Method and system of multi-round ticket booking intelligent agent based on large model

Abstract

The invention belongs to the technical field of artificial intelligence and intelligent customer service, and discloses a method and system for a multi-round ticket booking agent based on a large model, and the method comprises the following steps: S1, obtaining the definition of a multi-round ticket booking conversation for related elements, and taking the definition as the basis for constructing the multi-round ticket booking agent, the definition of the related elements comprises the definition of multiple rounds of interaction intentions, the jump relation of multiple rounds of states and the function definition of a ticket booking service API (Application Program Interface); according to the invention, the intention recognition module can represent the real demand of a user through a structured intention, and then can perform pruning optimization on the intention when a large model is used for intention recognition reasoning; the difficulty of intention recognition in a dialogue management task can be effectively reduced through supervised fine tuning, and the system interaction quality is optimized; each round of dialogue of the user can be replied and real-time flight information can be acquired through dialogue state updating tasks and API scheduling in the multi-round ticket booking agent system.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence and intelligent customer service technology, and more specifically, to a method and system for a multi-round ticket booking agent based on a large model. Background Art

[0002] As a breakthrough in the field of artificial intelligence, large models (generative large models, such as LLM) have demonstrated outstanding capabilities in text generation, language understanding, and dialogue management. By pre-training on large amounts of text data, these models can capture complex language patterns and generate responses that fit the context. If LLM is applied to multi-round ticket booking agents, it can effectively improve the intelligence level and interaction quality of the dialogue system.

[0003] With the rapid development of Internet technology, online ticket booking services have become a part of people's daily lives, especially playing an important role in the fields of tourism and transportation.

[0004] However, traditional ticket booking systems are mostly based on single-round interactions, and users need to manually search and filter information. This process is not only time-consuming but may also lead to errors due to improper user operations, affecting the user experience. In addition, existing dialogue systems based on large models still face challenges in multi-round interactions and specific domain tasks, such as how to maintain contextual consistency in multi-round dialogues, how to handle complex user intentions and preferences, and how to provide accurate information and decision support during the dialogue process. In addition, there are also limitations in personalized services and user intention prediction.

[0005] In view of this, the present invention proposes a method and system for multi-round ticket booking agent based on a large model to solve the above problems. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a method of multi-round ticket booking agent based on a large model, comprising the following steps:

[0007] S1. Obtain the definition of relevant elements of the multi-round ticket booking dialogue and use it as the basis for building a multi-round ticket booking agent. The definition of relevant elements includes the definition of multi-round interaction intentions, the jump relationship of multi-round states, and the function definition of the ticket booking service API;

[0008] S2. Training and fine-tuning of large models, wherein the large model fine-tuning includes task definition, dataset construction, and supervised fine-tuning;

[0009] S3. Establish a prompt word project and use it in combination with the fine-tuned large model. Establishing the prompt word project includes updating the dialogue status and calling the API.

[0010] Furthermore, the definition of the multi-round interaction intention includes:

[0011] Obtain the business requirements for multi-round ticket booking conversations, and define the intent labels and related slot information for each round in the multi-round ticket booking system based on the business requirements;

[0012] The question entered by the user corresponds to an intent label, and the candidate slot information in the question entered by the user is assigned a value.

[0013] Furthermore, the jump relationship of the multi-round state includes:

[0014] A mapping relationship between each round of dialogue and the dialogue state in multiple rounds of ticket booking dialogue is established, wherein each round of dialogue state and the next round of dialogue state are represented by a mapping relationship.

[0015] Furthermore, the function definition of the ticket booking service API includes:

[0016] Define the API input, output parameters and interaction methods to adapt the API input, output parameters and interaction methods to the large model.

[0017] Furthermore, the task definition includes:

[0018] Obtain and combine the business scenarios of multiple rounds of ticket booking conversations, decompose the multiple rounds of ticket booking conversations into structured data according to the business scenarios, and then obtain the definitions of intents, instruction sets, and large model reasoning tasks;

[0019] The intent is represented by structured Json data. For each intent, it needs to include the current dialogue state, vertical domain label, intent recognition label, input question, probability of intent recognition label and slot information related to the intent;

[0020] The instruction set is a normalized representation of the system output format in the dialogue system;

[0021] The large model reasoning task is to include an intention recognition module that combines context information in multiple rounds of ticket booking conversations, and then output structured intention information through the intention recognition module.

