing mode of the current consultation round according to the current inquiry information, the intent recognition agent of the intelligent interaction system can first determine whether the target user initiates the inquiry for the first time.

[0049] In the embodiment, the intent recognition agent can be implemented based on an intent recognition model, such as a recurrent neural network based on deep learning, a transformer, a long short-term memory artificial neural network, or other types of machine learning models. The model used to implement the intent recognition agent can be trained based on a second training sample set, which can include a plurality of health inquiry texts and corresponding intent labels of each health inquiry text. During the training of the model used to implement the intent recognition agent, the input of the model used to implement the intent recognition agent can be determined according to each health inquiry text, and the training target of the model used to implement the intent recognition agent can be determined according to the corresponding intent label, until the model used to implement the intent recognition agent reaches the corresponding training termination condition.

[0050] Therefore, in this step, the intent recognition agent can determine the input of the corresponding model according to the current inquiry information of the target user in the current consultation round and the inquiry content in the previous consultation round, and determine the intent recognition result of the current inquiry information based on the corresponding model.

[0051] In the embodiment, the target user initiating the inquiry for the first time can mean that the current inquiry information of the target user first shows a clear intent in this conversation, and further can mean that the current inquiry information of the target user first shows a clear health consultation intent in this conversation. For example, the current inquiry information of the target user is "I feel a little uncomfortable", and the intent recognition agent can determine that the current inquiry information has a clear intent, and further can determine whether this clear intent is first shown by the target user in this conversation in combination with the intent recognition result of the inquiry information in the previous consultation round.

[0052] Step S402, determine whether the target user initiates the inquiry for the first time.

[0053] If it is determined that the target user initiates the inquiry for the first time, step S403 can be performed; otherwise, step S404 can be performed. ​

[0054] In step S403, the inquiry agent determines the target user's symptom summary information and initial medication recommendation information based on the current inquiry information.

[0055] If the intent recognition agent determines that the target user is initiating a query for the first time, it can pass the current query information to the query agent. Therefore, in this step, the query agent can combine the current query information with the target user's historical health consultation information, historical medical information, and historical medication purchase records to determine the target user's symptom overview and initial medication recommendations.

[0056] The inquiry agent only needs to perform a preliminary screening of drugs based on the user's symptoms. Therefore, in this embodiment, the initial medication recommendation information for the target user may include information on various types of drugs that meet the target user's symptoms.

[0057] In this embodiment, the inquiry agent can also be implemented based on the large language model. The large language model used to implement the inquiry agent can be obtained by fine-tuning based on the third training sample set. The third training sample set may include multiple health inquiry texts, summary information corresponding to each health inquiry text, and a set of candidate drugs corresponding to the summary information, wherein the candidate drug set can be determined based on the summary information and the medical knowledge graph. In the process of fine-tuning the large language model used to implement the inquiry agent, the input of the large language model used to implement the inquiry agent can be determined based on each health inquiry text, and the training target of the large language model used to implement the inquiry agent can be determined based on the corresponding summary information and the set of candidate drugs, until the large language model used to implement the inquiry agent reaches the corresponding training termination condition.

[0058] Therefore, in this step, the inquiry agent can determine the prompt words of the large language model based on the current inquiry information of the target user in the current consultation round, and determine the symptom overview and initial medication recommendation information of the target user.

[0059] In step S404, the health consultation agent determines a processing method based on the current inquiry information and inquiry content.

[0060] If the intent recognition agent determines that this is not the target user's first inquiry, it can pass the current inquiry information to the health consultation agent. The inquiry agent can also pass the inquiry content from the previous consultation round, including the target user's symptom overview and initial medication recommendations, to the health consultation agent. Therefore, in this step, the health consultation agent can determine the prompt words of the large language model based on the inquiry content of the previous consultation round and the current inquiry information, and then determine the handling method for the current consultation round based on the large language model.

[0061] In step S303, in response to the processing mode including determining the health consultation judgment result, and the health consultation judgment result being ending health consultation, the drug recommendation agent obtains the medication recommendation scheme of the target user according to the inquiry content of the previous consultation round and the current inquiry information.

[0062] In this embodiment, if the health consultation agent determines that the processing mode of the current consultation round includes determining the health consultation judgment result, and the health consultation judgment result is ending health consultation, it means that the target user's physical condition can be clearly inferred according to the information provided by the target user, and there is no need to further consult the target user, and the target user can be given appropriate medication. Therefore, the health consultation agent can pass the inquiry content of the previous round and the current inquiry information to the drug recommendation agent.

[0063] In this step, the drug recommendation agent can obtain the medication recommendation scheme of the target user according to the inquiry content of the previous consultation round and the current inquiry information. The medication recommendation scheme can include the drug name of the recommended drug and the administration method of the recommended drug.

[0064] In determining the recommended drug in the medication recommendation scheme, the drug recommendation agent can make a rationality judgment on each drug in the initial medication recommendation information to ensure the safety of the target user's medication. The rationality judgment can include but is not limited to: special population contraindications, target user's historical allergy information, whether there is a compatibility contraindication between drugs, etc. Among them, the special population contraindication means that some types of drugs are contraindicated for pregnant women, lactating women, children, the elderly, patients with liver and kidney diseases, etc. The target user's historical allergy information means that the target user has an allergic reaction to the drug ingredients. Whether there is a compatibility contraindication between drugs means that the interaction between drugs will affect the efficacy of the drug or cause a toxic reaction.

[0065] In this embodiment, the drug recommendation agent can be implemented based on a large language model. The large language model used to implement the drug recommendation agent can be fine-tuned based on a fourth training sample set. The fourth training sample set can include a plurality of health inquiry texts, context information of each health inquiry text including abstract information and a candidate drug set, and actual drug purchase information of a user. In the process of fine-tuning the large language model used to implement the drug recommendation agent, the input of the large language model used to implement the drug recommendation agent can be determined according to each health inquiry text and the context information of each health inquiry text, and the training target of the large language model used to implement the drug recommendation agent can be determined according to the corresponding actual drug purchase information, until the large language model used to implement the drug recommendation agent reaches the corresponding training termination condition.

