Business scenario dialogue methods and related products

By using a thinking chain based on a large language model for dialogue processing in business scenarios such as business recommendation and medical follow-up, the problems of high labor costs and low efficiency in traditional methods are solved, and efficient and accurate business scenario dialogue is achieved.

CN117909463BActive Publication Date: 2024-12-03ZHUHAI HENGQIN SANMED AITECH INC
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202311765522.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-12-20
Publication Date
2024-12-03
Estimated Expiration
2043-12-20

AI Technical Summary

Technical Problem

In business scenarios such as business recommendation and medical follow-up, traditional methods require a lot of manual participation, resulting in high labor costs and low efficiency, and inaccurate follow-up data, which is not accurate and comprehensive enough, and is easily affected by human factors.

Method used

By utilizing multiple thinking chains based on large language models, first use the business needs to understand the thinking chain to determine whether the target user has business needs. If the business needs are confirmed, then use the business needs to understand the thinking chain to initiate a business dialogue request to the user equipment and generate a Business demand guidance information, and then use the business question and reply thinking chain to complete the dialogue process, so as to actively guide users to participate in business scenario dialogue and reduce labor costs.

Benefits of technology

It has achieved no manual intervention throughout the entire process of business scenario dialogue, reduced labor costs, improved dialogue efficiency and data accuracy, and ensured the efficiency and accuracy of business scenario dialogue.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117909463B_ABST
    Figure CN117909463B_ABST
Patent Text Reader

Abstract

The present disclosure provides a business scenario dialogue method and related products. A specific implementation of the method uses multiple thinking chains based on a large language model, first uses a business demand understanding thinking chain, performs semantic understanding based on the user information of the target user and business scenario related information, and determines whether the target user has business needs; if it is confirmed that there is a business need, then uses the business demand understanding thinking chain to initiate a business dialogue request to the user device corresponding to the target user, and generates business demand guidance information, and sends the business demand guidance information as the first business dialogue information of the business dialogue to the user device. After that, the dialogue process with the user device is completed through the business question thinking chain and the business answer thinking chain respectively. It realizes the active guidance of users to participate in business scenario dialogues, and uses multiple thinking chains based on large language models to conduct dialogue processes, reducing the manpower cost of business scenario dialogues.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The embodiments of the present disclosure relate to the field of natural language processing technology, and in particular to business scenario dialogue methods and related products. Background Art

[0002] With the development of Large Language Model (LLM), various conversational applications based on LLM have emerged.

[0003] However, most of the current dialogue applications involve users actively asking questions to large language model applications, and large language model applications passively answering questions. The above dialogue method is not applicable to many business scenario dialogues. Summary of the invention

[0004] The embodiments of the present disclosure provide a business scenario dialogue method, device, electronic device, storage medium, and computer program product.

[0005] In a first aspect, an embodiment of the present disclosure provides a business scenario dialogue method, the method comprising:

[0006] Obtain user information and business scenario-related information of the target user;

[0007] Using the business demand understanding thinking chain, semantic understanding is performed based on the user information of the target user and relevant information of the business scenario to obtain the business demand information of the target user, where the business demand information includes demand confirmation information indicating whether the business demand is met;

[0008] In response to the target user's demand confirmation information indicating that there is a business demand, initiating a business session request to a user device corresponding to the target user by using the business demand understanding thinking chain;

[0009] In response to the successful establishment of the business dialogue, using a business question thinking chain, generating business demand guidance information based on the user information of the target user, and sending the business demand guidance information as the first business dialogue information of the business dialogue to the user device;

[0010] For the business dialogue, the business question thinking chain and the business answer thinking chain are used to complete the dialogue process with the user device, wherein the business answer thinking chain is used to perform the following reply information generation operation: receiving and recording the current user dialogue information fed back by the user device in response to the business dialogue, based on the historical dialogue record of the business dialogue, searching the current user dialogue information in the search engine and the business domain knowledge base respectively to obtain the first search result and the second search result, generating business dialogue intermediate information based on the first search result and the second search result, and sending the business dialogue intermediate information and the current user dialogue information to the business question thinking chain; the business question thinking chain is used to generate business dialogue information based on the user information and business scenario related information of the target user, the historical dialogue record of the business dialogue, the business dialogue intermediate information and the current user dialogue information, and send the business dialogue information to the user device.

[0011] In some optional implementations, the business demand understanding thinking chain, the business question thinking chain, and the business answer thinking chain are artificial intelligence agents generated based on a large language model and a preset business question and answer process template.

[0012] In some optional implementations, the business requirement understanding thinking chain and the business question thinking chain are the same thinking chain.

[0013] In some optional implementations, the method further includes: in response to the business dialogue ending, generating business processing information based on a dialogue record corresponding to the business dialogue.

[0014] In some optional implementations, the method further includes: determining a service processing terminal device according to the service processing information; and sending the service processing information to the service processing terminal device.

