Logic processing method, system and equipment of intelligent question and answer model, and medium

Abstract

The invention belongs to the technical field of intelligent logistics, and particularly relates to a logic processing method, system and device for an intelligent question and answer model in a freight scene and a medium, and the method comprises the steps: when a user initiates a dialogue, analyzing a business target according to the dialogue content of the user; performing multiple rounds of conversations with the user based on the business target, and collecting related information of the business target from the multiple rounds of conversations; wherein for each round of dialogue of the user, the correlation between the current dialogue content and the service target is evaluated: if the dialogue content is not related to the service target, a chat mode is started to call corresponding data information in a database to make a reply according to the demand of the user; and if the dialogue content is related to the business target, collecting and updating the related information of the business target, and re-evaluating the information collection completion degree of the business target after the information collection completion degree of the business target so as to perform reflection according to the information collection completion degree of the business target and confirm whether a user needs to be guided to provide the related information of business target missing.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent logistics technology, and specifically relates to a logic processing method, system, device, and medium of an intelligent question-answering model in a freight scenario. Background Art

[0002] In recent years, with the emergence of large language models (LLM), AI applications based on LLM have exploded, and agent is an important form of AI application. Agent is mainly composed of LLM, task planning (Planning), memory (Memory) and tools (Tools). Among them, LLM is the core brain, which combines planning tasks with historical memory and calls tools to achieve business goals.

[0003] However, the agent application based on LLM currently has the following two problems: First, the LLM illusion leads to poor question-answering results: Agents with LLM as their core brain often give irrelevant answers and poor question-answering results in real business questions and answers due to the LLM illusion problem.

[0004] Second, the agent has a poor sense of purpose and cannot effectively achieve business goals: the agent uses Planning to break down tasks and call tools to achieve business goals, but tool parsing errors and lack of focus in multi-round conversation memory often occur during execution. The agent has a poor sense of purpose and cannot effectively achieve business goals. Summary of the invention

[0005] In view of the shortcomings of the prior art mentioned above, the purpose of the present invention is to provide a logical processing method that can be applied to agent applications, adding processing processes such as current information collection, effect progress evaluation, reflection, and next step direction planning, so as to solve the problem of poor question and answer effects caused by hallucinations in the agent's LLM calls, and improve the processing efficiency and accuracy of the business goals required by users.

[0006] To achieve the above-mentioned purpose and other related purposes, the present invention provides a logical processing method of an intelligent question-answering model in a freight scenario, including: when a user initiates a conversation, analyzing the business goal according to the user's conversation content; conducting multiple rounds of conversations with the user based on the business goal, and collecting relevant information of the business goal from the multiple rounds of conversations; wherein, for each round of conversations with the user, evaluating the relevance of the current conversation content to the business goal: if the conversation content is not relevant to the business goal, turning on the chat mode to call the corresponding data information in the database to respond to the user's needs; if the conversation content is relevant to the business goal, collecting and updating the relevant information of the business goal, and re-evaluating the completion of the information collection of the business goal after the update, so as to reflect on the completion of the information collection of the business goal, and confirm whether it is necessary to guide the user to provide the relevant information missing from the business goal.

[0007] According to a specific embodiment of the present invention, when the conversation content is related to the business goal, the relevant information about the business goal is collected by analyzing the conversation content, and after the collection is completed, it also includes: reflecting based on the collected relevant information to confirm the completeness and accuracy of the information, and updating the relevant information of the business goal when the completeness and accuracy of the information meet the standards.

[0008] According to a specific embodiment of the present invention, reflection is performed based on the collected relevant information to confirm the information completeness and information accuracy, and the step of updating the relevant information of the business goal when the information completeness and information accuracy meet the standards includes: when the information is incomplete, guiding the user to supplement it; and / or when the information accuracy is lower than a preset threshold, requesting the user to verify.

[0009] According to a specific embodiment of the present invention, after the chat mode is turned on, it also includes: detecting the number of conversation rounds that the user has had in chatting, and when the number of conversation rounds exceeds a preset threshold, reflecting on the completion of information collection of the business goal to confirm whether it is necessary to guide the user's conversation content to return to the business goal, and provide relevant information that is missing from the business goal.

