Intelligent file handling guiding method and device, electronic equipment and storage medium

Abstract

The invention provides an intelligent file handling guiding method and device, electronic equipment and a storage medium, relates to the technical field of self-service, is applied to a consultation file handling guiding robot interaction system with a built-in large language model LLM, and extracts a guiding word and a data analysis rule of a corresponding item from a business management platform according to an obtained item name; injecting the guide word into a cue word of a large language model LLM, performing closed domain dialogue with a user based on the large language model LLM, and guiding the user to provide necessary file handling information until the necessary file handling information is complete; and converting the necessary file handling information into a data format consistent with a business handling system according to the data analysis rule, automatically filling the data format into the business handling system, and displaying a file handling information filling result to a user. Therefore, through intelligent guidance of multiple rounds of conversations with the user and automatic data analysis and filling, the file handling efficiency is improved, and the user experience is enhanced.

Description

Technical Field

[0001] The present application relates to the technical field of self-service, and in particular, to an intelligent guidance method for handling cases, a device, an electronic device, and a storage medium. Background Art

[0002] In traditional government affairs services or other business handling scenarios, users often face problems such as complex business processes and unclear information filling requirements. The existing interaction methods between business handling systems and users are relatively single, mostly relying on manual consultation or simple online guidance, which are difficult to meet the needs of users to handle business quickly and accurately. Moreover, in the process of multi-round dialogue interaction, it is difficult to ensure the coherence of the dialogue logic and the effective utilization of information, and it is impossible to accurately guide users to provide the information required for business handling. In addition, it is difficult to integrate between business systems, and there are large differences in data formats and interface standards between different systems, resulting in the need to carry out large-scale transformation of existing business systems when introducing new intelligent services, with high costs and long implementation cycles. Summary of the Invention

[0003] In view of this, the purpose of the present application is to provide an intelligent guidance method for handling cases, a device, an electronic device, and a storage medium, which can improve the efficiency of business handling and the user experience.

[0004] In a first aspect, an embodiment of the present application provides an intelligent guidance method for handling cases, which is applied to a consultation and guidance robot interaction system with a built-in large language model LLM (Large Language Model). The method includes the following steps:

[0005] Extract the guiding words and data parsing rules for the corresponding matters from the business management platform according to the obtained matter names;

[0006] Inject the guiding words into the prompt words of the large language model LLM, and conduct a closed-domain dialogue with the user based on the large language model LLM to guide the user to provide the necessary case handling information until the necessary case handling information is complete;

[0007] Convert the necessary case handling information into a data format consistent with the business handling system according to the data parsing rules, automatically fill it into the business handling system, and display the case handling information filling result to the user.

[0008] In some embodiments, the guiding words are configured on the business management platform according to the information filling requirements of the business handling system for handling each matter, and are used to indicate the necessary case handling information for each matter.

[0009] In some embodiments, the conducting a closed-domain dialogue with the user based on the large language model LLM includes the following steps:

[0010] Obtain the user's historical conversation information from the cached data, and inject the historical conversation information into the prompt words of the large language model (LLM);

[0011] Based on the large language model (LLM), logically complete the sentences of the user's current conversation information, analyze the obtained completion result, summarize the known necessary document processing information, and store the completion result as historical conversation information in the cached data;

[0012] Compare the known necessary document processing information with the guiding words to determine whether the necessary information is complete. If it is complete, switch to the document processing interface of the business processing system; if it is not complete, based on the large language model (LLM), generate conversation content according to the guiding words to guide the user to provide the missing necessary document processing information.

[0013] In some embodiments, after generating the conversation content based on the large language model (LLM) according to the guiding words, the following steps are included:

[0014] Obtain the text data of the conversation content, convert the text data into audio, and generate the corresponding lip movements, body movements or facial expressions of the digital human according to the text data, and output them synchronously through audio and video.