[0015] In some optional embodiments, the business scenario is a patient follow-up scenario, and the business scenario related information includes at least one of the following: clinical examination results, clinical diagnosis results, and follow-up related information, and the business demand is a follow-up demand.

[0016] In some optional implementations, the preset business question and answer process template is a preset follow-up question and answer process template.

[0017] In a second aspect, an embodiment of the present disclosure provides a business scenario dialogue device, the device comprising:

[0018] An acquisition module, configured to acquire user information of a target user and information related to a business scenario;

[0019] A demand understanding module is configured to use a business demand understanding thinking chain to perform semantic understanding based on the user information of the target user and relevant information of the business scenario to obtain the business demand information of the target user, wherein the business demand information includes demand confirmation information indicating whether the business demand is met;

[0020] A dialog request module is configured to initiate a business dialog request to a user device corresponding to the target user by using the business demand understanding thinking chain in response to the target user's demand confirmation information indicating that the target user has a business demand;

[0021] a dialogue guiding module configured to generate business demand guiding information based on the user information of the target user by using a business question thinking chain in response to the business dialogue being successfully established, and to send the business demand guiding information as the first business dialogue information of the business dialogue to the user device;

[0022] The business dialogue module is configured to complete the dialogue process between the user device and the business dialogue using the business question thinking chain and the business answer thinking chain, wherein the business answer thinking chain is used to perform the following reply information generation operations: receiving and recording the current user dialogue information fed back by the user device in response to the business dialogue, based on the historical dialogue record of the business dialogue, searching the current user dialogue information in a search engine and a business domain knowledge base respectively to obtain a first search result and a second search result, generating business dialogue intermediate information based on the first search result and the second search result, and sending the business dialogue intermediate information and the current user dialogue information to the business question thinking chain; the business question thinking chain is used to generate business dialogue information based on the user information and business scenario related information of the target user, the historical dialogue record of the business dialogue, the business dialogue intermediate information and the current user dialogue information, and send the business dialogue information to the user device.

[0023] In some optional implementations, the business demand understanding thinking chain, the business question thinking chain, and the business answer thinking chain are artificial intelligence agents generated based on a large language model and a preset business question and answer process template.

[0024] In some optional implementations, the business requirement understanding thinking chain and the business question thinking chain are the same thinking chain.

[0025] In some optional implementations, the device further includes: an information generation module configured to generate business processing information based on a conversation record corresponding to the business conversation in response to the end of the business conversation.

[0026] In some optional implementations, the apparatus further includes: a device determination module configured to determine a service processing terminal device based on the service processing information; and an information sending module configured to send the service processing information to the service processing terminal device.

[0027] In some optional embodiments, the business scenario is a patient follow-up scenario, and the business scenario related information includes at least one of the following: clinical examination results, clinical diagnosis results, and follow-up related information, and the business demand is a follow-up demand.

[0028] In some optional implementations, the preset business question and answer process template is a preset follow-up question and answer process template.

[0029] In a third aspect, an embodiment of the present disclosure provides an electronic device, comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner in the first aspect.

[0030] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by one or more processors, implements the method described in any implementation manner in the first aspect.

[0031] In a fifth aspect, an embodiment of the present disclosure provides a computer program product, including a computer program / instruction, which implements the method described in any implementation manner in the first aspect when the computer program / instruction is executed by a processor.

[0032] In the actual production, life and work environment, there are some business scenarios that require the business party to actively initiate a dialogue with the user. For example, in some business recommendation scenarios or medical follow-up scenarios, the business party or hospital as the business party needs to hire relevant personnel to actively communicate with the user, such as through phone calls, text messages, instant messaging services (Instant Messaging Service, IMS), faxes, etc. However, a large amount of manual participation is required, which leads to increased labor costs.

[0033] For example, patient follow-up is an indispensable part of the treatment and rehabilitation process. Through follow-up, doctors can understand the patient's condition and treatment status in a timely manner, and adjust the treatment plan in a timely manner, thereby improving the treatment effect and patient satisfaction. However, traditional follow-up methods often require doctors to manually make phone calls or communicate face-to-face for manual consultations. The follow-up efficiency is low, and the follow-up method is easily affected by human factors, making the follow-up data inaccurate and incomplete, and it is easy to have problems with irregular follow-up data and difficult analysis, making it difficult to provide doctors with accurate patient health information.