[0066] Therefore, in this step, the drug recommendation agent can determine the prompt word of the large language model according to the current inquiry information of the target user in the current consultation round and the inquiry content in the previous consultation round, and determine the drug use recommendation scheme of the target user based on the large language model.

[0067] Optionally, in order to enable the drug recommendation agent to further guarantee the drug safety of the target user, the embodiment can also score the large language model used to implement the drug recommendation agent, and fine-tune the large language model used to implement the drug recommendation agent based on the score. The scoring rule corresponding to the drug recommendation agent can be set according to actual needs, and the embodiment does not limit this. In an optional implementation manner, the scoring rule corresponding to the drug recommendation agent can refer to the following table: In step S304, in response to obtaining the drug use recommendation scheme, the information integration agent obtains health consultation feedback information according to the drug use recommendation scheme.

[0068] After the drug recommendation agent determines the drug use recommendation scheme of the target user, the drug recommendation agent can pass the inquiry content in the previous consultation round and the drug use recommendation scheme to the information integration agent. Therefore, in this step, the information integration agent can obtain the health consultation feedback information of the target user according to the inquiry content in the previous consultation round and the drug use recommendation scheme.

[0069] In this embodiment, the health consultation feedback information can include the drug recommendation explanation corresponding to the drug use recommendation scheme, helping the target user to better understand and accept the drug use recommendation scheme. Optionally, the health consultation feedback information can also include the illness summary information of the target user, helping the user to better understand the physical condition of the user.

[0070] In this embodiment, the information integration agent can be implemented based on a large language model. The large language model used to implement the information integration agent can be obtained by fine-tuning based on a fifth training sample set. The fifth training sample set can include a plurality of health inquiry texts, context information of each health inquiry text including actual drug purchase information of the user, and health consultation summary information corresponding to the health inquiry text. In the process of fine-tuning the large language model used to implement the information integration agent, the input of the large language model used to implement the information integration agent can be determined according to each health inquiry text and the context information of each health inquiry text, and the training target of the large language model used to implement the information integration agent can be determined according to the corresponding health consultation summary information, until the large language model used to implement the information integration agent reaches the corresponding training termination condition.

[0071] Therefore, in this step, the information integration agent can determine the prompt word of the large language model according to the current inquiry information of the target user in the current consultation round and the inquiry content in the previous consultation round, and determine the health consultation feedback information of the target user based on the large language model.

[0072] Optionally, in order to make the health consultation feedback information output by the information integration agent more suitable for the drug use recommendation scheme, the large language model used to implement the information integration agent can also be scored, and the large language model used to implement the information integration agent can be fine-tuned based on the score. The scoring rule corresponding to the information integration agent can be set according to actual needs, and the present embodiment does not limit this. In an optional implementation manner, the scoring rule corresponding to the information integration agent can refer to the following table: Figure 5 is a flowchart of the multi-agent-based interaction method of the embodiment of the present application. As shown in Figure 5 In an optional implementation manner of the present embodiment, step S304 can further include the following steps: Step S501, in response to obtaining the drug use recommendation scheme, the qualification audit agent determines the drug sale information corresponding to the drug use recommendation scheme according to the drug use recommendation scheme.

[0073] In an optional implementation manner of the present embodiment, in order to further improve the rationality of drug recommendation and reduce the difficulty of drug purchase of the target user, after determining the drug use recommendation scheme of the target user, the drug recommendation agent can pass the inquiry content in the previous consultation round and the drug use recommendation scheme to the qualification audit agent. Therefore, in this step, the qualification audit agent can determine the corresponding drug sale information according to the drug use recommendation scheme output by the drug recommendation agent.

[0074] In the present embodiment, the drug sale information is used to represent whether the target user has the purchase qualification of the recommended drug. The purchase qualification can include whether the recommended drug is sold in the area where the target user is located, whether the recommended drug is sold on the instant e-commerce platform, whether the recommended drug is a network prohibited drug, etc.

[0075] Optionally, the qualification audit agent can be implemented based on a pre-set drug list and a matching rule. For example, if the drug list is the list of sellable drugs on the instant e-commerce platform, the matching rule can be: if the recommended drug matches the drug list, it can be determined that the drug sale information represents that the target user has the purchase qualification of the recommended drug; if the drug list is a network prohibited drug, the matching rule can be: if the recommended drug matches the drug list, it can be determined that the drug sale information represents that the target user does not have the purchase qualification of the recommended drug.

[0076] In step S502, in response to the drug sale information indicating that the target user has the purchase qualification, the information integration agent acquires the health consultation feedback information.

[0077] If the qualification auditing agent determines that the drug sale information corresponding to the medication recommendation scheme indicates that the target user has the purchase qualification, the inquiry content and the medication recommendation scheme in the previous consultation round can be passed to the information integration agent. Therefore, in this step, the information integration agent can acquire the health consultation feedback information.

[0078] In step S305, the health consultation feedback information is sent to the terminal.

[0079] After acquiring the health consultation feedback information, the information integration agent can send the health consultation feedback information to the terminal through the server.

[0080] In this embodiment, the information integration agent can create and update the health consultation feedback information according to the information type of the health consultation feedback information by calling a predetermined toolkit, so as to improve the flexibility of information interaction. The information type of the health consultation feedback information can include at least one of streaming, non-streaming, and exception. The predetermined toolkit can be an IMSDK (Instant Messaging Software Development Kit, instant messaging software development kit).

