Intention order query interaction method and system, electronic equipment and storage medium

Acquisition and verification of customer information through generative artificial intelligence technology, and the use of large language models to determine the intended orders for insurance policy inquiry and answer questions, solving the problem that traditional systems cannot flexibly adapt to customer needs, and improving the accuracy and customer experience of insurance policy inquiry .

CN119088916APending Publication Date: 2024-12-06SUNSHINE DIGITAL INTELLIGENCE TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411090156.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-09
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The existing insurance policy query technology relies on traditional dialogue robot systems. Predefined response templates cannot flexibly adapt to customer specific needs, resulting in incomplete matching of answers and reducing customer experience.

Method used

Generative artificial intelligence technology is used to obtain the complete intention information and slot information of customers, perform identity verification, and use large language models to determine intention orders and answer customers' questions to improve the accuracy and naturalness of answers.

Benefits of technology

It significantly improves customer experience, provides more accurate answers by accurately obtaining intention and slot information, and improves customer satisfaction and service efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119088916A_ABST
    Figure CN119088916A_ABST
Patent Text Reader

Abstract

The invention discloses an intention order query interaction method and system, electronic equipment and a storage medium, and relates to the technical field of business interaction, and the method comprises the steps: obtaining the complete intention information and complete slot position information of a customer according to the input information of the customer through employing a generative artificial intelligence technology; obtaining identity information of the client according to the complete slot position information; verifying the identity information; after the client passes the identity verification, determining an intention order according to the input information of the client, the complete intention information and the complete slot position information by using a generative artificial intelligence technology, and providing the intention order to the client; and when the customer puts forward a question about the intentional order, determining an answer to the question by using a generative artificial intelligence technology, and providing the answer to the customer. According to the invention, intention orders and answers can be accurately determined, and the customer experience is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Background Art

[0002] At present, insurance policy inquiry technology mainly relies on the traditional conversational robot (chatbot) system, which responds to customers' specific inquiry needs through predefined response templates. However, the predefined response templates cannot flexibly adapt to customers' specific inquiry needs, resulting in the content of the answer may not fully match the questions raised by the customer, reducing the customer experience. Summary of the invention

[0003] The technical problem to be solved by the present invention is to address the deficiencies of the prior art and specifically provide a query interaction method, system, electronic device and storage medium for an intended order, which are as follows:

[0004] 1) In the first aspect, the present invention provides a query interaction method for an intended order, and the specific technical solution is as follows:

[0005] Based on the customer's input information and using generative artificial intelligence technology, obtain the customer's complete intent information and complete slot information;

[0006] Get the customer's identity information based on the complete slot information;

[0007] Verify identity information;

[0008] When the customer passes identity authentication, based on the customer’s input information, complete intent information, and complete slot information, and using generative AI technology, the intended order is determined and provided to the customer;

[0009] When a customer asks a question about an intended order, generative AI technology is used to determine the answer to the question and provide it to the customer.

[0010] The beneficial effects of the query interaction method for intended orders provided by the present invention are as follows:

[0011] On the one hand, based on generative artificial intelligence technology, it is possible to accurately obtain the customer's complete intention information and complete slot information, and accurately determine the intended order; on the other hand, when customers ask questions about intended orders, the use of generative artificial intelligence technology can determine more accurate answers, greatly improving the customer experience.

[0012] On the basis of the above scheme, the query interaction method of an intended order of the present invention can also be improved as follows.

[0013] Furthermore, based on the customer's input information and using generative artificial intelligence technology, the customer's complete intent information and complete slot information are obtained, including:

[0014] Collect the customer's current input information, and use generative artificial intelligence technology to obtain intent-related data and slot-related data from the current input information and analyze them to obtain information in a structured format, parse the structured format information to obtain current intent information and current slot information; determine whether the current intent information is complete to obtain a first judgment result, and determine whether the current slot information is complete to obtain a second judgment result; when the first judgment result and the second judgment result are both yes, determine the current intent information as complete intent information, and determine the current slot information as complete slot information; when at least one of the first judgment result and the second judgment result is no, determine the missing information, and continue to collect the customer's current input information until the first judgment result and the second judgment result are both yes.