[0034] In order to solve the problem of high labor cost and low efficiency in business scenarios where the business party actively initiates the dialogue, the business scenario dialogue method, device, electronic device, storage medium and computer program product provided by the embodiments of the present disclosure, by using multiple thinking chains based on large language models, first use the business demand understanding thinking chain, based on the user information of the target user and the business scenario related information to perform semantic understanding, determine whether the target user has business needs; if it is confirmed that there is a business need, then use the business demand understanding thinking chain to initiate a business dialogue request to the user device corresponding to the target user, and generate business demand guidance information, and send the business demand guidance information as the first business dialogue information of the business dialogue to the user device. After that, the business question thinking chain and the business answer thinking chain are used to complete the dialogue process with the user device. Thereby, it is realized to actively guide users to participate in the business scenario dialogue, and use multiple thinking chains based on large language models to conduct the dialogue process, without manual intervention throughout the process, reducing the labor cost of the business scenario dialogue. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Other features, objects and advantages of the present disclosure will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings. The drawings are only for the purpose of illustrating specific embodiments and are not to be considered as limiting the present invention. In the drawings:

[0036] Figure 1 is an exemplary system architecture diagram in which an embodiment of the present disclosure may be applied;

[0037] Figure 2 is a flow chart of an embodiment of a business scenario dialogue method according to the present disclosure;

[0038] Figure 3 is a structural diagram of an embodiment of a business scenario dialogue device according to the present disclosure;

[0039] Figure 4 It is a structural diagram of a computer system of an electronic device suitable for implementing an embodiment of the present disclosure. DETAILED DESCRIPTION

[0040] The present disclosure is further described in detail below in conjunction with the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings.

[0041] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present disclosure may be combined with each other. The present disclosure will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0042] Figure 1 An exemplary system architecture 100 is shown to which embodiments of the business conversation method, apparatus, electronic device, and storage medium of the present disclosure can be applied.

[0043] like Figure 1 As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0044] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as natural language processing applications, voice recognition applications, short video social applications, audio and video conferencing applications, web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0045] Terminal devices 101, 102, 103 can be hardware or software. When terminal devices 101, 102, 103 are hardware, they can be various electronic devices with information input devices (e.g., keyboard, mouse, touch screen, microphone, camera, etc.) and information output devices (e.g., display screen, speaker, etc.), including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III, Moving Picture Experts Compression Standard Audio Layer 3), MP4 (Moving Picture Experts Group Audio Layer IV, Moving Picture Experts Compression Standard Audio Layer 4) players, laptop computers and desktop computers, etc. When terminal devices 101, 102, 103 are software, they can be installed in the terminal devices listed above. It can be implemented as multiple software or software modules (for example, used to provide business scenario dialogue services), or it can be implemented as a single software or software module. No specific limitation is made here.

[0046] In some cases, the service conversation method provided by the present disclosure may be executed by the terminal devices 101, 102, 103, and accordingly, the service conversation apparatus may be disposed in the terminal devices 101, 102, 103. In this case, the system architecture 100 may not include the server 105.

[0047] In some cases, the business dialogue method provided by the present disclosure can be jointly performed by the terminal devices 101, 102, 103 and the server 105. For example, the step of "obtaining user information of the target user and business scenario related information" can be performed by the terminal devices 101, 102, 103, and the steps of "using the business demand understanding thinking chain, performing semantic understanding based on the user information of the target user and the business scenario related information, and obtaining the business demand information of the target user" can be performed by the server 105. The present disclosure does not limit this. Accordingly, the business dialogue device can also be respectively set in the terminal devices 101, 102, 103 and the server 105.

[0048] In some cases, the business dialogue method provided by the present disclosure may be executed by the server 105 , and accordingly, the business dialogue device may also be set in the server 105 . In this case, the system architecture 100 may not include the terminal devices 101 , 102 , and 103 .

[0049] It should be noted that the server 105 can be hardware or software. When the server 105 is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or it can be implemented as a single server. When the server 105 is software, it can be implemented as multiple software or software modules (for example, for providing distributed services), or it can be implemented as a single software or software module. No specific limitation is made here.

[0050] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to the implementation requirements.

[0051] Continue to refer Figure 2 , which shows a process 200 of an embodiment of a business scenario dialogue method according to the present disclosure, the business scenario dialogue method comprises the following steps:

[0052] Step 201: Obtain user information and business scenario related information of a target user.

[0053] In this embodiment, the execution subject of the business dialogue method (for example Figure 1 The server shown in the figure) can obtain the user information and business scenario-related information of the target user locally or remotely from other electronic devices connected to the execution subject network.

[0054] It is understandable that the specific content of the user information of the target user and the business scenario related information to be obtained can be set according to the needs of the business scenario.

[0055] For example, for business recommendation scenarios such as product or service recommendations, user information may include user name, contact information, gender, age, occupation, income status, etc., and business scenario related information may include the user's historical purchase or consumption records of the product or service category that the user wishes to recommend in the corresponding business scenario, user consultation records, user consumption tendency profile, etc. Accordingly, the business demand in the business recommendation scenario is the business recommendation demand.

[0056] For another example, when the business scenario is a patient follow-up scenario, user information may include user name, contact information, gender, age, eating habits, body fat percentage, blood sugar index, blood pressure index, lifestyle and changes in condition, etc., and business scenario related information may include at least one of the following: clinical examination results, clinical diagnosis results and follow-up related information. Accordingly, the business demand in the patient follow-up scenario is a follow-up demand. Here, the patient follow-up scenario means that medical institutions usually follow up with patients after the visit to investigate the patient's satisfaction with the visit, the treatment process, or the health status after the visit.