[0022] Furthermore, the input of the intention recognition module includes: the input question of the user's current state, the intention state information of the previous round, and the instruction set information of the previous round, which is expressed as:

[0023] f(query n , Intent n-1 , Skill n-1 )=Intent n

[0024] In the formula, query n Indicates the input question of the user's current status in the nth round of dialogue, Intent n-1 Indicates the intention status information of the n-1th round of dialogue, Skill n-1 Indicates the instruction set information of the n-1th round of dialogue, f(query n , Intent n-1 , Skill n-1 ) represents the input of the intent recognition module, Intent n It indicates that the intention recognition module outputs the intention information of the nth round of dialogue.

[0025] Furthermore, the dataset construction includes the following process:

[0026] Step 1: Use a small number of manually annotated samples as a seed dataset, and use a large model to infer the intent and instruction set representation corresponding to the sample data as the first version of the benchmark dataset;

[0027] Step 2: Use the GPT-4o large model to imitate the input query in the benchmark dataset to generate similar dialogues, and generate corresponding intents and instruction sets for the generated dialogue flow to obtain the dataset generated by the large model;

[0028] Step 3: Manually perform quality inspection on the data in step 2, fix the problems in the generated data set, and store the manually reviewed and proofread data in the database to obtain the second version of the data set;

[0029] Step 4: For the second version of the data set, use the GPT-4o large model to rewrite synonymous sentences, including rewriting the user input query and information in the conversation history to generate a candidate synonymous sentence data set.

[0030] Step 5: Manually perform quality inspection on the data in step 4, fix the problems in the generated data set, and store the manually reviewed and proofread data in the database to obtain the third version of the data set.

[0031] Step 6: Repeat the process from step 2 to step 5, and continue to expand the data set until the sample size of the data set reaches the preset expectation.

[0032] Furthermore, the supervised fine-tuning includes:

[0033] Based on the constructed dataset, a parameter-efficient fine-tuning method is used to fine-tune the large model.

[0034] Furthermore, the dialogue state update is a task of identifying dialogue state information in multiple rounds of ticket booking dialogue scenarios;

[0035] The intent recognition module and the dialogue management module in the large model reasoning task are combined into one reasoning module, which is used as a task for updating the dialogue state.

[0036] That is, the input of the dialogue state update task needs to include the user's current input, the dialogue state of the previous round, and the instructions completed by the system in the previous round. Then, its expression is:

[0037] Prompt words = character information + instruction description + intent set definition + dialogue state definition + context + guide words;

[0038] The API call is the core module for the interaction between the big model and the external ticket booking service. The big model can be used to call the third-party API in real time.

[0039] The system of multi-round ticket booking agent based on large model includes:

[0040] The task definition module obtains the definition of relevant elements of the multi-round ticket booking dialogue and uses it as the basis for building a multi-round ticket booking agent. The definition of relevant elements includes the definition of multi-round interaction intentions, the jump relationship of multi-round states, and the function definition of the ticket booking service API;

[0041] A large model fine-tuning module is used to train and fine-tune a large model, including task definition, data set construction, and supervised fine-tuning.

[0042] A prompt word project establishment module is used to establish a prompt word project and use it in combination with the fine-tuned large model. The establishment of the prompt word project includes dialogue status updates and API calls.

[0043] The technical effects and advantages of the method and system of the present invention based on a large model of multi-round ticket booking agent are as follows:

[0044] 1. Use the big model to complete the entire business process of multi-round ticket booking dialogues. The big model plays a key role in multi-round ticket booking dialogues (multi-round ticket booking business). It can not only generate coherent and natural multi-round reply scripts to meet user consultation and interaction needs, but also efficiently dispatch third-party APIs to achieve full process automation such as ticket inquiry, reservation, and payment. This integrated solution greatly improves ticket booking efficiency, optimizes user experience, and reduces labor costs, demonstrating the application potential of artificial intelligence in the field of customer service.

[0045] 2. A new paradigm of intent system is proposed through the intent recognition module, which can express the real needs of users with structured intents, and then optimize the intent pruning when using a large model for intent recognition reasoning; supervised fine-tuning can effectively reduce the difficulty of intent recognition in dialogue management tasks and optimize the system interaction quality; through the dialogue state update task and API scheduling in the multi-round ticket booking agent system, it can reply to each round of user dialogue and obtain real-time flight information. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a flow chart of the method of multi-round ticket booking agent based on a large model of the present invention;

[0047] Figure 2 It is a schematic diagram of the actual use process of the present invention;

[0048] Figure 3 It is a schematic diagram of the system of the present invention based on a large model and a multi-round ticket booking agent. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] Example 1

[0051] See also Figure 1 As shown, the method of multi-round ticket booking agent based on a large model described in this embodiment includes the following steps:

[0052] S1. Obtain the definition of relevant elements of the multi-round ticket booking dialogue (multi-round dialogue) and use it as the basis for building the multi-round ticket booking agent. The definition of relevant elements includes the definition of multi-round interaction intentions, the jump relationship of multi-round states, and the function definition of the ticket booking service API;

[0053] S2. Train and fine-tune the large model. The large model fine-tuning includes task definition, dataset construction, and supervised fine-tuning.