[0015] Furthermore, the customer's input information includes voice information and / or text information.

[0016] Furthermore, the intention order is a target intention insurance policy.

[0017] 2) In the second aspect, the present invention also provides a query interaction system for intended orders, and the specific technical solution is as follows:

[0018] It includes an intention slot information acquisition module, an identity information acquisition module, an identity authentication module, an intention order determination module and an interaction module;

[0019] The intent slot information acquisition module is used to: obtain the customer's complete intent information and complete slot information based on the customer's input information and using generative artificial intelligence technology;

[0020] The identity information acquisition module is used to: obtain the customer's identity information based on the complete slot information;

[0021] The identity authentication module is used to: verify identity information;

[0022] The intended order determination module is used to: after the customer passes identity authentication, determine the intended order based on the customer's input information, complete intent information and complete slot information, and use generative artificial intelligence technology to provide it to the customer;

[0023] The interactive module is used to: When a customer asks a question about an intended order, generative artificial intelligence technology is used to determine the answer to the question and provide it to the customer.

[0024] On the basis of the above scheme, the query interaction system for intended orders of the present invention can also be improved as follows.

[0025] Furthermore, the intent slot information acquisition module is specifically used to:

[0026] Collect the customer's current input information, and use generative artificial intelligence technology to obtain intent-related data and slot-related data from the current input information and analyze them to obtain information in a structured format, parse the structured format information to obtain current intent information and current slot information; determine whether the current intent information is complete to obtain a first judgment result, and determine whether the current slot information is complete to obtain a second judgment result; when the first judgment result and the second judgment result are both yes, determine the current intent information as complete intent information, and determine the current slot information as complete slot information; when at least one of the first judgment result and the second judgment result is no, determine the missing information, and continue to collect the customer's current input information until the first judgment result and the second judgment result are both yes.

[0027] Furthermore, the customer's input information includes voice information and / or text information.

[0028] Furthermore, the intention order is a target intention insurance policy.

[0029] 3) In a third aspect, the present invention further provides an electronic device, comprising a processor, the processor being coupled to a memory, the memory storing at least one computer program, the at least one computer program being loaded and executed by the processor, so that the electronic device can implement any of the above-mentioned query interaction methods for intended orders.

[0030] 4) In a fourth aspect, the present invention further provides a computer-readable storage medium, in which at least one computer program is stored, and the at least one computer program is loaded and executed by a processor so that the computer implements any of the above-mentioned query interaction methods for intended orders.

[0031] It should be noted that the beneficial effects achieved by the technical solutions of the second to fourth aspects of the present invention and the corresponding possible implementation methods can be found in the above-mentioned technical effects of the first aspect and its corresponding possible implementation methods, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0033] Figure 1 A flowchart of a query interaction method for an intended order according to an embodiment of the present invention;

[0034] Figure 2 A schematic diagram of the structure of an inquiry and interaction system for an intended order according to an embodiment of the present invention;

[0035] Figure 3 The figure is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0036] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0037] like Figure 1 As shown, a query interaction method for an intended order according to an embodiment of the present invention includes the following steps:

[0038] S1. Based on the customer's input information and using generative artificial intelligence technology, obtain the customer's complete intention information and complete slot information;

[0039] The customer's input information includes voice information and / or text information.

[0040] Among them, the intention order is an intention insurance policy, and it can also be other orders such as orders to purchase daily necessities, telecommunications package orders, etc.

[0041] Among them, intent refers to: the intention actively proposed by the customer. For example, when communicating with the customer about the insurance policy, intent refers to: the customer's input information such as "want to check the policy", "want to renew the policy", and the sentence or voice indicating that he is currently busy.

[0042] Among them, the slot refers to: when questions about identity information are asked using pre-configured business scripts, the valid information in the customer's answers to these questions. For example, the question is: "What is your name?" The customer answers: "I am Zhang San." At this time, "Name" is the slot name and "Zhang San" is the slot value. It should be noted that the ID number and mobile phone number can also be the slot name, and the corresponding specific data is the slot value.

[0043] S2. Obtain the customer's identity information based on the complete slot information. The specific acquisition method is as follows:

[0044] 1) The first acquisition method: through the pre-defined slots, determine the corresponding slot value to obtain the customer's identity information, the customer's identity information includes: name, ID number, mobile phone number, gender and age, etc.