[0057] It should be noted that the user information of the target user and the business scenario related information may be in various data formats, which is not specifically limited in the present disclosure, and may include but is not limited to text, images, sounds, videos, etc.

[0058] Step 202 , using the business demand understanding thinking chain, performs semantic understanding based on the user information of the target user and the business scenario related information to obtain the business demand information of the target user.

[0059] Here, the chain-of-thought (CoT) for understanding business requirements refers to the chain of thought obtained by decomposing and symbolizing the business requirements understanding process of the corresponding business scenarios based on the business domain knowledge base.

[0060] It should be noted that here, the business demand understanding thinking chain can be directly used to semantically understand the user information of the target user and the business scenario related information based on the large language model to obtain the business demand information of the target user. The large language model refers to a neural network model trained by a large amount of text data, which is used to process and generate natural language text. These models usually use deep learning methods, such as recurrent neural networks (RNN) or transformers to model the statistical laws and semantic relationships of language. The business demand understanding agent can also be used to semantically understand the user information of the target user and the business scenario related information to obtain the business demand information of the target user. Among them, the business demand understanding agent is an artificial intelligence agent (Artificial Intelligence Agent) pre-built based on the business demand understanding thinking chain and the large language model. The agent can include three elements: memory, tool, and profile. The agent can call some tools in a targeted manner based on these three elements to complete the tasks that need to be delivered to the agent.

[0061] In order to adapt to specific business scenarios, the large language model used in the embodiments of the present disclosure may also be a model that has undergone prompt tuning. Prompt Tuning refers to the process of adjusting and optimizing the understanding and processing of questions or instructions input by users in a dialogue system. Prompt Tuning fine-tuning of a large language model refers to further optimizing an already trained large language model to adapt to specific tasks or application processes. In Prompt Tuning, a set of appropriate prompts or samples are usually selected as input data, and the model is fine-tuned by training on this data. These prompts can be questions, instructions or examples related to specific business scenarios, which are used to guide the reasoning and generation of the model in specific business scenarios.

[0062] The business demand understanding agent can be an agent with intent understanding capabilities after the business demand understanding thinking chain is based on prompts and fine-tuning of the large language model. Specifically, some sample user information and business scenario-related information that have been manually annotated or intelligently processed can be used as input data, and the corresponding business demand annotated information can be used as output data labels. Based on the business demand understanding thinking chain, the large language model can be trained to fine-tune the business demand understanding agent.

[0063] Here, the business requirement information includes requirement confirmation information for indicating whether the business requirement is met.

[0064] In some optional implementations, the business demand information may also include specific information about the business demand, such as time, location, business type, and other related information related to the business demand.

[0065] It should be noted that the business demand information may also be in various data formats, which is not specifically limited in the present disclosure, and may include but is not limited to text, images, sounds, videos, etc.

[0066] Step 203 : In response to the target user's requirement confirmation information indicating that the target user has a service requirement, a service session request is initiated to a user device corresponding to the target user by using a service requirement understanding thinking chain.

[0067] Here, the user equipment of the target user may be various electronic devices, such as a fixed-line phone, a mobile phone, an intelligent electronic device, etc.

[0068] Specifically, a service conversation request may be initiated to the user device corresponding to the target user in various ways, such as calling the target user's mobile phone number, sending an email to the target user's email address, sending an instant messaging message to the target user's bound instant messaging application account, or sending a message to the target user's bound preset service scenario application account, etc.

[0069] It should be noted that, here, the business demand understanding thinking chain can be directly used, or the above-mentioned business demand understanding agent can be used to initiate a business dialogue request to the user device corresponding to the target user.

[0070] Step 204, in response to the successful establishment of the business dialogue, using the business question thinking chain, generating business demand guidance information based on the user information of the target user, and sending the business demand guidance information as the first business dialogue information of the business dialogue to the user device.

[0071] Here, the business question thinking chain refers to the thinking chain obtained by breaking down and symbolizing the business demand guidance process of the corresponding business scenario based on the business domain knowledge base.

[0072] It should be noted that here, the business question thinking chain can be directly used based on the large language model, or the business demand understanding agent can be used to generate business demand guidance information based on the user information of the target user, and the business demand guidance information is sent to the user device as the first business dialogue information of the business dialogue. Among them, the business questioning agent is an artificial intelligence agent pre-built based on the business question thinking chain and the large language model.

[0073] Optionally, the business questioning agent can also be an agent that has the function of generating language according to the context after the business questioning thinking chain is based on the prompt fine-tuning of the large language model. For example, some sample user information and business scenario related information that have been manually annotated or intelligently processed can be used as input data, and the corresponding annotated business demand guidance information can be used as the output result label. Based on the business questioning thinking chain, the large language model can be trained to fine-tune the business questioning agent.