[0081] Optionally, if the information type is streaming, during the streaming response process, when the large language model of the information integration agent returns the streaming output, the stack SDK can trigger the onMessage(content) event processing function according to the streaming output, and pass the content as a callback parameter to the information integration agent. After the information integration agent receives the content, it first calls the createMessage method of the IMSDK to create an unfinished streaming message, and sends the unfinished streaming message to the terminal. Subsequently, the information integration agent can call the updateMessageContent method of the IMSDK to incrementally append the content to the unfinished message body and synchronize to the terminal; after the streaming response ends, the stack SDK will trigger the onAgentFinish event and pass the structured result to the information integration agent. After receiving the onAgentFinish event, the information integration agent can call the updateExt method of the IMSDK to insert the structured result into the extension area of the streaming message through protocol matching rules (such as the mapping of drug recommendation instructions and card templates), and send it to the terminal. If the information type is non-streaming, the stack SDK can return the complete message (result) as a callback parameter to the information integration agent. After receiving the complete message, the information integration agent can call the createMessage method to create a message and send the complete message to the terminal. If the information type is an error, it indicates that the information integration agent fails to reason. At this time, the stack SDK will trigger the onError event and pass the error code to the information integration agent. After receiving the error code, the information integration agent generates user-readable error information through protocol matching, and calls the SendErrorMessage method of the IMSDK to create a message and send it to the terminal.

[0082] Figure 6 is a flowchart of the intelligent interaction system of the embodiment of the present application sending a message to a terminal. As Figure 6As shown, if the message type is streaming and in the streaming response, the stack SDK (not shown) triggers the onMessage(content) event handler and returns the content callback parameter to the intelligent interaction system 61. The intelligent interaction system 61 can first call the createMessage method of IMSDK62 to create an unfinished streaming message, and then call the updateMessageContent method of IMSDK62 to append the content to the unfinished message body and synchronize the message to the terminal 63. If the message type is streaming and the streaming response ends, the stack SDK triggers the onAgentFinish event and passes the structured result to the intelligent interaction system 61. The intelligent interaction system 61 can then call the updateExt method of IMSDK to insert the structured data into the extension area of ​​the streaming message and send it to the terminal. If the message type is non-streaming, after the stack SDK passes the result callback function to the intelligent interaction system 61, the intelligent interaction system 61 can call the createMessage method to create the message and send the complete message to the terminal 63. If the message type is an error, the stack SDK triggers the onError event and passes the error code to the intelligent interaction system 61. The intelligent interaction system 61 can generate user-readable error information through protocol matching, and call the SendErrorMessage method of IMSDK to create a message, and then send the message to the terminal.

[0083] Step S306: in response to receiving the health consultation feedback information of the current inquiry information, display the health consultation feedback information.

[0084] In this step, after the terminal receives the health consultation feedback information of the current inquiry information, the health consultation feedback information can be displayed on the interactive page of the intelligent interactive system to recommend medicines to the target user.

[0085] Figure 7 FIG is a flow chart of a multi-agent interaction method according to an embodiment of the present invention. Figure 7 As shown, in an optional implementation, the method of this embodiment may further include the following steps: Step S701: In response to the drug sales approval information indicating that the target user is not eligible to purchase the drug, the drug recommendation agent obtains a medication recommendation plan again.

[0086] In an optional implementation of this embodiment, if the qualification review agent determines that the drug sales approval information indicates that the target user is not eligible to purchase the drug, the review result can be fed back to the drug recommendation agent. The drug recommendation agent can then obtain a medication recommendation plan for the target user based on the query content of the previous consultation round and the current query information.

[0087] It is easy to understand that in the present embodiment, step S701 can be executed after step S501.

[0088] Figure 8 is a flowchart of the multi-agent based interaction method of the embodiments of the present application. As shown in the figure, in an alternative implementation, the method of the present embodiment can further include the following steps: Figure 8 Step S801, in response to the processing mode including determining the health consultation judgment result and determining the drug suitability result, and the health consultation judgment result representing not ending the health consultation, and the drug suitability result representing that the target user is not suitable for drug use, the artificial health consultation agent obtains the artificial consultation prompt information according to the current inquiry information.

[0089] In an alternative implementation of the present embodiment, if the processing mode output by the health consultation agent includes determining the health consultation judgment result and determining the drug suitability result, and the health consultation judgment result represents not ending the health consultation, and the drug suitability result represents that the target user is not suitable for drug use, it indicates that the condition of the target user is difficult to be relieved by the drug, or the complexity of the condition of the target user is high, therefore the inquiry content of the previous consultation round and the current inquiry information can be passed to the artificial health consultation agent. Therefore in the present step, the artificial health consultation agent can obtain the artificial consultation prompt information according to the inquiry content of the previous consultation round and the current inquiry information.

[0090] In the present embodiment, the artificial health consultation agent can also be implemented based on a large language model. The large language model used to implement the artificial health consultation agent can be fine-tuned based on the sixth training sample set, which can include a plurality of health inquiry texts, context information of each health inquiry text including the summary information of the user and the candidate drug set, and artificial consultation guidance information corresponding to the health inquiry text. In the process of fine-tuning the large language model used to implement the artificial health consultation agent, the input of the large language model used to implement the artificial health consultation agent can be determined according to each health inquiry text and the context information of each health inquiry text, and the training target of the large language model used to implement the artificial health consultation agent can be determined according to the corresponding artificial consultation guidance information, until the large language model used to implement the artificial health consultation agent reaches the corresponding training termination condition.

[0091] Therefore in the present step, the artificial health consultation agent can determine the prompt word of the large language model according to the current inquiry information of the artificial health consultation agent in the current consultation round, and determine the artificial consultation prompt information. Optionally, the artificial consultation prompt information can include a jump link of the artificial consultation interaction page.

[0092] ​It is easy to understand that in the embodiment, the step S801 can be executed after the step S302.

[0093] In step S802, the artificial consultation prompt information is sent to the terminal.

[0094] After obtaining the artificial consultation prompt information, the artificial health consultation agent can send the artificial consultation prompt information to the terminal through the server. In this step, the sending mode of the artificial consultation prompt information can refer to the sending mode of the health consultation feedback information, which will not be described here.

[0095] In step S803, the artificial consultation prompt information is displayed in response to receiving the artificial consultation prompt information.

[0096] In this step, after the terminal receives the artificial consultation prompt information, the artificial consultation prompt information can be displayed in the interactive page of the intelligent interactive system to prompt the user to select the artificial health consultation.