[0045] 2) The second acquisition method: Use generative artificial intelligence technology to analyze the complete slot information and determine the customer's identity information.

[0046] S3. Verify the identity information, which is specifically achieved through the following methods:

[0047] 1) The first verification method: Send the customer's identity information to a third party with identity authentication authority, and verify the identity information through the third party. Specifically:

[0048] First, verify the consistency of the identity information. If it is inconsistent, the verification fails. If it is consistent, continue to query whether the customer is a dishonest person based on the identity information. If so, the verification fails. If not, the verification passes, that is, the customer passes the identity authentication.

[0049] 2) The second verification method: verify identity information through some pre-set regulations, specifically:

[0050] First, verify the consistency of the identity information. Specifically, verify whether the name, ID number, mobile phone number, gender and age are consistent. If they are inconsistent, the verification fails. If they are consistent, continue to query whether the customer is a dishonest person based on the identity information. If so, the verification fails. If not, the verification passes, that is, the customer passes the identity verification.

[0051] S4. After the customer passes the identity verification, the intended order is determined based on the customer's input information, complete intent information and complete slot information, and generative artificial intelligence technology is used to provide the customer with the following information:

[0052] By utilizing the reference resolution capability of the large language model (LLM), the intended order can be accurately determined based on the customer's input information, complete intent information, and complete slot information, which can significantly improve accuracy. The large language model (LLM) can specifically adopt ChatGPT, etc.

[0053] S5. When a customer asks a question about an intended order, we use generative AI technology to determine the answer to the question and provide it to the customer, specifically:

[0054] For example, when the intended order is an intended insurance policy, the customer can freely ask questions about the intended insurance policy. The present invention uses the language generation capability of the large language model and combines RAG (retrieval enhanced generation) technology to comprehensively consider the customer's questions, the customer's input information, and the policy and product terms information of the intended insurance policy returned by the insurance company's policy query system to generate accurate answers that fit the customer's questions. This process not only improves the accuracy of the answer, but also greatly improves the naturalness of the answer and customer satisfaction.

[0055] Optionally, in S1, based on the customer's input information and using generative artificial intelligence technology, complete intention information and complete slot information of the customer are obtained, including:

[0056] S10. Collect the customer's current input information, and use generative artificial intelligence technology to obtain intent-related data and slot-related data from the current input information and analyze them to obtain information in a structured format, parse the structured format information to obtain current intent information and current slot information; determine whether the current intent information is complete to obtain a first judgment result, and determine whether the current slot information is complete to obtain a second judgment result; when the first judgment result and the second judgment result are both yes, determine the current intent information as complete intent information, and determine the current slot information as complete slot information; when at least one of the first judgment result and the second judgment result is no, determine the missing information, and continue to collect the customer's current input information until the first judgment result and the second judgment result are both yes.

[0057] The intent-related data and the slot-related data are obtained from the current input information and analyzed to obtain information in a structured format. The structured format information is parsed to obtain the current intent information and the current slot information. The specific implementation process is as follows:

[0058] Through the prompt project, construct the information extraction prompt, fill the current intent information and current slot information that need to be collected into the prompt, and require the large language model to input in a structured format such as JSON. After the large language model returns all, perform the structured parsing post-processing process to finally obtain the current intent information and slot information.

[0059] The process of determining whether the current intent information is complete is as follows:

[0060] An intention information template is set in advance, and the intention information template includes multiple data items such as "whether to renew the insurance", "which type of insurance is needed", etc. When the intention information template is obtained from the customer's current input information and all the data items that need to be filled in can be determined, the intention information before the judgment is complete, otherwise, it is judged to be incomplete.

[0061] It should be noted that there may be a scenario in the intent information template where it is sufficient to "fill in one data item in a group of data items". After filling in any one of the data items in the group, there is no need to fill in other data items in the group; moreover, there may be data items in the intent information template that must be filled in. Therefore, there is a situation where "there is no need to fill in all data items in the intent information template".