[0010] According to a specific embodiment of the present invention, after updating the relevant information of the business goal and re-evaluating the completion of information collection of the business goal, it also includes: calling an interface to loop back the user's historical conversations for deep reflection to correct the hallucination analysis of the user's conversation content; wherein, the deep reflection is thinking, deducing, observing, and reflecting based on the user's historical conversations.

[0011] According to a specific embodiment of the present invention, it also includes: after the relevant information of the business target is collected, providing intelligent services to the user according to all the relevant information of the business target.

[0012] According to a specific embodiment of the present invention, the intelligent question-answering model adopts an LLM-based agent application.

[0013] A logic processing system of an intelligent question-answering model in a freight scenario, comprising: a target determination module, which is used to analyze the current business target according to the user's conversation content when a user initiates a conversation; an information collection module, which is used to conduct multiple rounds of conversations with the user based on the business target, and collect relevant information of the business target from the multiple rounds of conversations; wherein, for each round of conversations with the user, the relevance of the current conversation content to the business target is evaluated: if the conversation content is not relevant to the business target, the chat mode is turned on to call the corresponding data information in the database to respond to the user's needs; if the conversation content is relevant to the business target, the relevant information of the business target is collected and updated, and the completion of the information collection of the business target is re-evaluated after the update, so as to reflect on the completion of the information collection of the business target and confirm whether it is necessary to guide the user to provide the relevant information missing from the business target.

[0014] An electronic device includes a processor, wherein the processor is coupled to a memory, and the memory stores program instructions. When the program instructions stored in the memory are executed by the processor, the above-mentioned method is implemented.

[0015] A computer-readable storage medium includes a program. When the program is run on a computer, the computer is enabled to execute the method described above.

[0016] The present invention provides a logical processing method of an intelligent question-answering model in a freight scenario. The method is based on the target reflection agent of LLM, which confirms the relevance of the user's current conversation content with his or her business goal, distinguishes normal conversations and casual conversations about business needs, and collects relevant information about the goal from them. The accuracy and completeness of information collection are ensured through reflection, and finally, the user's conversation content is guided to return to his or her business needs through reflection before information collection is completed.

[0017] At the same time, after the chat ends or the chat conversation exceeds a certain number of rounds, reflection is carried out, and the user's conversation content is further guided back to its business needs, specifically solving the problem that the agent cannot return to the business process normally after chatting with the user.

[0018] The present invention effectively reduces the frequency of LLM hallucination through multiple reflections at different times, improves the accuracy of text parsing by the agent, and further improves the question-answering effect of the agent. By evaluating the completion of information collection, the agent realizes real-time information status query and interaction synchronization between multiple rounds of dialogue, and specifically solves the problem that the current agent has a poor sense of goal and cannot achieve business goals. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 A flowchart of a specific embodiment of a logic processing method of an intelligent question-answering model in a freight scenario provided by the present invention; Figure 2 A flowchart of another specific embodiment of a logic processing method of an intelligent question-answering model in a freight scenario provided by the present invention; Figure 3 A schematic diagram of a question-answering method in practical application of the intelligent question-answering model provided by the present invention; Figure 4 Another question-answering schematic diagram of the intelligent question-answering model provided by the present invention in practical applications; Figure 5 This is a structural schematic diagram of a specific embodiment of a logic processing system of an intelligent question-answering model in a freight scenario provided by the present invention; Figure 6 The present invention provides a structural block diagram of a specific embodiment of an electronic device. DETAILED DESCRIPTION

[0019] In order to facilitate understanding of the present application, the present application will be described more fully below with reference to the relevant drawings. Embodiments of the present application are provided in the drawings. However, the present application can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present application more thorough and comprehensive.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application.

[0021] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0022] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it is obvious to those skilled in the art that the embodiments of the present invention can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present invention difficult to understand.

[0023] Example 1 See also Figure 1 The logical processing method of the intelligent question-answering model in a freight scenario shown includes: Step S100: when a user initiates a conversation, the business goal is analyzed according to the conversation content of the user.

[0024] First of all, it should be noted that the intelligent question and answer model mentioned in this embodiment adopts an LLM-based agent application and is mainly used in freight scenarios. Specifically, it serves as an intelligent AI assistant in the logistics platform, such as an intelligent customer service AI assistant, an intelligent marketing AI assistant, an intelligent invitation AI assistant, etc., to provide users with convenient logistics services.