[0097] Figure 9 is a flowchart of the multi-agent-based interaction method of the embodiment of the application. As shown in Figure 9 In an optional implementation, the method of the embodiment can further include the following steps: In step S901, in response to the processing mode being to determine the health consultation judgment result and the medication appropriateness result, and the health consultation judgment result representing that the health consultation is not ended and the medication appropriateness result representing that the target user is appropriate for medication, the health consultation agent determines that the current consultation round is ended.

[0098] In an optional implementation of the embodiment, if the processing mode output by the health consultation agent includes determining the health consultation judgment result and the medication appropriateness result, and the health consultation judgment result represents that the health consultation is not ended and the medication appropriateness result represents that the target user is appropriate for medication, it indicates that the health consultation agent has not fully understood the symptoms of the target user, and therefore the current consultation round can be determined to be ended.

[0099] It is easy to understand that in the embodiment, the step S901 can be executed after the step S302.

[0100] Figure 10 is a flowchart of the multi-agent-based interaction method of the embodiment of the application. As shown in Figure 10 In an optional implementation, the method of the embodiment can further include the following steps: In step S1001, in response to the processing mode being to output the question information, the health consultation agent determines the current question information of the current consultation round.

[0101] In an optional implementation of the embodiment, if the processing manner output by the health consultation agent includes outputting the question information, the health consultation agent can determine the current question information of the current consultation round according to the current inquiry information and the inquiry content of the previous consultation round, so as to understand the condition of the target user through the current question information.

[0102] It is easy to understand that in the embodiment, the step S1001 can be executed after the step S302.

[0103] In step S1002, the current question information is sent to the terminal.

[0104] After obtaining the current question information, the health consultation agent can send the current question information to the terminal through the server. In this step, the sending manner of the current question information can refer to the sending manner of the health consultation feedback information, which will not be described herein.

[0105] In step S1003, the current question information is displayed in response to receiving the current question information of the current inquiry information.

[0106] In this step, after the terminal receives the current question information of the current inquiry information, the terminal can display the current question information in the interactive page of the intelligent interactive system, so as to prompt the user to input the response inquiry information according to the current question information.

[0107] Figures 11-12 is a schematic diagram of the interaction process of the multi-agent of the embodiment of the application. It is easy to understand that, Figure 11 is a schematic diagram of the first half of the interaction process, Figure 12 is a schematic diagram of the second half of the interaction process, that is, Figure 12 is Figure 11 is a continuation of Figures 11-12As shown, after the intelligent interaction system receives the current inquiry information of the target user, the intent recognition intelligent agent 1101 can take the previous inquiry content (i.e. the inquiry content of the previous consultation round) and the current inquiry information as input, and output the intent recognition result of the current inquiry information, i.e. whether the target user initiates the inquiry for the first time (i.e. the first inquiry). If yes, the intent recognition intelligent agent 1101 can pass the previous inquiry content and the current inquiry information to the inquiry intelligent agent 1102; if no, the intent recognition intelligent agent 1101 and the inquiry intelligent agent 1102 can pass the previous inquiry content to the health consultation intelligent agent 1103. The inquiry intelligent agent 1102 can take the previous inquiry content and the current inquiry information as input, and output the symptom summary information of the target user and the initial medication recommendation information, and then pass the current inquiry information of the target user, the symptom summary information and the initial medication recommendation information to the health consultation intelligent agent 1103 as part of the previous inquiry content. The health consultation intelligent agent 1103 takes the previous inquiry content as input, and outputs the question (i.e. the current question information) / whether to end (i.e. whether to end the health consultation) / whether to push the medicine (i.e. whether the target user is suitable for medication). If it is determined to end, the health consultation intelligent agent 1103 passes the current inquiry information of the target user, the symptom summary information and the initial medication recommendation information to the drug recommendation intelligent agent 1104 as part of the previous inquiry content. The drug recommendation intelligent agent 1104 takes the previous inquiry content as input, and outputs the medication recommendation scheme, and then passes the medication recommendation scheme to the qualification audit intelligent agent 1105 as part of the previous inquiry content. The qualification audit intelligent agent 1105 determines the drug sale information according to the medication recommendation scheme in the previous inquiry content. If it is determined that the drug sale information indicates that the target user has the qualification to purchase the medicine, the qualification audit intelligent agent 1105 passes the medication recommendation scheme to the information integration intelligent agent 1106 as part of the previous inquiry content; if it is determined that the drug sale information indicates that the target user does not have the qualification to purchase the medicine, the audit result is fed back to the drug recommendation intelligent agent 1104. The drug recommendation intelligent agent 1104 again takes the previous inquiry content as input, and outputs the medication recommendation scheme. The information integration intelligent agent 1106 can take the previous inquiry content as input, and output the health consultation feedback information, while determining that the current consultation round is ended. If it is determined not to end, and the target user is not suitable for pushing the medicine, the health consultation intelligent agent 1103 passes the current inquiry information of the target user, the symptom summary information and the initial medication recommendation information to the artificial health consultation intelligent agent 1107 as part of the previous inquiry content. The artificial health consultation intelligent agent 1107 takes the previous inquiry content as input, and outputs the artificial consultation prompt information, while determining that the current consultation round is ended. If it is determined not to end, and the target user is suitable for pushing the medicine, the health consultation intelligent agent 1103 can determine that the current consultation round is ended.

[0108] Figure 13is a flowchart of the multi-agent based interaction method of an embodiment of the present application. As shown in Figure 13 In an alternative implementation, the method of the present embodiment can further include the following steps: Step S1301, in response to the page jump control being triggered, display the drug purchase page corresponding to the health consultation feedback information.

[0109] In an alternative implementation of the present embodiment, the health consultation feedback information can further include a page jump control. Therefore, in the present step, if the terminal detects that the page jump control is triggered, for example, the target user clicks the page jump control, the drug purchase page corresponding to the health consultation feedback information can be jumped to and displayed.

[0110] The drug purchase page can include at least one recommended drug in the medication recommendation scheme and an order submission control. The setting of the drug purchase page can be set according to actual needs, and the present embodiment does not limit this.