[0054] S3. Establish a prompt word project and use it in conjunction with the fine-tuned large model. Establishing the prompt word project includes conversation status updates and API calls.

[0055] Furthermore, the definition of multi-round interaction intention includes:

[0056] Obtain the business requirements for multi-round ticket booking conversations, and define the intent labels and related slot information for each round in the multi-round ticket booking system based on the business requirements;

[0057] The question entered by the user corresponds to an intent label, and the candidate slot information in the question entered by the user is assigned a value.

[0058] Furthermore, the jump relationship of multiple rounds of states includes:

[0059] Establishing a mapping relationship between each round of dialogue and the corresponding dialogue state in multiple rounds of ticket booking dialogues, wherein each round of dialogue state and the next round of dialogue state are represented by a mapping relationship;

[0060] Specifically, it refers to the mapping process from the state of the current dialogue round (including user input, system response and context information) to the state of the next dialogue round. This process describes how the dialogue system updates and predicts the state of the next dialogue round based on the user input and the current dialogue context, thereby achieving the continuity and logic of the dialogue, ensuring that the dialogue can flow reasonably according to the context state, avoiding the system from falling into a repetitive or unavailable state, and maintaining the smoothness of the user experience.

[0061] Furthermore, the function definition of the ticket booking service API includes:

[0062] Define the API input, output parameters and interaction methods to adapt them to the big model;

[0063] Specifically, for example, the system can adapt to the resource services of third-party APIs and provide large models for calling, including ticket query, ticket information retrieval and booking functions, to complete the complete multi-round ticket booking process.

[0064] Furthermore, the task definition includes:

[0065] Obtain and combine the business scenarios of multiple rounds of ticket booking conversations, decompose the multiple rounds of ticket booking conversations into structured data according to the business scenarios, and then obtain the definitions of intents, instruction sets, and large model reasoning tasks;

[0066] Among them, the intent is represented by structured Json data. For each intent, it needs to include the current dialogue state, vertical domain label, intent recognition label, input question, probability of intent recognition label, and slot information (or slot KV) related to the intent. The definition of the intent involved in multi-round ticket booking is shown in the following table:

[0067]

[0068]

[0069] The instruction set is a normalized representation of the system output format in the dialogue system. In the application scenario of multi-round ticket booking dialogue, it mainly involves the TTS voice reply content and the content of the ticket booking result query result form, so it can be represented by two instructions. Among them, the instruction set parameter set consists of the instruction type and the instruction header. The instruction type can include TTS instructions and form instructions; the instruction set involved in the system is defined as follows:

[0070]

[0071] The large model reasoning task is to include an intent recognition module that combines context information in a multi-round ticket booking dialogue (multi-round ticket booking agent task), and then output structured intent information through the intent recognition module;

[0072] The input of the intention recognition module includes: the user's current state input question, the intention state information of the previous round, and the instruction set information of the previous round. Its expression is:

[0073] f(query n , Intent n-1 , Skill n-1 )=Intent n

[0074] In the formula, query n Indicates the input question of the user's current status in the nth round of dialogue, Intent n-1 Indicates the intention status information of the n-1th round of dialogue, Skill n-1 Indicates the instruction set information of the n-1th round (previous round) of dialogue, f(query n , Intent n-1 , Skill n-1 ) represents the input of the intent recognition module, Intent n It indicates that the intention recognition module outputs the intention information of the nth round of dialogue.

[0075] Furthermore, the dataset construction includes the following processes:

[0076] Step 1: Use a small number of manually annotated samples as a seed dataset, and use a large model to infer the intent and instruction set representation corresponding to the sample data as the first version of the benchmark dataset;

[0077] Step 2: Use the GPT-4o large model to imitate the input query in the benchmark dataset to generate similar dialogues, and generate corresponding intents and instruction sets for the generated dialogue flow to obtain the dataset generated by the large model;

[0078] Step 3: Manually perform quality inspection on the data in step 2, fix the problems in the generated data set, and store the manually reviewed and proofread data in the database to obtain the second version of the data set;

[0079] Step 4: For the second version of the data set, use the GPT-4o large model to rewrite synonymous sentences, including rewriting the user input query and information in the conversation history to generate a candidate synonymous sentence data set.