[0074] Here, the business demand guidance information is used to guide the user into a business dialogue. In traditional artificial intelligence dialogue applications, users often actively ask questions or demands to the artificial intelligence dialogue application to initiate a dialogue, and the artificial intelligence dialogue application responds.

[0075] Different from the conventional AI dialogue application scenarios, for scenarios where the business party needs to actively initiate a dialogue, the user will not actively initiate a dialogue. Therefore, here, the business question thinking chain or business question intelligent agent is used to generate business demand guidance information based on the user information of the target user, and the business demand guidance information is sent to the user device as the first business dialogue information of the business dialogue. In this way, the target user can obtain the business demand guidance information in the business dialogue through the user device, and participate in the business dialogue under the guidance of the business demand guidance information.

[0076] Taking the product or service recommendation scenario as an example, the business demand guidance information can be "Hello Mr. XX, we have noticed that your mobile network traffic has exceeded the limit in your mobile phone bills in recent months. In order to save you money, we recommend that you change your current package, which includes more mobile network traffic. Do you want to know more?" After seeing the above business demand guidance information, the target user can participate in the conversation. For example, the target user can enter "need", "please tell me the specific package situation", "no need", "thank you, no need", etc. on the user device through text, emoticons, or voice to conduct business conversations.

[0077] In some optional implementations, various implementations may be used to pre-formulate a preset business process question and answer template corresponding to a business scenario based on the needs of a specific business scenario. Accordingly, the business question thinking chain may be a thinking chain generated according to the business demand guidance part in the preset business process question and answer template corresponding to the business scenario.

[0078] As an example, in a patient follow-up business scenario, the preset business process question and answer template can be a preset follow-up question and answer process template, which automatically converts the business question and answer process described by professional and technical personnel into a decision-making process, and forms a prompt engineering model through pseudocode language.

[0079] Step 205, for the business dialogue, use the business question thinking chain and the business answer thinking chain to complete the dialogue process with the user device.

[0080] Here, the business response thinking chain refers to the thinking chain obtained by decomposing and symbolizing the business question and answer process of the corresponding business scenario based on the business domain knowledge base.

[0081] It should be noted that, here, using the business response thinking chain can refer to using the business response thinking chain based on the large language model, or it can also refer to using the business response intelligent agent. Among them, the business response intelligent agent is an artificial intelligence agent pre-built based on the business response thinking chain and the large language model.

[0082] During the business dialogue process, the business reply thinking chain or business reply agent is used to perform the following reply information generation operations:

[0083] First, current user dialogue information fed back by the user equipment in response to the service dialogue is received and recorded.

[0084] The business reply thought chain or business reply agent may be used to add the current user conversation information to the historical conversation record of the business conversation after receiving the current user conversation information fed back by the user device in response to the business conversation.

[0085] Optionally, various vector databases can be used to store historical conversation records of business conversations. Furthermore, business answer thinking chains or business answer agents and business question thinking chains or business question agents can share historical conversation records of business conversations through vector database technology.

[0086] Then, based on the historical dialogue records of the business dialogue, the current user dialogue information is searched in the search engine and the business domain knowledge base to obtain a first search result and a second search result.

[0087] Here, the business domain knowledge base refers to a knowledge base of a business domain corresponding to a business scenario. For example, for a patient follow-up business scenario, the business domain knowledge base may be a medical knowledge base.

[0088] In some optional implementations, the business response thinking chain or business response agent performs the process of "searching the current user conversation information on the search engine and the business domain knowledge base based on the historical conversation records of the business conversation to obtain the first search result and the second search result", which can also be performed as follows: first, the business demand understanding thinking chain or the business demand understanding agent is used to perform semantic understanding based on the user information of the target user and the business scenario related information, the historical conversation records of the business conversation and the current user conversation information to obtain the business demand information of the target user. Then, based on the business demand information of the target user, the current user conversation information is searched on the search engine and the business domain knowledge base to obtain the first search result and the second search result. That is, searching based on the understanding of the target user's needs can improve the efficiency and accuracy of the search.

[0089] Next, business session intermediate information is generated based on the first search result and the second search result.

[0090] As an example, the first search result and the second search result may be combined as the business conversation intermediate information.

[0091] It should be noted that the business response agent here can also be an agent based on the business response thinking chain. After the large language model is prompted and fine-tuned, it has the function of searching and generating information according to historical conversation records.

[0092] Finally, the intermediate information of the business dialogue and the current user dialogue information are sent to the business questioning agent.

[0093] It is understandable that the various steps in the above-mentioned reply information generation operation can be implemented by utilizing the business reply thinking chain based on the large language model, or utilizing the business reply intelligent agent, by calling various tools, applications or application programming interfaces with corresponding functions.

[0094] During the business conversation process, the business question thinking chain can be used based on the large language model, or the business question intelligent agent can be used to generate business conversation information based on the user information of the target user and business scenario-related information, the historical conversation records of the business conversation, the intermediate information of the business conversation and the current user conversation information, and send the business conversation information to the user device.