[0111] It is easy to understand that step S1301 can be executed after step S306.

[0112] Step S1302, in response to the order submission control being triggered, send an order submission request according to the drug purchase information.

[0113] In the present step, if the order submission control is triggered, the terminal can determine the drug purchase information according to the selected recommended drug, and send an order submission request to the server.

[0114] Figure 14 is a flowchart of the multi-agent based interaction method of an embodiment of the present application. As shown in Figure 14 In an alternative implementation, step S1302 can include the following steps: Step S1401, in response to the order submission control being triggered, determine whether the prescription drug is included in the drug purchase information.

[0115] In an alternative implementation of the present embodiment, the medication recommendation scheme can further include a prescription drug, so after the order submission control is triggered, the terminal can determine whether the prescription drug is included in the drug purchase information.

[0116] Step S1402, in response to the prescription drug being included in the drug purchase information, send a prescription issuing request according to the symptom summary information of the target user and the drug identification of the prescription drug.

[0117] In the present step, if the prescription drug is included in the drug purchase information, the terminal can send a prescription issuing request according to the symptom summary information of the target user and the drug identification of the prescription drug, so that the target user can have the purchase qualification of the prescription drug.

[0118] Step S1403, in response to receiving the target prescription, sending an order submission request according to the drug purchase information and the target prescription.

[0119] After the server issues a prescription for a prescription drug (i.e., a target prescription) through an artificial or intelligent agent, the target prescription can be fed back to the terminal. Therefore, after receiving the target prescription, the terminal can send an order submission request according to the drug purchase information and the target prescription.

[0120] In the embodiments of the present application, the intention recognition, health consultation, drug recommendation and other processes can be automatically performed based on the inquiry of the target user in a multi-agent interaction manner, which reduces the time cost of online health consultation of the user and helps to quickly and accurately select and purchase symptomatic drugs and submit a drug purchase order, thereby effectively improving the drug selection and purchase efficiency of the user.

[0121] After the terminal of the embodiment of the present application sends the inquiry information to the server, the health consultation intelligent agent on the server side determines the processing mode of the current consultation round according to the inquiry information. If it is determined that the current processing mode is to end the health consultation, the drug recommendation intelligent agent obtains the user's medication recommendation scheme according to the inquiry content and the inquiry information of the previous consultation round, and then the information integration intelligent agent obtains health consultation feedback information including drug recommendation instructions corresponding to the medication recommendation scheme according to the medication recommendation scheme, and sends the health consultation feedback information to the terminal. After the terminal receives the health consultation feedback information, the health consultation feedback information is displayed. The embodiment of the present application can recommend drugs to the user based on the multi-agent interaction manner, so as to improve the drug selection and purchase efficiency of the user when the user cannot judge the required drug, and improve the drug purchase experience of the user.

[0122] Figure 15 is a flowchart of the interaction method based on multiple agents on the server side of the embodiment of the present application. As shown in Figure 15 The method of the present embodiment on the server side includes the following steps: Step S1501, in response to receiving the current inquiry information of the target user in the current consultation round, the health consultation intelligent agent determines the processing mode of the current consultation round according to the current inquiry information.

[0123] In the present embodiment, the implementation mode of step S1501 is the same as that of step S302, which will not be repeated here.

[0124] Step S1502, in response to the processing mode including determining the health consultation judgment result, and the health consultation judgment result being to end the health consultation, the drug recommendation intelligent agent obtains the medication recommendation scheme of the target user according to the inquiry content and the current inquiry information of the previous consultation round.

[0125] In the present embodiment, the implementation mode of step S1502 is the same as that of step S303, which will not be repeated here.

[0126] In response to obtaining the medication recommendation scheme, the information integration agent obtains health consultation feedback information according to the medication recommendation scheme.

[0127] In this embodiment, the implementation manner of step S1503 is the same as that of step S304, and details are not repeated here.

[0128] Step S1504, sending the health consultation feedback information to the terminal.

[0129] In this embodiment, the implementation manner of step S1504 is the same as that of step S305, and details are not repeated here.

[0130] After the terminal of the embodiment of the present application sends the inquiry information to the server, the health consultation agent on the server side determines the processing mode of the current consultation round according to the inquiry information. If it is determined that the current processing mode is to end the health consultation, the drug recommendation agent obtains the medication recommendation scheme of the user according to the inquiry content and the inquiry information of the previous consultation round. Then, the information integration agent obtains the health consultation feedback information including the drug recommendation instructions corresponding to the medication recommendation scheme according to the medication recommendation scheme, and sends the health consultation feedback information to the terminal. After the terminal receives the health consultation feedback information, the health consultation feedback information is displayed. The embodiment of the present application can recommend drugs to the user based on the multi-agent interaction mode, so as to improve the drug selection and purchase efficiency of the user when the user cannot judge the required drug, and improve the drug purchase experience of the user.

[0131] Figure 16 is a flowchart of the method of the embodiment of the present application on the terminal side. As shown in Figure 16 The method of the embodiment of the present application on the terminal side includes the following steps: Step S1601, sending the current inquiry information.

[0132] In this embodiment, the implementation manner of step S1601 is the same as that of step S301, and details are not repeated here.

[0133] Step S1602, in response to receiving the health consultation feedback information of the current inquiry information, displaying the health consultation feedback information.

[0134] In this embodiment, the implementation manner of step S1602 is the same as that of step S306, and details are not repeated here.

[0135] After the terminal of the embodiment of the present application sends the inquiry information to the server, the health consultation intelligent agent of the server end determines the processing mode of the current consultation round according to the inquiry information. If it is determined that the current processing mode is to end the health consultation, the medicine recommendation intelligent agent obtains the user's medication recommendation scheme according to the inquiry content and the inquiry information of the previous consultation round, and then the information integration intelligent agent obtains the health consultation feedback information including the medicine recommendation explanation corresponding to the medication recommendation scheme according to the medication recommendation scheme, and sends the health consultation feedback information to the terminal. After the terminal receives the health consultation feedback information, the health consultation feedback information is displayed. The embodiment of the present application can recommend medicines to users based on the multi-agent interaction mode, so as to improve the medicine selection efficiency of users when they cannot judge the required medicine, and improve the user's shopping experience.