[0062] The process of determining whether the current slot information is complete is as follows:

[0063] The slot information is set in advance, and the intention information template includes multiple data items such as "name", "ID number", "mobile phone number", etc. When the slot information template is obtained from the customer's current input information and all the data items that need to be filled in can be determined, the slot information before the judgment is complete, otherwise, it is judged to be incomplete.

[0064] It should be noted that there may be a scenario in the slot information template where it is sufficient to "fill in one data item in a group of data items", and there may be data items in the slot information template that must be filled in. Therefore, there is a situation where "there is no need to fill in all data items in the slot information template". After filling in any one of the data items in the group, there is no need to fill in other data items in the group. Therefore, there is a situation where "there is no need to fill in all data items in the slot information template".

[0065] Optionally, in the above technical solution, a system that uses a large language model and a rule / ML model is used in combination. In this solution, a large language model can be used to handle complex natural language understanding and generation tasks, while rules or traditional ML models can be used to handle tasks that require extremely high accuracy, such as identity authentication, data verification, etc. This hybrid approach can fully utilize the advantages of each model to improve the performance and accuracy of the overall system.

[0066] In another embodiment, comprising:

[0067] 1) Customer identity verification:

[0068] Through conversations with customers, we collect information such as their ID numbers for identity verification to ensure that customers have the right to make insurance policy inquiries. Compared with traditional identity verification methods, this invention uses the excellent semantic understanding ability of large language models to accurately extract customer identity information by constructing multiple slot extraction prompts, while reducing misunderstandings and information extraction errors.

[0069] 2) Positioning of the intended insurance policy:

[0070] After the customer passes the identity verification, it is necessary to accurately locate the insurance policy that the customer wants to query, that is, the intended insurance policy (also called the target insurance policy). The present invention collects the customer's needs in detail through multiple rounds of dialogue, and uses the reference resolution ability of the large language model to accurately understand and locate the intended insurance policy. This step significantly improves the positioning accuracy and avoids multiple rounds and misunderstandings in traditional methods.

[0071] 3) Question and answer interaction:

[0072] After successfully locating the intended insurance policy, the customer can freely ask questions about the intended insurance policy. The present invention utilizes the language generation capability of the large language model, combined with RAG (retrieval enhanced generation) technology, to comprehensively consider the customer's questions, conversation records, and the policy and product terms information returned by the policy query system, and generate accurate and appropriate answers to the customer's questions. This process not only improves the accuracy of the answer, but also greatly improves the naturalness of the answer and customer satisfaction.

[0073] Alternatively, by building a knowledge graph for the domain, the entities and relationships in the graph can be used to understand the customer's query and provide accurate answers. The knowledge graph can help improve the understanding ability of the dialogue system, especially in handling complex queries and providing fact-based answers. This approach requires a lot of domain knowledge engineering work to build and maintain the knowledge graph.

[0074] In another embodiment, comprising:

[0075] 1) Extract intent information and slot information based on the customer's input information, specifically:

[0076] According to the business script, the current conversation information (voice information) is collected, and the information extraction prompt is constructed through the prompt project. The current intent information and current slot information to be collected are filled into the prompt, and the large language model is required to be input in a structured format such as JSON. After the large language model is fully returned, the structured parsing post-processing process is performed to finally obtain the current intent information and slot information.

[0077] 2) The conversation state is transferred according to the conversation state jump logic in the business script. For example, when obtaining the customer's name, the system will ask the customer's name, analyze the customer's reply, and try to extract the customer's name information. If the name is extracted, it will enter the next node. If not, other corresponding feedback will be given. Then it will be determined whether the conversation process can be advanced. According to the business script, if the collected customer information can trigger further conversation state transfer, the corresponding transfer will be performed. If the transfer cannot be made, the current conversation state will continue to be maintained, and the information required for the conversation state transfer will continue to be tried. If the customer asks a question, the customer's question will be answered accordingly.

[0078] 3) Based on the dialogue status, such as the currently collected customer information such as intent and slot, the selective use of the dialogue process configuration determines whether to use the business core system call tool, and obtains the policy and product terms information of the intended insurance policy based on the customer information and the customer's question. Specifically, the relevant knowledge is recalled from the knowledge base in the form of question-answer pairs based on the customer's question, including insurance company or insurance field knowledge. The returned policy information, product terms information, and knowledge recalled from the knowledge base are filled into the prompt template together with the customer's question, and the large language model is called to obtain the answer, which is returned to the customer.