[0095] In some optional implementations, the process of generating business dialogue information based on the target user's user information and business scenario related information, historical dialogue records of business dialogues, business dialogue intermediate information and current user dialogue information by using a business question thinking chain or a business question intelligent agent based on a large language model can also be performed as follows: first, semantic understanding is performed based on the target user's user information and business scenario related information, historical dialogue records of business dialogues and current user dialogue information using a business demand understanding thinking chain or a business demand understanding intelligent agent to obtain the target user's business demand information. Then, business dialogue information is generated based on the target user's business demand information and the business dialogue intermediate information. That is, based on the target user's demand understanding and the retrieved business dialogue intermediate information, business dialogue information is generated to better meet the actual needs of the target user. It should be noted that the business demand understanding intelligent agent and the business question intelligent agent can be the same intelligent agent or different intelligent agents, and the present disclosure does not make specific limitations on this.

[0096] As an example, some sample user information and business scenario-related information in the same business scenario that have been manually labeled or intelligently processed, sample business conversation history records, and business conversation intermediate information input data can be used, and the corresponding business conversation information can be used as the output result label. The large language model can be trained based on the business response thinking chain to fine-tune the business response intelligent agent.

[0097] Through steps 201 to 205, it is possible to actively guide the target user to participate in the business scenario dialogue, and utilize multiple intelligent agents based on a large language model to conduct the dialogue process without human intervention, thereby reducing the labor cost of the business scenario dialogue.

[0098] In some optional implementations, the process 200 of the business scenario dialogue method may further include the following step 206:

[0099] Step 206: In response to the end of the business conversation, business processing information is generated based on the conversation record corresponding to the business conversation.

[0100] Here, the business processing information may be generated based on the conversation record corresponding to the business conversation according to the preset business processing information data format rule.

[0101] Business processing information is used to indicate the initiator of the business scenario, that is, how the business party specifically executes the business for the target user. For example, business processing information may include but is not limited to: business executor identification, business execution time, and specific business model type, etc. When the business scenario is a patient follow-up scenario, the business processing information may include the name of the follow-up doctor, follow-up time, follow-up frequency, follow-up method (telephone follow-up, instant messaging service follow-up, or door-to-door follow-up, etc.), follow-up content, and follow-up suggestions (advising patients to increase exercise, adjust diet, maintain a regular schedule, take relevant medications appropriately, etc.).

[0102] In the traditional follow-up process, the follow-up method is easily affected by human factors, making the follow-up data inaccurate and incomplete, and the follow-up data is prone to irregularities and difficulty in analysis, making it difficult to provide doctors with accurate patient health information. In the patient follow-up business scenario, the business processing information generated by step 206 is generated according to the preset business processing information data format rules, the data format is standardized, and is not related to personal factors, so that doctors can be provided with accurate patient health information.

[0103] In some optional implementations, the process 200 of the business scenario dialogue method may further include the following steps 207 and 208:

[0104] Step 207: Determine the service processing terminal device according to the service processing information.

[0105] Here, the corresponding business processing terminal device may be determined according to the business executor identifier in the business processing information. For example, when the business scenario is a patient follow-up scenario, the business processing terminal device may be a terminal device of the follow-up doctor.

[0106] Step 208: Send the service processing information to the service processing terminal device.

[0107] Here, the service processing information can be sent to the service processing terminal device in various ways, such as by making a phone call, sending a text message, sending an email, or sending an instant messaging message. This disclosure does not specifically limit this.

[0108] Through step 207 and step 208, the relevant terminal device of the business party can receive the business processing information, and then the relevant business executor or business person in charge can obtain the business processing information, and can provide corresponding services to the target user through the above business processing information. For example, in the patient follow-up scenario, the follow-up doctor can obtain the corresponding follow-up related information, such as the follow-up time, follow-up frequency and follow-up content.

[0109] The business dialogue method provided by the above-mentioned embodiment of the present disclosure utilizes multiple agents based on a large language model. First, a business demand understanding agent is used to perform semantic understanding based on the user information of the target user and the business scenario related information to determine whether the target user has a business demand. If it is confirmed that there is a business demand, the business demand understanding agent is used to initiate a business dialogue request to the user device corresponding to the target user, generate business demand guidance information, and send the business demand guidance information as the first business dialogue information of the business dialogue to the user device. After that, the dialogue process with the user device is completed through the business questioning agent and the business answering agent respectively. Thereby, it is realized to actively guide users to participate in the business scenario dialogue, and utilize multiple agents based on a large language model to conduct the dialogue process, without manual intervention throughout the process, thereby reducing the labor cost of the business scenario dialogue.