[0080] Step 5: Manually perform quality inspection on the data in step 4, fix the problems in the generated data set, and store the manually reviewed and proofread data in the database to obtain the third version of the data set.

[0081] Step 6: Repeat the process from step 2 to step 5, and continue to expand the data set until the sample size of the data set reaches the preset expectation. The preset expectation can be set to 50,000 samples in the data set.

[0082] Furthermore, supervised fine-tuning includes:

[0083] Based on the constructed dataset, the large model (such as ChatGLM3) is fine-tuned using parameter-efficient fine-tuning methods (such as Parameter-Efficient Fine-Tuning, PEFT, Lora algorithm);

[0084] Specifically, the PEFT method is characterized by fixing most of the pre-trained parameters and only fine-tuning a small number or additional model parameters, which greatly reduces the computing and storage costs and training time. At the same time, the advanced PEFT technology can also be comparable to full fine-tuning in terms of performance. The fine-tuning process is mainly to improve the reasoning accuracy of large models in complex tasks such as intent recognition, so as to achieve the goal of industrializing the business.

[0085] Furthermore, the conversation state update is a task to identify the conversation state information in the multi-round ticket booking conversation scenario;

[0086] The intent recognition module and the dialogue management module in the large model reasoning task are combined into one reasoning module, which is used as a task for updating the dialogue state.

[0087] That is, the input of the dialogue state update task needs to include the user's current input, the dialogue state of the previous round, and the instructions completed by the system in the previous round. Then, its expression is:

[0088] Prompt words = character information + command description + intent set definition + dialogue state definition + context + guide words.

[0089]

[0090]

[0091]

[0092]

[0093] Specifically, the prompt word engineering of the present invention is also a key component of the multi-round ticket booking intelligent entity, which can directly affect whether the large model can understand user needs and complete the complete ticket booking business process; the prompt word engineering guides the large model to output unified generation and preparation of efficient instructions according to system requirements, so as to better intelligently collaborate with users to complete voice interactive ticket booking needs.

[0094] Furthermore, API calls are the core part of the interaction between the big model and external ticket booking services. The big model can be used to call third-party APIs in real time to obtain real-time flight information.

[0095] Specifically, in the present invention, the API involved mainly includes the flight information API of XX Airlines. The optimization of the prompt words of the API call is mainly to infer the corresponding API call parameter information from the result of the dialogue state update, and then complete the intelligent interaction under the current dialogue state through the ability of calling the API by the big model; the main API parameters involved in the present invention are as follows:

[0096] parameter Parameter Type Parameter Description Example leaveTime String Earliest departure time "2024-08-2915:00:00” arriveTime String Latest arrival time "2024-08-2920:00:00” fromCityId Int Departure City ID 12 toCityId Int Destination City Id 23 minPrice Float Lowest Price 500 maxPrice Float Highest Price 1000

[0097] Furthermore, after combining the fine-tuned large model with the prompt word engineering, the present invention also supports multiple deployment methods, including command line calls, GUI deployment, and API deployment, so as to build a multi-round ticket booking agent with voice interaction capabilities, providing users with highly intelligent ticket booking services in commercial scenarios.

[0098] It should be noted that the reference Figure 2 For example, if a user wants to book a flight from Hangzhou to Beijing through an intelligent ticket booking system, the user will have multiple rounds of booking dialogues with the system through natural language to complete the booking process. The process is as follows:

[0099] The task definition first defines the user's intent system and output structure. For example, the intent of "booking a flight ticket" and the related slot information, such as departure place, destination, date, etc., are defined. For example, when a user says, "Book a flight ticket to Beijing for me," the system recognizes that this is an intent of "booking a flight ticket" and extracts the slot information such as departure place, destination, etc.

[0100] Generate benchmark datasets using large models (such as GPT-4o), and then continuously expand and improve the datasets through manual review and modification. For example, the system generates multiple queries like "help me book a flight to Beijing" and annotates these queries with intent and instruction sets;

[0101] Use efficient fine-tuning methods (such as the PEFT method) to fine-tune the large model to improve the accuracy of intent recognition; for example, the fine-tuned model can more accurately recognize the intent of user input and generate corresponding instruction sets, such as TTS instructions and form instructions;

[0102] Update clear prompts through the dialogue state to guide the large model to understand the user's needs and update the dialogue state. For example, when the user says "depart from Hangzhou", the prompt guides the model to update the departure slot information and ask the user for the next slot information (such as date).