[0025] Therefore, when users use the logistics platform to handle related business, they can consult and understand through the intelligent AI assistant, and will initiate a dialogue with the intelligent AI assistant and ask questions accordingly. In this regard, the intelligent AI assistant needs to first identify and analyze the user's business needs based on the user's dialogue content, that is, the business goals that need to be handled, such as freight business, including intra-city freight, inter-city freight, etc.

[0026] What can be understood here is that usually when a user initiates a conversation with an intelligent AI assistant, he or she will simply explain his or her needs, such as "How much is the freight from XX to XX", or "How long does it take to ship XX to XX", or "I need to place an order", etc. Based on the above or similar conversation content, the intelligent AI assistant can identify and analyze the user's business goals. However, when the content of the conversation initiated by the user does not contain relevant information about his or her business needs, the intelligent AI assistant can guide the user to provide detailed explanations, such as replying "Hello, what business do you need to handle this time", and then identify and analyze the business goals from the user's new round of conversation.

[0027] Step S200, based on the business goal, multiple rounds of dialogue are conducted with the user, and relevant information of the business goal is collected from the multiple rounds of dialogue; wherein, for each round of dialogue of the user, the relevance of the current dialogue content with the business goal is evaluated: Step S210: If the conversation content is not related to the business goal, the chat mode is turned on to call the corresponding data information in the database to respond to the user's needs.

[0028] Step S220, if the conversation content is related to the business goal, collect and update the relevant information of the business goal, and re-evaluate the completion of information collection of the business goal after the update, so as to reflect on the completion of information collection of the business goal and confirm whether it is necessary to guide the user to provide the relevant information missing from the business goal.

[0029] After confirming the user's business goal, relevant information about the business goal can be collected accordingly to provide intelligent services to the user. To this end, the intelligent AI assistant can continuously collect all relevant information about the business goal through multiple rounds of dialogue with the user. In this process, the intelligent AI assistant mainly responds to the user's dialogue content, and then guides the user to provide relevant information about the business goal on this basis.

[0030] It should be noted here that the determination of the user's business goal and the collection of relevant information about the business goal can be performed simultaneously, and the specific identification and analysis is based on the user's conversation content. For example, when the conversation content initiated by the user contains detailed information, that is, "from No. 27 Changbai Road, Fengtai District, Beijing to Changping District, Beijing, transport 30 kilograms of cement, small van model", it can be determined that the user's business goal this time is intra-city freight, and the relevant information about the business goal includes "starting point: No. 27 Changbai Road, Fengtai District, Beijing", "end point: Changping District, Beijing", "cargo: cement", "cargo weight: 30 kilograms", and "car model: small van". It can be seen that when the user's conversation content is more detailed, the number of conversation rounds that need to collect relevant information can be greatly reduced. Of course, if the above business goal is to be achieved, the relevant information about the business goal includes but is not limited to the above, so it may be necessary to further collect relevant information until all relevant information about the business goal is collected.

[0031] For details, see Figure 2 As shown, in the process of collecting the user's business goal, for each round of conversation with the user, it is first necessary to evaluate the relevance of the user's current conversation content to the business goal. It can be understood here that the user may consult the intelligent AI assistant for relevant information other than the freight business included in the current logistics platform, such as understanding the weather of the day, the news of the day, etc. Correspondingly, the conversation content initiated by the user does not include relevant information about the business goal, and it cannot be collected from it.

[0032] Therefore, when the user's conversation content is not related to the business goal, it is recognized that the user is currently chatting, and the chat mode can be turned on to make adaptive responses based on the user's actual needs. Of course, since the main function of the intelligent AI assistant is to provide users with consulting responses on the current logistics platform-related business information, the corresponding configured data information is mostly related to the freight business, and the user's chat conversation content needs to be queried and called from the internal database, and the corresponding data information is responded to in turn. However, when the database does not store relevant data information, it is necessary to make an explanatory response to the user, while maintaining the user's good mood as much as possible, and guide the user's conversation content back to the business goal, such as "I'm sorry, I don't provide any functions related to XX, but if you have other business issues, you can consult", to ensure the user's experience of the current logistics platform and maintain smooth conversation.