[0110] Further references Figure 3 As an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of a business scenario dialogue device, and the device embodiment is Figure 3 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0111] like Figure 3As shown, the business scenario dialogue device 300 of this embodiment includes: an acquisition module 301, a demand understanding module 302, a request dialogue module 303, a dialogue guiding module 304 and a business dialogue module 305. Among them, the acquisition module 301 is configured to acquire the user information of the target user and the business scenario related information; the demand understanding module 302 is configured to use the business demand understanding agent to perform semantic understanding based on the user information of the target user and the business scenario related information to obtain the business demand information of the target user, and the business demand information includes demand confirmation information indicating whether the business demand is met; the request dialogue module 303 is configured to respond to the demand confirmation information of the target user indicating that the business demand is met, and use the business demand understanding agent to initiate a business dialogue request to the user device corresponding to the target user; the dialogue guiding module 304 is configured to respond to the successful establishment of the business dialogue, and use the business questioning agent to generate business demand guidance information based on the user information of the target user, and send the business demand guidance information as the first business dialogue information of the business dialogue to the user device; the business dialogue module 305 is configured to For the business dialogue, the business questioning agent and the business answering agent are used to complete the dialogue process with the user device, wherein the business answering agent is used to perform the following reply information generation operations: receiving and recording the current user dialogue information fed back by the user device in response to the business dialogue, based on the historical dialogue record of the business dialogue, searching the current user dialogue information in a search engine and a business domain knowledge base respectively to obtain a first search result and a second search result, generating business dialogue intermediate information based on the first search result and the second search result, and sending the business dialogue intermediate information and the current user dialogue information to the business questioning agent; the business questioning agent is used to generate business dialogue information based on the user information and business scenario related information of the target user, the historical dialogue record of the business dialogue, the business dialogue intermediate information and the current user dialogue information, and send the business dialogue information to the user device.

[0112] In this embodiment, the specific processing of the acquisition module 301, the demand understanding module 302, the request dialogue module 303, the dialogue guidance module 304 and the business dialogue module 305 of the business scenario dialogue device 300 and the technical effects thereof can be referred to respectively. Figure 2 The relevant descriptions of step 201, step 202, step 203, step 204 and step 205 in the corresponding embodiment are not repeated here.

[0113] In some optional implementations, the business requirement understanding agent, the business questioning agent, and the business answering agent may be artificial intelligence agents generated based on a large language model and a preset business question and answer process template.

[0114] In some optional implementations, the business requirement understanding agent and the business questioning agent may be the same agent.

[0115] In some optional implementations, the business scenario dialogue device 300 may further include: an information generation module ( Figure 3 ), configured to generate business processing information based on the conversation record corresponding to the business conversation in response to the end of the business conversation.

[0116] In some optional implementations, the business scenario dialogue device 300 may further include: a device determination module ( Figure 3 ), configured to determine a service processing terminal device according to the service processing information; and an information sending module ( Figure 3 ), configured to send the business processing information to the business processing terminal device.

[0117] In some optional implementations, the business scenario may be a patient follow-up scenario, and the business scenario related information may include at least one of the following: clinical examination results, clinical diagnosis results, and follow-up related information, and the business demand is a follow-up demand.

[0118] In some optional implementations, the preset business question and answer process template may be a preset follow-up question and answer process template.

[0119] It should be noted that the implementation details and technical effects of each module in the business dialogue device provided by the embodiment of the present disclosure can refer to the description of other embodiments in the present disclosure, and will not be repeated here.

[0120] Reference below Figure 4 , which shows a schematic diagram of the structure of a computer system 400 suitable for implementing the electronic device of the present disclosure. Figure 4 The computer system 400 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0121] like Figure 4 As shown, the computer system 400 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 401, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 402 or a program loaded from a storage device 408 to a random access memory (RAM) 403. Various programs and data required for the operation of the computer system 400 are also stored in the RAM 403. The processing device 401, the ROM 402, and the RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0122] Typically, the following devices may be connected to the I / O interface 405: input devices 406 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, etc.; output devices 407 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 408 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 409. The communication devices 409 may allow the computer system 400 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 4 The computer system 400 of an electronic device having various devices is shown, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0123] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device 409, or installed from a storage device 408, or installed from a ROM 402. When the computer program is executed by the processing device 401, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0124] It should be noted that the computer-readable medium mentioned above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. Computer readable signal media may also be any computer readable medium other than computer readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0125] The computer-readable medium may be included in the electronic device, or may exist independently without being installed in the electronic device.

[0126] The computer readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device implements the following Figure 2 The illustrated embodiment and its optional implementations illustrate a business scenario dialogue method.

[0127] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Python, Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0128] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0129] The modules involved in the embodiments described in the present disclosure may be implemented by software or hardware. The name of a module does not limit the module itself in some cases. For example, an acquisition module may also be described as a "module for acquiring user information of a target user and information related to a business scenario".