[0015] In some embodiments, when the large language model (LLM) conducts a closed-domain conversation with the user, the following steps are further included:

[0016] Based on the large language model (LLM), conduct semantic analysis on the user's current conversation information in each round to identify whether the user requests to terminate the current document processing. If so, stop the document processing.

[0017] In some embodiments, the data parsing rules are configured on the business management platform according to the data format requirements of the business processing system for each item, and are used to guide the translation of the necessary document processing information provided by the user in natural language into the code representation recognizable by the business processing system.

[0018] In some embodiments, the method further includes the following steps:

[0019] In response to the instruction that the filled information is correct sent by the user, submit the document processing.

[0020] In a second aspect, an embodiment of the present application provides an intelligent guiding document processing device, which is applied to a consulting and guiding robot interaction system built with a large language model (LLM). The device includes:

[0021] An extraction module, configured to extract the corresponding guiding words and data parsing rules of the corresponding item from the business management platform according to the obtained item name;

[0022] A guidance module, configured to inject the guiding words into the prompt words of the large language model (LLM), and conduct a closed-domain conversation with the user based on the LLM to guide the user to provide necessary document processing information until the necessary document processing information is complete;

[0023] A document processing module, configured to convert the necessary document processing information into a data format consistent with the business processing system according to the data parsing rule, automatically fill it into the business processing system, and display the document processing information filling result to the user.

[0024] In a third aspect, an electronic device provided by an embodiment of the present application includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device runs, the processor communicates with the memory through the bus. When the machine-readable instructions are executed by the processor, the intelligent guidance document processing method according to any one of the above first aspects is executed.

[0025] In a fourth aspect, a computer-readable storage medium provided by an embodiment of the present application stores a computer program. When the computer program is run by a processor, the intelligent guidance document processing method according to any one of the above first aspects is executed.

[0026] An intelligent guidance document processing method, device, electronic device, and storage medium provided by the present application are applied to a consultation and guidance robot interaction system built with a large language model (LLM). The guiding words and data parsing rules of the corresponding matters are extracted from the business management platform according to the obtained matter names; the guiding words are injected into the prompt words of the LLM, and a closed-domain conversation is conducted with the user based on the LLM to guide the user to provide necessary document processing information until the necessary document processing information is complete; the necessary document processing information is converted into a data format consistent with the business processing system according to the data parsing rule, automatically filled into the business processing system, and the document processing information filling result is displayed to the user. Therefore, through the intelligent guidance of multiple rounds of conversations with the user, as well as automatic data parsing and filling, the document processing efficiency is improved and the user experience is enhanced. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the accompanying drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0028] Figure 1 Shows a flowchart of the intelligent guidance document processing method described in the embodiments of the present application;

[0029] Figure 2 The figure shows a schematic diagram of the interface for configuring guiding words and data parsing rules in an embodiment of the present application;

[0030] Figure 3 The figure shows a flowchart of a closed - domain conversation between the user and an embodiment of the present application based on a large - language model (LLM);

[0031] Figure 4 The figure shows a flowchart of an embodiment of the present application for generating conversation content based on a large - language model (LLM) according to the guiding words to guide the user to provide missing necessary document - handling information;

[0032] Figure 5 The figure shows a schematic diagram of the interface for a closed - domain conversation between an embodiment of the present application and the user;

[0033] Figure 6 The figure shows a schematic diagram of the structure of the intelligent guiding document - handling device according to an embodiment of the present application;

[0034] Figure 7 The figure shows a block diagram of the structure of the electronic device according to an embodiment of the present application. Detailed implementation manners

[0035] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. It should be understood that the accompanying drawings in the present application are only for the purposes of illustration and description, and are not used to limit the protection scope of the present application. Additionally, it should be understood that the schematic drawings are not drawn to actual scale. The flowcharts used in the present application illustrate the operations implemented according to some embodiments of the present application. It should be understood that the operations in the flowchart may not be implemented in sequence, and steps without logical context relationships may be reversed in order or implemented simultaneously. In addition, those skilled in the art can add one or more other operations to the flowchart or remove one or more operations from the flowchart under the guidance of the content of the present application.