[0103] Through the guidance of key information, such as prompt words, including person information, instruction description, intent set definition, etc., this information helps the model understand its role and task. For example, the prompt words clearly tell the model that it is an airline ticket salesperson and needs to update slot information based on user input.

[0104] The output format is specified through prompt word engineering to ensure that the output is structured and easy-to-parse instructions; for example, the output dialogue status update results must follow a specific JSON format to facilitate system parsing and execution, so that the large model can call third-party APIs in real time to obtain real-time flight information.

[0105] In this embodiment, the accuracy of intent recognition is improved by focusing on the construction of the data set and the fine-tuning of the model through large model fine-tuning; the prompt word engineering focuses on carefully designed prompt words to guide the model to understand and respond to user needs and standardize the output format; supervised learning can improve the accuracy of the model in intent recognition tasks; the standardized output format and clear task guidance of the prompt word engineering can improve the interaction efficiency and experience between users and the system; by continuously expanding the data set and optimizing the prompt words, the present invention can adapt to more business scenarios and user needs.

[0106] Example 2

[0107] See also Figure 3 As shown, the system of multi-round ticket booking agent based on a large model described in this embodiment includes:

[0108] The task definition module obtains the definition of relevant elements of the multi-round ticket booking dialogue and uses it as the basis for building a multi-round ticket booking agent. The definition of relevant elements includes the definition of multi-round interaction intentions, the jump relationship of multi-round states, and the function definition of the ticket booking service API;

[0109] A large model fine-tuning module is used to train and fine-tune a large model, including task definition, data set construction, and supervised fine-tuning.

[0110] The prompt word project establishment module establishes the prompt word project and uses it in combination with the fine-tuned large model. The establishment of the prompt word project includes dialog state updates and API calls.

[0111] Specifically, in the present invention, the large model can be flexibly configured according to business needs and application performance, and is not limited to a certain model; for example, the large models that support localized deployment and supervised fine-tuning mainly include BaiChuan2, ChatGLM4 and Qwen; the invention mainly uses ChatGLM4-6B as the base large model for fine-tuning; if the business needs require calling the reasoning service of the third-party large model in the form of API (such as Qwen2-72B, etc.), the locally deployed reasoning service can be replaced with the corresponding third-party large model reasoning service API for application; in addition, in the fine-tuning process of the large model, the PEFT algorithm used is not limited to the LoRA algorithm, such as QLoRA, AdaLoRA, P-Tuning, Prefix-Tuning, Prompt-Tuning, etc. are all commonly used efficient fine-tuning algorithms, which can be comprehensively considered and selected according to the actual situation such as the fine-tuning data set, LLM model characteristics, and performance factors to achieve better results;

[0112] It should be noted that through the comprehensive optimization of the large model fine-tuning and prompt word engineering, the large model infers each process of the multi-round dialogue and gradually completes the complete business process of ticket booking; this innovation can be widely extended to other multi-round interaction scenarios, such as booking hotels and ordering takeout; at the same time, in the application field, this innovation can be well applied in various intelligent products, such as smart assistants, digital figures, etc. Users can interact with the system through voice interaction, providing a complete end-to-end voice ticket booking experience.

[0113] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0114] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only one, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0115] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

[0116] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A multi-round ticket booking agent method based on a large model, characterized in that: The following steps are involved: S1. Obtain the definition of relevant elements of the multi-round ticket booking dialogue and use it as the basis for building a multi-round ticket booking agent. The definition of relevant elements includes the definition of multi-round interaction intentions, the jump relationship of multi-round states, and the function definition of the ticket booking service API; S2. Training and fine-tuning of large models, wherein the large model fine-tuning includes task definition, dataset construction, and supervised fine-tuning; S3. Establish a prompt word project and use it in combination with the fine-tuned large model. Establishing the prompt word project includes updating the dialogue status and calling the API.

2. The method of multi-round ticket booking agent based on a large model according to claim 1 is characterized in that: The definition of the multi-round interaction intention includes: Obtain the business requirements for multi-round ticket booking conversations, and define the intent labels and related slot information for each round in the multi-round ticket booking system based on the business requirements; The question entered by the user corresponds to an intent label, and the candidate slot information in the question entered by the user is assigned a value.