[0079] In another embodiment, the collection slots: ① query insurance type, including values: auto insurance, medical life insurance, accident health insurance, universal insurance; ② policy tail number; ③ insurance name; ④ whether the insured is the person himself, the method includes:

[0080] S101. When the customer searches for an insurance policy, the insurance policy is searched according to the slot information provided by the customer, and the insurance type is specifically searched in the insurance type list;

[0081] S102, when the customer inquires about which type of insurance to inquire about, the insurance policy is inquired according to the slot, and then S103 is executed; when the customer inquires whether it is a certain type of insurance policy to be inquired about, then S103 is executed;

[0082] S103, asking for ID number, at this time, the customer provides the ID number or the ID number in the insurance policy;

[0083] S104, asking the customer whether he is the insured person, if so, asking which policy to query, when the customer provides the policy tail number, policy list, insurance name, policy list, insured or policy list, query the policy according to the slot, and specifically query whether the insurance type is in the non-motor insurance list, the query insurance type list or the life insurance list; if not, confirm again whether it is the policy, when the insurance is in the motor insurance list, ask the vehicle brand, if not in the motor insurance list, query whether the insurance type is in the non-motor insurance list, the query insurance type list or the life insurance list;

[0084] You can also check the policy (intention insurance policy) based on the slot and then respond to the questions raised by the customer.

[0085] S105, Policy Q&A, set up the Policy Q&A node: If the customer does not ask any questions, you need to ask the customer what he wants to inquire, otherwise answer the customer's questions.

[0086] When inquiring about premium calculation, inform that premium calculation cannot be performed and ask whether a salesperson has been arranged. If so, ask whether the contact number can be an incoming call number. If so, inform the customer that a salesperson has been arranged. Check whether the license plate number in the policy is the license plate number of the new car or the transferred car, and ask whether the license plate number should be re-entered. If so, confirm again whether the license plate number is correct. If so, modify the license plate number in the policy information, inform the customer that the re-entry has been successful, and ask if there are any other questions.

[0087] When asking whether the mobile phone number can be changed, ask whether to change the mobile phone number of the policyholder or the insured. After receiving the customer's answer, ask for the new mobile phone number, collect the new mobile phone number, modify the new mobile phone number in the policy information, inform the customer that the contact information has been changed, and ask if they need any help. If there are no other questions, the process ends.

[0088] The query interaction method of an intended order of the present invention aims to optimize the process of traditional insurance policy query service by using a large language model (LLM) such as ChatGPT, etc., and improve customer service efficiency and experience. The introduction of large model technology on the basis of the traditional intelligent dialogue system not only significantly improves the accuracy of information extraction and intent recognition, but also has a significant improvement in process flexibility and intelligence. Compared with the classic agent solution based on large models, the use of clear script process control guides the behavior of the agent to achieve 100% controllability of the process. By utilizing the powerful semantic understanding and information extraction capabilities of the large language model, the present invention significantly improves the accuracy of identity authentication and target policy positioning. With the natural language generation capability of the large language model, the present invention can generate more natural answers that fit the actual problems of customers, greatly improving the customer experience. By integrating the large language model and RAG technology, the present invention realizes fast and accurate question and answer generation, effectively shortens customer waiting time, and improves service efficiency. Combined with the knowledge base recall step, the present invention can provide more comprehensive and detailed insurance knowledge to meet the diverse information needs of customers.

[0089] The present invention is an information extraction and intent recognition method optimized by a big model in an insurance policy query scenario. Compared with traditional methods based on NER, text classification and regular matching, the present invention uses a big model for information extraction, which improves the accuracy and efficiency of customer identity verification and target policy positioning. The key is to use the natural language understanding ability of the big model to accurately identify the customer's intentions and required information in a structured format through customized prompt engineering. On the basis of traditional chatbot solutions, the present invention optimizes the management of dialogue state and the integration of business processes through a big model. This includes dynamic dialogue state management using a big model, and process jump logic based on customer intent and slot information, which effectively advances the dialogue process while reducing dialogue loops caused by misunderstanding customer intent.