[0036] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application described and illustrated in the accompanying drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application claimed, but merely represents the selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative efforts fall within the protection scope of the present application.

[0037] It should be noted that the term "including" will be used in the embodiments of the present application to indicate the existence of the features stated thereafter, but does not exclude the addition of other features.

[0038] In view of the technical problems proposed in the background art, the present application provides an intelligent guided case handling method, device, electronic device and storage medium, which can improve the efficiency of business handling and user experience.

[0039] See the attached Figure 1 In the intelligent guided case handling method provided by an embodiment of the present application, which is applied to a consultation and guidance robot interaction system built with a large language model LLM, the method includes the following steps:

[0040] S1. Extract the guiding words and data parsing rules of the corresponding matter from the business management platform according to the obtained matter name;

[0041] S2. Inject the guiding words into the prompt words of the large language model LLM, and conduct a closed-domain conversation with the user based on the large language model LLM to guide the user to provide necessary case handling information until the necessary case handling information is complete;

[0042] S3. Convert the necessary case handling information into a data format consistent with the business handling system according to the data parsing rule, automatically fill it into the business handling system, and display the case handling information filling result to the user.

[0043] Specifically, in step S1, the business management platform is a supporting management platform for the business handling system, and is used in this application for business administrator users to configure matters that can be handled, their corresponding guiding words, data parsing rules, etc.

[0044] Through these configurations, the rapid integration and adaptation of the consultation and guidance robot interaction system with various business systems are achieved. There is no need to conduct large-scale transformation on the existing business systems, reducing the cost and difficulty of system integration, and enabling the entire consultation and guidance robot human-computer interaction system to flexibly connect to different business scenarios. Among them, the guiding words are configured on the business management platform according to the information filling requirements of the business handling system for each matter handling, and are used to indicate the necessary document information for each matter. When handling a specific matter, the large language model LLM will ask questions to the user based on the guiding words, guiding the user to gradually provide various information required to complete the matter handling, so as to ensure that the information provided by the user meets the requirements of the business handling system and improve the accuracy and efficiency of business handling. The data parsing rules are also configured on the business management platform. After the large language model LLM collects the user's document information, according to the name of the current matter being handled, the corresponding matter data parsing rules are obtained from the business management platform. These rules are used to guide the translation of the user's document information provided in natural language form into a format (such as json) that the business handling system can recognize and process, and map it to the corresponding parameters of the business handling system, realizing the data format conversion and docking between the user information and the business handling system, and ensuring the smooth progress of the business handling process.

[0045] See the attached drawings of the specification Figure 2 In one embodiment, for example, for the matter of "vehicle annual inspection", the guiding words configured clearly list the necessary information required for document handling, namely "license plate number, license plate color", and give example statements to guide the user's reply. For example, when asking about the license plate color, a selectable color range is provided, and when asking about the license plate number, etc., with the aim of guiding the user to provide information that meets the business handling requirements. In the configured data parsing rules, the parsed KEY list "carCode, carColor" is shown, as well as the explanation for each KEY, "carCode: license plate number; carColor: license plate color". This indicates that after obtaining the information provided by the user, the information described by the user in natural language (such as the license plate number and color informed by the user) will be parsed and mapped into specific code identifiers (carCode, carColor) according to these rules, so that the business system can recognize and process it, which conforms to the definition in the document that the data parsing rules translate the user's document information into a format recognizable by the business system.

[0046] In addition, it should be noted that the matter name can be obtained by the consultation and guidance robot interaction system through analyzing the content input by the user (such as conversation information, selection operations, etc.). For example, by analyzing the intention behind the user's expression content with the help of the large language model LLM, inferring the business that the user wants to handle; or the user makes a selection among the business options provided by the consultation and guidance robot interaction system, and determines the matter name according to the selected option by the user.