[0136] Figure 17 is a schematic diagram of the multi-agent based interaction device of the embodiment of the present application, which is suitable for a server. As shown in Figure 17 , the multi-agent based interaction device of the embodiment of the present application includes a processing mode determination unit 1701, a scheme determination unit 1702, a feedback information determination unit 1703 and an information sending unit 1704.

[0137] The processing mode determination unit 1701 is configured to, in response to receiving the current inquiry information of the target user in the current consultation round, determine the processing mode of the current consultation round by the health consultation intelligent agent according to the current inquiry information, the processing mode including at least one of determining a health consultation judgment result, outputting question information and determining a medication suitability result, the health consultation judgment result being used to represent whether to end the health consultation, and the medication suitability result being used to represent whether the target user is suitable for medication; the scheme determination unit 1702 is configured to, in response to the processing mode including determining the health consultation judgment result and the health consultation judgment result being to end the health consultation, obtain the medication recommendation scheme of the target user by the medicine recommendation intelligent agent according to the inquiry content and the current inquiry information of the previous consultation round; the feedback information determination unit 1703 is configured to, in response to obtaining the medication recommendation scheme, obtain the health consultation feedback information according to the medication recommendation scheme by the information integration intelligent agent, the health consultation feedback information including the medicine recommendation explanation corresponding to the medication recommendation scheme; and the information sending unit 1704 is configured to send the health consultation feedback information to the terminal.

[0138] After the terminal of the embodiment of the present application sends the inquiry information to the server, the health consultation intelligent agent of the server end determines the processing mode of the current consultation round according to the inquiry information. If it is determined that the current processing mode is to end the health consultation, the drug recommendation intelligent agent obtains the drug use recommendation scheme of the user according to the inquiry content of the previous consultation round and the inquiry information, and then the information integration intelligent agent obtains the health consultation feedback information including the drug recommendation instructions corresponding to the drug use recommendation scheme according to the drug use recommendation scheme, and sends the health consultation feedback information to the terminal. After the terminal receives the health consultation feedback information, the health consultation feedback information is displayed. The embodiment of the present application can recommend drugs to the user based on the multi-agent interaction mode, so as to improve the drug selection and purchase efficiency of the user when the user cannot judge the required drug, and improve the drug purchase experience of the user.

[0139] Figure 18 is a schematic diagram of the multi-agent based interaction device of the embodiment of the present application, which is suitable for a terminal. As shown in Figure 18 , the multi-agent based interaction device of the embodiment of the present application includes an information sending unit 1801 and an information receiving unit 1802.

[0140] Among them, the information sending unit 1801 is used to send the current inquiry information; the information receiving unit 1802 is used to display the health consultation feedback information in response to receiving the health consultation feedback information of the current inquiry information, the health consultation feedback information is obtained by the information integration intelligent agent according to the drug use recommendation scheme of the target user, the drug use recommendation scheme obtains the health consultation feedback information by the drug recommendation intelligent agent according to the inquiry content of the previous consultation round and the current inquiry information, the drug use recommendation scheme is in response to the processing mode of the current consultation round including determining the health consultation judgment result, and the health consultation judgment result is determined to end the health consultation, the processing mode is determined by the health consultation intelligent agent according to the current inquiry information, the processing mode includes at least one of determining the health consultation judgment result, outputting the question information and determining the drug suitability result, the health consultation judgment result is used to represent whether to end the health consultation, the drug suitability result is used to represent whether the target user is suitable for drug use, and the health consultation feedback information includes the drug recommendation instructions corresponding to the drug use recommendation scheme.

[0141] After the terminal of the embodiment of the present application sends the inquiry information to the server, the health consultation intelligent agent of the server end determines the processing mode of the current consultation round according to the inquiry information, if it is determined that the current processing mode is to end the health consultation, the drug recommendation intelligent agent obtains the user's medication recommendation scheme according to the inquiry content and the inquiry information of the previous consultation round, and then the information integration intelligent agent obtains the health consultation feedback information including the drug recommendation information corresponding to the medication recommendation scheme according to the medication recommendation scheme, and sends the health consultation feedback information to the terminal. After the terminal receives the health consultation feedback information, the health consultation feedback information is displayed. The embodiment of the present application can recommend drugs to users based on the multi-agent interaction mode, so as to improve the drug selection efficiency of users when they cannot judge the required drugs, and improve the drug purchase experience of users.

[0142] Figure 19 is a schematic diagram of an electronic device of the embodiment of the present application. In the present embodiment, the electronic device includes a server, a terminal and the like. As shown in the figure, the electronic device includes at least one processor 1901, and a memory 1902 connected with the at least one processor 1901, and a communication component 1903 connected with the scanning device, the communication component 1903 receives and sends data under the control of the processor 1901; wherein the memory 1902 stores instructions executable by the at least one processor 1901, and the instructions are executed by the at least one processor 1901 to implement the above-mentioned multi-agent based interaction method. Figure 19

[0143] Specifically, the electronic device includes one or more processors 1901 and a memory 1902, Figure 19 In the embodiment, the processor 1901 is taken as an example. The processor 1901 and the memory 1902 can be connected through a bus or other ways, Figure 19 In the embodiment, the connection through the bus is taken as an example. The memory 1902 is a kind of non-volatile computer readable storage medium, which can be used to store non-volatile software programs, non-volatile computer executable programs and modules. The processor 1901 executes the various functions of the device and data processing by running the non-volatile software programs, instructions and modules stored in the memory 1902, that is, implements the above-mentioned multi-agent based interaction method.