[0033] Secondly, when the number of rounds of small talk by users exceeds a certain number, it means that the content of the user's conversation is gradually deviating from the current business goal. It can be understood here that the number of rounds of small talk by users is a conversation initiated by the user and is small talk, and the intelligent AI assistant makes a corresponding reply, which is considered a round of conversation.

[0034] Therefore, when the intelligent AI assistant detects that the number of rounds of small talk exceeds the preset threshold, it will learn human thinking and reflect accordingly, that is, reflect based on the user's historical conversation content and the conversation content about the user's expected response to the current conversation. The main purpose of this reflection is to continue to guide the user's conversation content back to the business goal and provide relevant information missing from the business goal. The intelligent AI assistant needs to make an adaptive and comprehensive response to the above content. In this way, while maintaining the user experience as much as possible, it can also improve the efficiency of collecting information related to business goals.

[0035] It can also be understood that the comprehensive reply is to add additional reply content to guide the user to provide information related to the business goal on the basis of the reply content required by the user's current conversation, so as to form a comprehensive reply.

[0036] When the user's conversation content is related to the business goal, it means that it may contain relevant information about the business goal, and the relevant information can be collected through identification and analysis. It should be added here that after collecting relevant information, the intelligent AI assistant also needs to reflect on the collected relevant information, and the main purpose of this reflection is to confirm the completeness and accuracy of the information. For example, based on the above "starting point: No. 27, Changbai Road, Fengtai District, Beijing" and "end point: Changping District, Beijing", it can be seen that the address information of the freight end point lacks a detailed specific location, or the user's conversation content is "No. 21, Shangdi 6th Street, Haidian District", and the intelligent AI assistant analyzes the address through other information of the user as "No. 21, Shangdi 6th Street, Haidian District, Beijing". Therefore, the accuracy of the relevant information obtained by identification and analysis will be calculated, and the user will be requested to verify when the accuracy is lower than the preset threshold. Therefore, in either of the above situations, an adaptive comprehensive response is required, that is, based on the content that the user needs to reply to in the current conversation, the user is guided to supplement or request verification based on the actual situation.

[0037] Of course, if the information completeness and accuracy of the collected relevant information meet the standards, the relevant information of the business goal will be updated. At the same time, after the information is updated, it is necessary to re-evaluate the information collection completion of the business goal to confirm whether the information has been collected. Secondly, before making a reply, the intelligent AI assistant needs to call the interface to loop back the user's historical conversation for deep reflection, which is achieved through think, act, obversion, and reflection. On the one hand, it corrects the common illusion problems when LLM replies, and on the other hand, it improves efficiency, reduces the number of cycles, and improves the accuracy of text parsing to avoid the intelligent AI assistant thinking too much and hallucinating, resulting in incorrect parsing of the user's conversation content. Furthermore, the intelligent AI assistant also needs to reflect on the current information collection completion of the business goal, and the main purpose of this reflection is to confirm whether it is necessary to continue to guide the user to provide relevant information missing from the business goal, so as to make an adaptive and comprehensive reply.

[0038] Based on the above, after the relevant information of the business goal is collected, the intelligent AI assistant can stop reflecting, that is, no matter what the user's conversation content is, it only needs to make a corresponding reply, unless a new business goal is identified and analyzed, then continue to make a comprehensive reply according to the above logic. Without making too many restrictions on this, the modifications and embellishments made to the embodiments of the present invention by those skilled in the art without departing from the spirit of the present invention still fall within the scope of the invention patent application of the present invention.

[0039] Step S300: After the relevant information of the business target is collected, intelligent services are provided to the user based on all the relevant information of the business target.

[0040] It is understandable that after collecting all relevant information of the user's business goals, convenient freight services can be provided accordingly, for example, an order can be directly generated for user confirmation, etc., without making too many restrictions on this. The modifications and improvements made to the embodiments of the present invention by those skilled in the art without departing from the spirit of the present invention still fall within the scope of the invention patent application of the present invention.

[0041] In a specific embodiment, the specific practical application of the technical solution of this application can be seen in Figure 3 , 4 As shown in the figure, in the freight scenario, when the user needs to place an order, it means that the user's business goal this time is to handle freight business, and the freight starting point and freight destination information can be collected accordingly. At the same time, in the process of communicating with users, reflection is used to improve the efficiency and accuracy of target information collection, and users are guided to business goals, thereby effectively shortening the number of dialogue rounds for collecting information, avoiding ineffective communication with users, and efficiently realizing users' business needs while ensuring user experience.