[0047] See the attached instructions Figure 3 In step S2, the closed-domain conversation with the user based on the large language model LLM includes the following steps:

[0048] S201. Obtain the user's historical conversation information from the cached data and inject the historical conversation information into the prompt words of the large language model LLM;

[0049] S202. Based on the large language model LLM, perform sentence logic completion on the user's current conversation information, analyze the obtained completion result, summarize the known necessary document processing information, and store the completion result as historical conversation information in the cached data;

[0050] S203. Compare the known necessary document processing information with the guiding words to determine whether the necessary information is complete. If it is complete, switch to the document processing interface of the business processing system; if it is not complete, based on the large language model LLM, generate conversation content according to the guiding words to guide the user to provide the missing necessary document processing information.

[0051] Specifically, in steps S201 and S202, for each round of conversation content input by the user, first obtain the user's historical conversation information from the cached data and inject the historical conversation information into the prompt words of the large language model LLM. The large language model LLM integrates the user's current conversation information and historical conversation information, and updates the known necessary document processing information, so as to have a more comprehensive understanding of the user's needs and the information already provided in the follow-up, ensure that the conversation proceeds orderly around business processing, and enhance the coherence and effectiveness of the conversation.

[0052] In step S203, the large language model LLM collects the necessary information for the user to handle the case according to the guiding words regarding the requirements for filling in the current case handling information, checks the missing situation of the necessary case handling information according to the requirements, determines whether the case guiding process (closed-domain dialogue) is completed, and updates the guiding status flag value at the same time. For example, set the flag value of "is_all_known" to indicate that if all the necessary case handling information has been successfully collected, that is, there is no missing key information, then it can be determined that the case guiding process has been completed, and the flag value of "is_all_known" is set to "1"; on the contrary, as long as one or more necessary case handling information has not been obtained, it means that the guiding process needs to continue, and it is necessary to interact with the user further to obtain the missing information, and the flag value of "is_all_known" is set to "0". Among them, setting the guiding status can clearly know the stage of the case handling guiding process, whether it is collecting information, waiting for the user to supplement key information, or has completed the collection of all necessary information. Furthermore, the dialogue process can be dynamically adjusted in real time according to the user's input and the collection situation of the case handling information, realizing the automation and orderliness of business handling.

[0053] In one embodiment, refer to the attached drawings of the specification Figure 4 , which shows the process of guiding the user to handle the annual inspection of passenger vehicles: First, the large language model LLM recognizes that the user has the intention to handle the "annual inspection of passenger vehicles" case and prompts the user to reply "Handle immediately" for confirmation. And after receiving the user's reply "Handle immediately", indicating the willingness to handle the case, the guiding words (license plate number and vehicle color) for handling the "annual inspection of passenger vehicles" case are extracted from the business management platform. Then, the large language model LLM generates the dialogue content according to the guiding word requirements, and preferably asks the user a question for each information item each time. Here, the dialogue content generated according to the guiding word "license plate number" is "What is your license plate number?", and the user tells the license plate number is "E A12345". After the large language model LLM receives the user information, on the one hand, it puts the user's license plate number information into the cache data, and on the other hand, it processes it in combination with the information it needs to obtain (license plate number and vehicle color), outputs the flag value of "is_all_known" as "0", and a new question "What color is your license plate?", until all the necessary case handling information has been collected and the guiding process ends, outputting the flag value of "is_all_known" as "0", and at the same time generating an end prompt "Thank you, we are preparing to handle it for you".

[0054] Furthermore, in other embodiments, semantic analysis is performed on the current dialogue information of each round of the user based on the large language model LLM to identify whether the user requires the termination of the current case, and if so, the case is stopped. That is, the user is allowed to interrupt and terminate the case at any time, and the user's wishes are fully respected. During the process of handling business, the user may change his mind for various reasons, such as finding that the information is not fully prepared, or needing to leave for something temporarily, etc., so that the user can stop the case handling process in time without having to continue, providing users with a flexible and autonomous handling experience and enhancing the flexibility of human-computer interaction.