[0144] ​The memory 1902 can include a program storage area and a data storage area, where the program storage area can store an operating system, at least one application required by a function, and the data storage area can store an option list, etc. In addition, the memory 1902 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state memory device. In some embodiments, the memory 1902 can optionally include a memory disposed remotely with respect to the processor 1901, which can be connected to an external device through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0145] One or more modules are stored in the memory 1902, and when executed by the one or more processors 1901, perform the multi-agent based interaction method in any of the above method embodiments.

[0146] The above product can perform the method provided by the embodiments of the present application, has the corresponding function modules and beneficial effects of performing the method, and the technical details not described in detail in the embodiments can be referred to the method provided by the embodiments of the present application.

[0147] After the terminal of the embodiment of the present application sends the inquiry information to the server, the health consultation agent of the server end determines the processing mode of the current consultation round according to the inquiry information, if it is determined that the current processing mode is to end the health consultation, the drug recommendation agent obtains the user's medication recommendation scheme according to the inquiry content and the inquiry information of the previous consultation round, and then the information integration agent obtains the health consultation feedback information including the drug recommendation information corresponding to the medication recommendation scheme according to the medication recommendation scheme, and sends the health consultation feedback information to the terminal. After the terminal receives the health consultation feedback information, the health consultation feedback information is displayed. The embodiment of the present application can recommend drugs to the user based on the multi-agent interaction mode, so as to improve the drug selection efficiency of the user when the user cannot judge the required drug, and improve the drug purchase experience of the user.

[0148] Another embodiment of the present application relates to a non-volatile storage medium for storing a computer readable program for a computer to execute part or all of the above method embodiments.

[0149] That is, a person skilled in the art can understand that all or part of the steps in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a program stored in a storage medium, including a plurality of instructions for causing a device (which can be a single-chip microcomputer, a chip, etc.) or a processor to execute all or part of the steps of the methods described in the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0150] The above merely describes the preferred embodiments of the present application and is not intended to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A multi-agent based interaction method, applicable to a server, characterized in that: The method comprises: In response to receiving the current inquiry information of the target user in the current consultation round, the health consultation agent determines a processing method for the current consultation round based on the current inquiry information, the processing method including at least one of determining a health consultation judgment result, outputting question information, and determining a medication suitability result, the health consultation judgment result being used to indicate whether the health consultation is terminated, and the medication suitability result being used to indicate whether the target user is suitable for medication; In response to the processing method including determining a health consultation judgment result, and the health consultation judgment result is to end the health consultation, the drug recommendation agent obtains a medication recommendation plan for the target user based on the query content of the previous consultation round and the current query information; In response to obtaining the medication suggestion plan, the information integration agent obtains health consultation feedback information according to the medication suggestion plan, the health consultation feedback information including a drug recommendation description corresponding to the medication suggestion plan; Send the health consultation feedback information to the terminal.

2. The method according to claim 1, characterized in that In response to receiving the current inquiry information of the target user in the current consultation round, the health consultation agent determines the processing method of the current consultation round according to the current inquiry information, including: In response to receiving the current query information, the intention recognition agent determines a corresponding intention recognition result based on the current query information, where the intention recognition result is used to indicate whether the target user initiates the query for the first time; In response to the intention recognition result indicating that the target user has initiated a query for the first time, the query agent determines the target user's symptom summary information and initial medication recommendation information based on the current query information; In response to the target user not initiating the inquiry for the first time, the health consultation agent determines the processing method based on the current inquiry information and the inquiry content, where the inquiry content includes the symptom summary information and the initial medication recommendation information.

3. The method according to claim 1, characterized in that In response to obtaining the medication recommendation plan, the information integration agent obtains health consultation feedback information according to the medication recommendation plan, including: In response to obtaining the medication recommendation plan, the qualification review agent determines, based on the medication recommendation plan, the drug sales approval information corresponding to the medication recommendation plan, wherein the drug sales approval information is used to indicate whether the target user is eligible to purchase the recommended drug, and the recommended drug is a drug in the medication recommendation plan; In response to the drug sales approval information indicating that the target user is eligible to purchase the drug, the information integration agent obtains the health consultation feedback information.

4. The method according to claim 3, characterized in that The method further comprises: In response to the drug sales approval information indicating that the target user is not eligible to purchase the drug, the drug recommendation agent obtains the medication recommendation plan again.

5. The method according to claim 1, wherein The method further comprises: In response to the processing method including determining a health consultation judgment result and determining a medication suitability result, and the health consultation judgment result indicates that the health consultation is not terminated, and the medication suitability result indicates that the target user is not suitable for medication, the artificial health consultation agent obtains artificial consultation prompt information according to the current query information; Sending the manual consultation prompt information to the terminal.

6. The method according to claim 1, characterized in that The method further comprises: In response to the processing method of determining the health consultation judgment result and determining the medication suitability result, and the health consultation judgment result indicates that the health consultation is not terminated, and the medication suitability result indicates that the target user is suitable for medication, the health consultation agent determines that the current consultation round is ended.

7. The method according to claim 1, characterized in that The method further comprises: In response to the processing mode being to output question information, the health consultation agent determines current question information of the current consultation round; The current question information is sent to the terminal.

8. The method according to claim 7, characterized in that The sending the current question information to the terminal includes: According to the information type of the current question information, a predetermined toolkit is called to create and update the current question information, where the information type includes at least one of streaming, non-streaming, and exception.

9. A multi-agent based interaction method, applicable to a terminal, characterized in that: The method comprises: Send current inquiry information; In response to receiving the health consultation feedback information of the current inquiry information, the health consultation feedback information is displayed. The health consultation feedback information is obtained by the information integration intelligent body according to the medication recommendation plan of the target user. The medication recommendation plan obtains the health consultation feedback information by the drug recommendation intelligent body according to the inquiry content of the previous consultation round and the current inquiry information. The processing method of the medication recommendation plan in response to the current consultation round includes determining the health consultation judgment result, and the health consultation judgment result is determined to terminate the health consultation. The processing method is determined by the health consultation intelligent body according to the current inquiry information. The processing method includes determining the health consultation judgment result, outputting the question information and determining at least one of the medication suitability result. The health consultation judgment result is used to indicate whether the health consultation is terminated. The medication suitability result is used to indicate whether the target user is suitable for medication. The health consultation feedback information includes the drug recommendation instructions corresponding to the medication recommendation plan.