[0042] It should be noted that the step division of the above methods is only for clear description. When implemented, they can be combined into one step or some steps can be split and decomposed into multiple steps. As long as they contain the same logical relationship, they are all within the scope of protection of this application; adding insignificant modifications to the algorithm or process or introducing insignificant designs without changing the core design of the algorithm and process are all within the scope of protection of this application.

[0043] Example 2 See also Figure 5 As shown, this embodiment also provides a logic processing system of an intelligent question-answering model in a freight scenario, including: The target determination module 10 is used to analyze the current business target according to the content of the user's conversation when the user initiates the conversation.

[0044] The information collection module 20 is used to conduct multiple rounds of dialogues with the user based on the business goal, and collect relevant information of the business goal from the multiple rounds of dialogues; wherein, for each round of dialogue of the user, the relevance of the current dialogue content with the business goal is evaluated: If the content of the conversation is not relevant to the business goal, the chat mode is turned on to call the corresponding data information in the database to respond to the user's needs.

[0045] If the conversation content is related to the business goal, then collect and update the relevant information of the business goal, and re-evaluate the completion of information collection of the business goal after the update, so as to reflect on the completion of information collection of the business goal and confirm whether it is necessary to guide the user to provide the relevant information missing from the business goal.

[0046] It should be noted that the logic processing system of the intelligent question and answer model in the freight scenario provided in the above embodiment and the logic processing method of the intelligent question and answer model in the freight scenario provided in the above embodiment 1 belong to the same concept, wherein the specific manner in which each module and unit performs the operation has been described in detail in the method embodiment and will not be repeated here. In actual application, the logic processing method of the intelligent question and answer model in the freight scenario provided in the above embodiment 1 can distribute the above functions to different functional modules as needed, that is, divide the internal structure of the device into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0047] Example 3 See also Figure 6 As shown, an embodiment of the present application further provides an electronic device, comprising a memory 2, a processor 1, and a program stored in the memory and executable on the processor, wherein the processor implements the steps of any of the above methods when executing the program.

[0048] Among them, the memory includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (for example: SD or DX memory, etc.), magnetic memory, disk, optical disk, etc. In some embodiments, the memory can be an internal storage unit of the electronic device, such as a mobile hard disk of the electronic device. In other embodiments, the memory can also be an external storage device of the electronic device, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device. Further, the memory can also include both an internal storage unit of the electronic device and an external storage device. The memory can not only be used to store application software and various types of data installed in the electronic device, but also can be used to temporarily store data that has been output or is to be output.

[0049] In some embodiments, the processor may be composed of an integrated circuit, for example, a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and a combination of various control chips. The processor is the control core (Control Unit) of the electronic device, which connects the various components of the entire electronic device using various interfaces and lines, and executes various functions of the electronic device and processes data by running or executing programs or modules stored in the memory, and calling data stored in the memory.

[0050] The processor executes the operating system of the electronic device and various installed application programs. The processor executes the application programs to implement the steps in the above method embodiment.

[0051] Exemplarily, the program may be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of program instruction segments capable of completing specific functions, which are used to describe the execution process of the program in the electronic device.

[0052] The above-mentioned integrated unit implemented in the form of software function modules can be stored in a computer-readable storage medium. The above-mentioned software function modules are stored in a storage medium and include several instructions for enabling a computer device (which can be a personal computer, a computer device, or a network device, etc.) or a processor to perform part of the functions of various embodiments of the present invention.

[0053] In summary, the present invention provides a logical processing method of an intelligent question-answering model in a freight scenario. Based on the target reflection agent of LLM, by confirming the relevance of the user's current conversation content with his or her business goals, the normal conversation and the chat conversation about business needs are distinguished, and reflection is performed after the chat ends or after the chat conversation exceeds a certain number of rounds, and the user's conversation content is guided to return to its business needs, so as to solve the problem that the agent cannot return to the business process normally after chatting with the user.