[0130] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

Claims

1. A business scenario dialogue method, comprising: Obtain user information and business scenario-related information of the target user; Using the business demand understanding thinking chain, semantic understanding is performed based on the user information of the target user and the business scenario related information to obtain the business demand information of the target user, where the business demand information includes demand confirmation information indicating whether the business demand is met, wherein the business demand understanding thinking chain refers to a thinking chain obtained by performing step decomposition and symbolic expression on the business demand understanding process of the corresponding business scenario based on the business domain knowledge base; In response to the target user's demand confirmation information indicating that there is a business demand, initiating a business session request to a user device corresponding to the target user by using the business demand understanding thinking chain; In response to the successful establishment of the business dialogue, using a business question thinking chain, generating business demand guidance information based on the user information of the target user, and sending the business demand guidance information as the first business dialogue information of the business dialogue to the user device; For the business dialogue, the business question thinking chain and the business answer thinking chain are used to complete the dialogue process with the user device, wherein the business answer thinking chain is used to perform the following reply information generation operation: receiving and recording the current user dialogue information fed back by the user device in response to the business dialogue, based on the historical dialogue record of the business dialogue, searching the current user dialogue information in a search engine and a business domain knowledge base to obtain a first search result and a second search result, generating business dialogue intermediate information based on the first search result and the second search result, and sending the business dialogue intermediate information and the current user dialogue information to the business question thinking chain; the business question thinking chain is used to generate business dialogue information based on the user information and business scenario related information of the target user, the historical dialogue record of the business dialogue, the business dialogue intermediate information and the current user dialogue information, and send the business dialogue information to the user device; Among them, the business demand understanding thinking chain, the business question thinking chain, and the business answer thinking chain are thinking chains generated based on a large language model and a preset business question and answer process template.

2. The method according to claim 1, wherein: The business demand understanding thinking chain and the business questioning thinking chain are the same thinking chain.

3. The method according to claim 1, wherein: The method further comprises: In response to the business conversation ending, business processing information is generated based on the conversation record corresponding to the business conversation.

4. The method according to claim 3, wherein: The method further comprises: Determining a service processing terminal device according to the service processing information; The service processing information is sent to the service processing terminal device.

5. The method according to claim 1, wherein: The business scenario is a patient follow-up scenario, and the business scenario related information includes at least one of the following: clinical examination results, clinical diagnosis results, and follow-up related information, and the business demand is a follow-up demand.

6. The method according to claim 5, wherein: The preset business question and answer process template is a preset follow-up question and answer process template.

7. A business dialogue device, comprising: An acquisition module, configured to acquire user information of a target user and information related to a business scenario; A demand understanding module is configured to use a business demand understanding thinking chain to perform semantic understanding based on the user information of the target user and relevant information of the business scenario to obtain the business demand information of the target user, wherein the business demand information includes demand confirmation information indicating whether the business demand is met; The request dialogue module is configured to, in response to the target user's demand confirmation information indicating that there is a business demand, initiate a business dialogue request to the user device corresponding to the target user using the business demand understanding thinking chain, wherein the business demand understanding thinking chain refers to a thinking chain obtained by performing step decomposition and symbolic expression on the business demand understanding process of the corresponding business scenario based on a business domain knowledge base; a dialogue guiding module configured to generate business demand guiding information based on the user information of the target user by using a business question thinking chain in response to the business dialogue being successfully established, and to send the business demand guiding information as the first business dialogue information of the business dialogue to the user device; The business dialogue module is configured to complete the dialogue process between the user device and the business dialogue using the business question thinking chain and the business answer thinking chain, wherein the business answer thinking chain is used to perform the following reply information generation operation: receiving and recording the current user dialogue information fed back by the user device in response to the business dialogue, searching the current user dialogue information in a search engine and a business domain knowledge base to obtain a first search result and a second search result based on the historical dialogue record of the business dialogue, generating business dialogue intermediate information based on the first search result and the second search result, and sending the business dialogue intermediate information and the current user dialogue information to the business question thinking chain; the business question thinking chain is used to generate business dialogue information based on the user information and business scenario related information of the target user, the historical dialogue record of the business dialogue, the business dialogue intermediate information and the current user dialogue information, and send the business dialogue information to the user device; Among them, the business demand understanding thinking chain, the business question thinking chain, and the business answer thinking chain are thinking chains generated based on a large language model and a preset business question and answer process template.

8. An electronic device comprising: one or more processors; a storage device having one or more programs stored thereon, When the one or more programs are executed by the one or more processors, the one or more processors are enabled to implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, wherein: When the computer program is executed by one or more processors, the method according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program / instruction, wherein the computer program / instruction is executed by a processor to implement the method according to any one of claims 1 to 6.

Citation Information

Patent Citations

  • Active conversation initiation method and device based on user identity

    CN110265009A

  • Man-machine conversation method and device

    CN110704703A

  • Conversation setting method and device, server and computer readable storage medium

    CN111897931A

  • System and method for formulating intelligent return visit through big data

    CN115954116A

  • Language interaction method and device based on multi-perception ability, equipment and medium

    CN117093893A