[0055] In addition, see the instructions for Figure 5 In this application, the consulting robot interaction system interacts with the user through a digital human. The digital human obtains the text data of the reply content from the large language model LLM, converts the text into audio through the speech generation module, and generates the corresponding lip shape, body movements and facial expressions according to the text. Through the synchronous output of audio and video, the anthropomorphic interaction between the digital human and the user is realized. At the same time, the front-end of the consulting robot is synchronized, and the conversation process is recorded and displayed to the user in the form of a dialog box.

[0056] In step S3, when the large language model LLM confirms that all the necessary information for the matter has been collected, it first obtains the pre-configured corresponding matter data parsing rules from the business management platform according to the name of the current matter, so as to structure and convert the information provided by the user. Then the front end of the consulting and guiding robot interactive system switches from the interactive interface to the matter declaration interface corresponding to the business processing system, and intelligently fills the parsed user information into the corresponding input box or field according to the form structure and field requirements of the business processing system. There is no need for users to fill in manually one by one. This intelligent reporting method not only greatly saves users' time and energy, but also reduces the error rate that may occur due to manual filling, and improves the accuracy and efficiency of business processing.

[0057] After completing the intelligent reporting, the front end of the consulting and guiding robot interactive system will display the filled-in application information to the user on the declaration interface, that is, data echo. Users can intuitively view the content automatically filled in by the system and verify whether the information is accurate. Data echo gives users the opportunity to finally confirm and correct the information, guarantees the user's right to know and control the application information, and enhances the user's sense of participation and trust in the business processing process. When the user confirms that the information in the data echo is correct on the front-end interface, he can click the submit button to formally submit the application to the business processing system. After receiving the user's application, the business processing system will review and process the application in accordance with the established business rules and procedures.

[0058] It can be seen that an intelligent guided case handling method provided by the present application uses a large language model (LLM) for intelligent guidance based on pre-configured guiding words, integrates the user's current conversation information and historical conversation information to ensure the coherence of multi-round conversations, enables users to receive accurate guidance and clear prompts, and quickly collects user case handling information; and translates the collected user case handling information into a format recognizable by the business handling system according to pre-configured data parsing rules for intelligent filling. This improves the efficiency of business handling and the user experience.

[0059] Based on the same inventive concept, an intelligent guided case handling device is also provided in an embodiment of the present application. Since the principle of problem-solving of the device in the embodiment of the present application is similar to that of the above-mentioned intelligent guided case handling method in the embodiment of the present application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be elaborated.

[0060] As shown in the attached Figure 6 description of the present application, an intelligent guided case handling device is also provided, which is applied to a consulting and guiding robot interaction system with a built-in large language model (LLM). The device includes:

[0061] An extraction module 601, configured to extract corresponding guiding words and data parsing rules of an item from a business management platform according to the obtained item name;

[0062] A guiding module 602, configured to inject the guiding words into the prompt words of the large language model (LLM), and conduct a closed-domain conversation with the user based on the large language model (LLM) to guide the user to provide necessary case handling information until the necessary case handling information is complete;

[0063] A case handling module 603, configured to convert the necessary case handling information into a data format consistent with the business handling system according to the data parsing rules, automatically fill it into the business handling system, and display the case handling information filling result to the user.

[0064] In some embodiments, the guiding words are configured on the business management platform according to the information filling requirements of the business handling system for handling each item, and are used to indicate the necessary case handling information for each item. The data parsing rules are configured on the business management platform according to the data format requirements of the business handling system for each item, and are used to guide the translation of the necessary case handling information provided by the user in natural language into a code representation recognizable by the business handling system.