[0054] At the same time, through multiple reflections at different times, the frequency of LLM hallucinations can be effectively reduced, the accuracy of the agent's text parsing can be improved, and then the agent's question-and-answer effect can be improved. By evaluating the completion of information collection, the agent can achieve real-time information status query and interaction synchronization between multiple rounds of dialogue, and specifically solve the problem that the current agent has a poor sense of purpose and cannot achieve business goals. The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by a person of ordinary skill in the art without departing from the spirit and technical concept disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. A logical processing method of an intelligent question-answering model in a freight scenario, characterized in that: include: When a user initiates a conversation, analyze the business objectives based on the user's conversation content; Conduct multiple rounds of dialogue with the user based on the business goal, and collect relevant information of the business goal from the multiple rounds of dialogue; wherein, for each round of dialogue with the user, evaluate the relevance of the current dialogue content with the business goal: If the content of the conversation is not relevant to the business goal, the chat mode is turned on to call the corresponding data information in the database to respond to the user's needs; If the conversation content is related to the business goal, then collect and update the relevant information of the business goal, and re-evaluate the completion of information collection of the business goal after the update, so as to reflect on the completion of information collection of the business goal and confirm whether it is necessary to guide the user to provide the relevant information missing from the business goal.

2. The logic processing method of the intelligent question-answering model in the freight scenario according to claim 1 is characterized in that: When the conversation content is related to the business goal, relevant information about the business goal is collected by analyzing the conversation content, and after the collection is completed, the method further includes: Reflect on the collected relevant information to confirm the completeness and accuracy of the information, and update the relevant information of the business objectives when the completeness and accuracy of the information meet the standards.

3. The logic processing method of the intelligent question-answering model in the freight scenario according to claim 2 is characterized in that: The steps of reflecting on the collected relevant information to confirm the completeness and accuracy of the information, and updating the relevant information of the business objectives when the completeness and accuracy of the information meet the standards include: When information is incomplete, guide the user to provide additional information; and / or When the accuracy of the information is lower than the preset threshold, the user is requested to verify.

4. The logic processing method of the intelligent question-answering model in the freight scenario according to claim 1 is characterized in that: After turning on the chat mode, it also includes: Detect the number of conversation rounds that the user has engaged in small talk, and when the number of conversation rounds exceeds a preset threshold, reflect on the completion of information collection of the business goal to confirm whether it is necessary to guide the user's conversation content to return to the business goal, and provide relevant information missing from the business goal.

5. The logic processing method of the intelligent question-answering model in the freight scenario according to claim 1 is characterized in that: After updating the relevant information of the business goal and re-evaluating the completion of information collection of the business goal, it also includes: Call the interface to loop back the user's historical conversations for deep reflection to correct the hallucination analysis of the user's conversation content; Among them, the deep reflection is thinking, deduction, observation and reflection based on the user's historical conversations.

6. The logic processing method of the intelligent question-answering model in the freight scenario according to claim 1 is characterized in that: Also includes: When the relevant information of the business target is collected, intelligent services are provided to the user based on all the relevant information of the business target.

7. The logic processing method of the intelligent question-answering model in the freight scenario according to claim 1 is characterized in that: The intelligent question-answering model adopts an agent application based on LLM.

8. A logical processing system for an intelligent question-answering model in a freight scenario, characterized in that: include: The target determination module is used to analyze the business target of this session based on the content of the user's conversation when the user initiates the conversation; An information collection module is used to conduct multiple rounds of dialogues with the user based on the business goal, and collect relevant information of the business goal from the multiple rounds of dialogues; wherein, for each round of dialogue of the user, the relevance of the current dialogue content to the business goal is evaluated: If the content of the conversation is not relevant to the business goal, the chat mode is turned on to call the corresponding data information in the database to respond to the user's needs; If the conversation content is related to the business goal, then collect and update the relevant information of the business goal, and re-evaluate the completion of information collection of the business goal after the update, so as to reflect on the completion of information collection of the business goal and confirm whether it is necessary to guide the user to provide the relevant information missing from the business goal.

9. An electronic device, characterized in that: The method comprises a processor, wherein the processor is coupled to a memory, wherein the memory stores program instructions, and when the program instructions stored in the memory are executed by the processor, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The invention comprises a program, which, when being executed on a computer, causes the computer to execute the method according to any one of claims 1 to 7.

Images