3. The method of multi-round ticket booking agent based on a large model according to claim 1, characterized in that: The jump relationship of the multi-round state includes: A mapping relationship between each round of dialogue and the dialogue state in multiple rounds of ticket booking dialogue is established, wherein each round of dialogue state and the next round of dialogue state are represented by a mapping relationship.

4. The method of multi-round ticket booking agent based on a large model according to claim 1, characterized in that: The function definition of the ticket booking service API includes: Define the API input, output parameters and interaction methods to adapt the API input, output parameters and interaction methods to the large model.

5. The method of multi-round ticket booking agent based on a large model according to claim 1, characterized in that: The task definition includes: Obtain and combine the business scenarios of multiple rounds of ticket booking conversations, decompose the multiple rounds of ticket booking conversations into structured data according to the business scenarios, and then obtain the definitions of intents, instruction sets, and large model reasoning tasks; The intent is represented by structured Json data. For each intent, it needs to include the current dialogue state, vertical domain label, intent recognition label, input question, probability of intent recognition label and slot information related to the intent; The instruction set is a normalized representation of the system output format in the dialogue system; The large model reasoning task in multiple rounds of ticket booking dialogues includes an intent recognition module that combines context information, and then outputs structured intent information through the intent recognition module.

6. The method of multi-round ticket booking agent based on a large model according to claim 5 is characterized in that: The input of the intention recognition module includes: the user's current state input question, the intention state information of the previous round, and the instruction set information of the previous round, which is expressed as: f(query n ,Intent n-1 ,Skill n-1 )=Intent n In the formula, query n Indicates the input question of the user's current status in the nth round of dialogue, Intent n-1 Indicates the intention status information of the n-1th round of dialogue, Skill n-1 Indicates the instruction set information of the n-1th round of dialogue, f(query n , Intent n-1 , Skill n-1 ) represents the input of the intent recognition module, Intent n It indicates that the intention recognition module outputs the intention information of the nth round of dialogue.

7. The method of multi-round ticket booking agent based on a large model according to claim 6 is characterized in that: The dataset construction includes the following processes: Step 1: Use a small number of manually annotated samples as a seed dataset, and use a large model to infer the intent and instruction set representation corresponding to the sample data as the first version of the benchmark dataset; Step 2: Use the big model to imitate the input query in the benchmark dataset to generate similar dialogues, and generate corresponding intents and instruction sets for the generated dialogue flow to obtain the dataset generated by the big model; Step 3: Manually perform quality inspection on the data in step 2, fix the problems in the generated data set, and store the manually reviewed and proofread data in the database to obtain the second version of the data set; Step 4: For the second version of the dataset, use the large model to rewrite synonymous sentences, including rewriting the user input query and information in the conversation history to generate a candidate synonymous sentence dataset. Step 5: Manually perform quality inspection on the data in step 4, fix the problems in the generated data set, and store the manually reviewed and proofread data in the database to obtain the third version of the data set. Step 6: Repeat the process from step 2 to step 5, and continue to expand the data set until the sample size of the data set reaches the preset expectation.

8. The method of multi-round ticket booking agent based on a large model according to claim 7, characterized in that: The supervised fine-tuning includes: Based on the constructed dataset, a parameter-efficient fine-tuning method is used to fine-tune the large model.

9. The method of multi-round ticket booking agent based on a large model according to claim 8, characterized in that: The dialogue state update is a task to identify dialogue state information in a multi-round ticket booking dialogue scenario; The intent recognition module and the dialogue management module in the large model reasoning task are combined into one reasoning module, which is used as a task for updating the dialogue state. The API call is the core module for the interaction between the big model and the external ticket booking service. The big model can be used to call the third-party API in real time.

10. A system of multi-round ticket booking agents based on a large model, applied to a method of multi-round ticket booking agents based on a large model as claimed in any one of claims 1 to 9, characterized in that: include: The task definition module obtains the definition of relevant elements of the multi-round ticket booking dialogue and uses it as the basis for building a multi-round ticket booking agent. The definition of relevant elements includes the definition of multi-round interaction intentions, the jump relationship of multi-round states, and the function definition of the ticket booking service API; A large model fine-tuning module is used to train and fine-tune a large model, including task definition, data set construction, and supervised fine-tuning. The prompt word project establishment module establishes the prompt word project and uses it in combination with the fine-tuned large model. The establishment of the prompt word project includes dialog state updates and API calls.

Images