[0090] In the above embodiments, although the steps are numbered S1, S2, etc., these are only specific embodiments given by the present invention. Those skilled in the art may adjust the execution order of S1, S2, etc. according to actual conditions, which is also within the protection scope of the present invention. It can be understood that in some embodiments, some or all of the above embodiments may be included.

[0091] like Figure 2 As shown, an intended order query interaction system 200 according to an embodiment of the present invention includes an intended slot information acquisition module 201, an identity information acquisition module 202, an identity authentication module 203, an intended order determination module 204 and an interaction module 205;

[0092] The intention slot information acquisition module 201 is used to: obtain the complete intention information and complete slot information of the customer based on the customer's input information and using generative artificial intelligence technology;

[0093] The identity information acquisition module 202 is used to: obtain the identity information of the customer according to the complete slot information;

[0094] The identity authentication module 203 is used to: verify identity information;

[0095] The intended order determination module 204 is used to: after the customer passes identity authentication, determine the intended order based on the customer's input information, complete intention information and complete slot information, and use generative artificial intelligence technology to provide the intended order to the customer;

[0096] The interaction module 205 is used to: when a customer asks a question about an intended order, determine the answer to the question using generative artificial intelligence technology and provide it to the customer.

[0097] Optionally, in the above technical solution, the intention slot information acquisition module 201 is specifically used for:

[0098] Collect the customer's current input information, and use generative artificial intelligence technology to obtain intent-related data and slot-related data from the current input information and analyze them to obtain information in a structured format, parse the structured format information to obtain current intent information and current slot information; determine whether the current intent information is complete to obtain a first judgment result, and determine whether the current slot information is complete to obtain a second judgment result; when the first judgment result and the second judgment result are both yes, determine the current intent information as complete intent information, and determine the current slot information as complete slot information; when at least one of the first judgment result and the second judgment result is no, determine the missing information, and continue to collect the customer's current input information until the first judgment result and the second judgment result are both yes.

[0099] Optionally, in the above technical solution, the customer's input information includes voice information and / or text information.

[0100] Optionally, in the above technical solution, the intended order is a target intended insurance policy.

[0101] It should be noted that the beneficial effects of the query interaction system 200 for intended orders provided in the above embodiment are the same as the beneficial effects of the query interaction method for intended orders provided in the above embodiment, which will not be described in detail here. In addition, when the system provided in the above embodiment realizes its functions, it only uses the division of the above functional modules as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the system can be divided into different functional modules according to actual conditions to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be described in detail here.

[0102] like Figure 3 As shown, an electronic device 300 of an embodiment of the present invention includes a processor 320, the processor 320 is coupled to a memory 310, and the memory 310 stores at least one computer program 330, and the at least one computer program 330 is loaded and executed by the processor 320, so that the electronic device 300 implements any of the above-mentioned query interaction methods for intended orders, specifically:

[0103] The electronic device 300 may have relatively large differences due to different configurations or performances, and may include one or more processors 320 (Central Processing Units, CPU) and one or more memories 310, wherein the one or more memories 310 store at least one computer program 330, and the at least one computer program 330 is loaded and executed by the one or more processors 320, so that the electronic device 300 implements any query interaction method for the intended order provided in the above embodiments. Of course, the electronic device 300 may also have components such as a wired or wireless network interface, a keyboard, and an input and output interface for input and output. The electronic device 300 may also include other components for realizing the functions of the device, which will not be described in detail here. The electronic device may specifically be a computer, etc.

[0104] A computer-readable storage medium according to an embodiment of the present invention stores at least one computer program, and the at least one computer program is loaded and executed by a processor so that a computer implements any of the above-mentioned query interaction methods for intended orders.

[0105] Optionally, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, and the like.

[0106] In an exemplary embodiment, a computer program product or a computer program is also provided, the computer program product or the computer program including computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device executes any of the above-mentioned query interaction methods for intended orders.

[0107] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects and represent the definition of a specific order or sequence. The order of use of similar objects can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than the order shown or described.