10. The method according to claim 9, characterized in that The method further comprises: In response to receiving the current question information of the current inquiry information, the current question information is displayed, and the current question information is determined by the health consultation agent in response to the processing method of outputting the question information.

11. The method according to claim 9, characterized in that The method further comprises: In response to receiving the manual consultation prompt information, the manual consultation prompt information is displayed. The manual consultation prompt information is responded to by the artificial health consultation intelligent body in accordance with the processing method, including determining the health consultation judgment result and determining the medication suitability result, and the health consultation judgment result indicates that the health consultation is not terminated, and the medication suitability result indicates that the target user is not suitable for medication determination.

12. The method according to claim 9, characterized in that The health consultation feedback information includes a page jump control; The method further comprises: In response to the page jump control being triggered, displaying a medicine purchase page corresponding to the health consultation feedback information, the medicine purchase page including an order submission control; In response to the order submission control being triggered, an order submission request is sent according to the medicine purchasing information.

13. The method according to claim 12, characterized in that In response to the order submission control being triggered, sending the order submission request according to the drug purchase information includes: In response to the order submission control being triggered, determining whether the drug purchasing information includes prescription drugs; In response to the drug purchasing information including prescription drugs, sending a prescription request based on the symptom summary information of the target user and the drug identifier of the prescription drug; In response to receiving the target prescription, the order submission request is sent according to the drug purchasing information and the target prescription, where the target prescription is a prescription corresponding to the prescription drug.

14. A multi-agent based interaction method, characterized in that: The method comprises: Send current inquiry information; In response to receiving the current inquiry information of the target user in the current consultation round, the health consultation agent determines a processing method for the current consultation round based on the current inquiry information, the processing method including at least one of determining a health consultation judgment result, outputting question information, and determining a medication suitability result, the health consultation judgment result being used to indicate whether the health consultation is terminated, and the medication suitability result being used to indicate whether the target user is suitable for medication; In response to the processing method including determining a health consultation judgment result, and the health consultation judgment result is to end the health consultation, the drug recommendation agent obtains a medication recommendation plan for the target user based on the query content of the previous consultation round and the current query information; In response to obtaining the medication suggestion plan, the information integration agent obtains health consultation feedback information according to the medication suggestion plan, the health consultation feedback information including a drug recommendation description corresponding to the medication suggestion plan; Sending the health consultation feedback information to the terminal; In response to receiving the health consultation feedback information of the current inquiry information, the health consultation feedback information is displayed.

15. A multi-agent based interactive system, characterized in that: include: The server is configured to, in response to receiving the current inquiry information of the target user in the current consultation round, the health consultation agent determines the processing method of the current consultation round according to the current inquiry information, in response to the processing method including determining the health consultation judgment result, and the health consultation judgment result is to end the health consultation, the drug recommendation agent obtains the medication recommendation plan of the target user according to the inquiry content of the previous consultation round and the current inquiry information, in response to obtaining the medication recommendation plan, the information integration agent obtains health consultation feedback information according to the medication recommendation plan, and sends the health consultation feedback information to the terminal, the processing method includes determining at least one of the health consultation judgment result, outputting question information and determining the medication suitability result, the health consultation judgment result is used to indicate whether the health consultation is ended, the medication suitability result is used to indicate whether the target user is suitable for medication, and the health consultation feedback information includes the drug recommendation instructions corresponding to the medication recommendation plan; as well as The terminal is configured to send current inquiry information and, in response to receiving health consultation feedback information of the current inquiry information, display the health consultation feedback information.

16. A multi-agent based interactive device, suitable for a server, characterized in that: The device comprises: a processing method determining unit, configured to, in response to receiving current inquiry information of a target user in a current consultation round, determine, by the health consultation agent, a processing method for the current consultation round based on the current inquiry information, the processing method including at least one of determining a health consultation judgment result, outputting question information, and determining a medication suitability result, the health consultation judgment result being used to indicate whether the health consultation is terminated, and the medication suitability result being used to indicate whether the target user is suitable for medication; a plan determination unit, configured to, in response to the processing method including determining a health consultation judgment result, and the health consultation judgment result being to end the health consultation, obtain a medication recommendation plan for the target user based on the query content of the previous consultation round and the current query information; a feedback information determining unit, configured to, in response to obtaining the medication recommendation plan, obtain health consultation feedback information based on the medication recommendation plan, wherein the health consultation feedback information includes a drug recommendation description corresponding to the medication recommendation plan; The information sending unit is used to send the health consultation feedback information to the terminal.

17. A multi-agent based interactive device, suitable for a terminal, characterized in that: The device comprises: An information sending unit, used for sending current inquiry information; An information receiving unit is used to display the health consultation feedback information in response to the health consultation feedback information received from the current inquiry information. The health consultation feedback information is obtained by the information integration intelligent body according to the medication recommendation plan of the target user. The health consultation feedback information obtained by the medication recommendation plan is obtained by the drug recommendation intelligent body according to the inquiry content of the previous consultation round and the current inquiry information. The processing method of the medication recommendation plan in response to the current consultation round includes determining the health consultation judgment result, and the health consultation judgment result is determined to terminate the health consultation. The processing method is determined by the health consultation intelligent body according to the current inquiry information. The processing method includes determining the health consultation judgment result, outputting the question information and determining at least one of the medication suitability result. The health consultation judgment result is used to indicate whether the health consultation is terminated. The medication suitability result is used to indicate whether the target user is suitable for medication. The health consultation feedback information includes the drug recommendation instructions corresponding to the medication recommendation plan.

18. An electronic device comprising a memory and a processor, characterized in that: The memory is configured to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method according to any one of claims 1 to 14.

19. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 14 is implemented.

20. A computer program product, characterized in that The computer program product comprises a computer program / instructions, which implement the method according to any one of claims 1 to 14 when executed by a processor.

Images