[0108] Those skilled in the art know that the present invention can be implemented as a system, method or computer program product. Therefore, the present invention can be specifically implemented in the following forms, namely: it can be complete hardware, it can be complete software (including firmware, resident software, microcode, etc.), or it can be a combination of hardware and software, which is generally referred to as "circuit", "module" or "system" herein. In addition, in some embodiments, the present invention can also be implemented in the form of a computer program product in one or more computer-readable media, and the computer-readable medium contains computer-readable program code.

[0109] Any combination of one or more computer-readable media can be used. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can 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 (a non-exhaustive list) include: 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 invention, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0110] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A query interaction method for intended orders, characterized in that: include: Based on the customer's input information and using generative artificial intelligence technology, obtain the customer's complete intent information and complete slot information; Obtaining the identity information of the customer according to the complete slot information; Verifying the identity information; When the customer passes identity authentication, the intended order is determined based on the customer's input information, the complete intent information and the complete slot information, and by using generative artificial intelligence technology, and is provided to the customer; When the customer asks a question about the intended order, the answer to the question is determined using generative artificial intelligence technology and provided to the customer.

2. The query interaction method for intended orders according to claim 1, characterized in that: Based on the customer's input information and using generative artificial intelligence technology, the customer's complete intent information and complete slot information are obtained, including: Collect the customer's current input information, and use generative artificial intelligence technology to obtain intent-related data and slot-related data from the current input information and analyze them to obtain information in a structured format, parse the structured format information to obtain current intent information and current slot information; judge whether the current intent information is complete to obtain a first judgment result, and judge whether the current slot information is complete to obtain a second judgment result; when the first judgment result and the second judgment result are both yes, determine the current intent information as the complete intent information, and determine the current slot information as the complete slot information; when at least one of the first judgment result and the second judgment result is no, determine the missing information, and continue to collect the customer's current input information until the first judgment result and the second judgment result are both yes.

3. A query interaction method for intended orders according to claim 1 or 2, characterized in that: The customer's input information includes voice information and / or text information.

4. The query interaction method for intended orders according to claim 1 or 2, characterized in that: The intention order is a target intention insurance policy.

5. A query and interaction system for intended orders, characterized in that: It includes an intention slot information acquisition module, an identity information acquisition module, an identity authentication module, an intention order determination module and an interaction module; The intention slot information acquisition module is used to: obtain the complete intention information and complete slot information of the customer based on the customer's input information and using generative artificial intelligence technology; The identity information acquisition module is used to: obtain the identity information of the customer according to the complete slot information; The identity verification module is used to: verify the identity information; The intended order determination module is used to: after the customer passes identity authentication, determine the intended order based on the customer's input information, the complete intention information and the complete slot information, and provide it to the customer by using generative artificial intelligence technology; The interactive module is used to: when the customer asks a question about the intended order, determine the answer to the question using generative artificial intelligence technology and provide it to the customer.

6. The query and interactive system for intended orders according to claim 5, characterized in that: The intention slot information acquisition module is specifically used for: Collect the customer's current input information, and use generative artificial intelligence technology to obtain intent-related data and slot-related data from the current input information and analyze them to obtain information in a structured format, parse the structured format information to obtain current intent information and current slot information; judge whether the current intent information is complete to obtain a first judgment result, and judge whether the current slot information is complete to obtain a second judgment result; when the first judgment result and the second judgment result are both yes, determine the current intent information as the complete intent information, and determine the current slot information as the complete slot information; when at least one of the first judgment result and the second judgment result is no, determine the missing information, and continue to collect the customer's current input information until the first judgment result and the second judgment result are both yes.

7. The query interactive system for intended orders according to claim 5 or 6, characterized in that: The customer's input information includes voice information and / or text information.

8. The query and interactive system for intended orders according to claim 5 or 6, characterized in that: The intention order is a target intention insurance policy.

9. An electronic device, characterized in that: The electronic device includes a processor, the processor is coupled to a memory, at least one computer program is stored in the memory, and the at least one computer program is loaded and executed by the processor so that the electronic device implements a query interaction method for intended orders as described in any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that: At least one computer program is stored in the computer-readable storage medium, and the at least one computer program is loaded and executed by the processor to enable the computer to implement a query interaction method for intended orders as described in any one of claims 1 to 4.