[0065] In some embodiments, the guiding module 602 conducts a closed-domain conversation with the user based on a large language model (LLM), including: obtaining the user's historical conversation information from cached data and injecting the historical conversation information into the prompt words of the large language model (LLM); logically completing the sentences of the user's current conversation information based on the large language model (LLM), analyzing the obtained completion result, summarizing the known necessary document processing information, and storing the completion result as historical conversation information in the cached data; comparing the known necessary document processing information with the guiding words to determine whether the necessary information is complete. If it is complete, switch to the document processing interface of the business processing system; if it is not complete, generate conversation content based on the guiding words using the large language model (LLM) to guide the user to provide the missing necessary document processing information.

[0066] In some embodiments, after the guiding module 602 generates conversation content based on the guiding words using the large language model (LLM), it further includes: obtaining the text data of the conversation content, converting the text data into audio, and generating the corresponding lip movements, body movements, or facial expressions of the digital human based on the text data, and outputting them synchronously through audio and video.

[0067] In some embodiments, when the guiding module 602 conducts a closed-domain conversation with the user based on the large language model (LLM), it further includes: semantically analyzing the user's current conversation information for each round based on the large language model (LLM) to identify whether the user requests to terminate the current document processing. If so, stop the document processing.

[0068] In some embodiments, the device further includes a submission module for submitting the document processing in response to the instruction that the filled information sent by the user is correct.

[0069] An intelligent guiding document processing device provided by the present application is applied to a consultation and guiding robot interaction system with a built-in large language model (LLM). The extraction module extracts the guiding words and data parsing rules of the corresponding matters from the business management platform according to the obtained matter names; the guiding module injects the guiding words into the prompt words of the large language model (LLM) and conducts a closed-domain conversation with the user based on the large language model (LLM) to guide the user to provide the necessary document processing information until the necessary document processing information is complete; the document processing module converts the necessary document processing information into a data format consistent with the business processing system according to the data parsing rules, automatically fills it into the business processing system, and displays the document processing information filling result to the user. Thus, through the intelligent guidance of multiple rounds of conversations with the user, as well as automatic data parsing and filling, the document processing efficiency is improved and the user experience is enhanced.

[0070] Based on the same concept of the present invention, the specification appendix Figure 7As shown, the structure of an electronic device 700 provided by an embodiment of the present application. The electronic device 700 includes: at least one processor 701, at least one network interface 704 or other user interfaces 703, a memory 705, and at least one communication bus 702. The communication bus 702 is used to realize the connection and communication between these components. Optionally, the electronic device 700 includes a user interface 703, including a display (such as a touch screen, LCD, CRT, holographic imaging, or projector, etc.), a keyboard, or a pointing device (such as a mouse, trackball, touchpad, or touch screen, etc.).

[0071] The memory 705 may include a read-only memory and a random access memory, and provide instructions and data to the processor 701. A part of the memory 705 may also include a non-volatile random access memory (NVRAM).

[0072] In some embodiments, the memory 705 stores the following elements, protection modules, or data structures, or subsets thereof, or extended sets thereof:

[0073] An operating system 7051, including various system programs, used to implement various basic services and process hardware-based tasks;

[0074] An application program module 7052, including various application programs, such as a launcher, a media player, a browser, etc., used to implement various application services.

[0075] In the embodiment of the present application, by invoking the programs or instructions stored in the memory 705, the processor 701 is used to execute a method for intelligent guided document handling, which can improve the efficiency of document handling and the user experience.

[0076] The present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is run by a processor, it executes the method for intelligent guided document handling.

[0077] Specifically, the storage medium can be a general storage medium, such as a mobile disk, a hard disk, etc. When the computer program on the storage medium is run, it can execute the above-mentioned method for intelligent guided document handling.

[0078] In the embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For another example, 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 displayed or discussed coupling or direct coupling or communication connection between each other can be through some communication interfaces. The indirect coupling or communication connection of devices or units can be in electrical, mechanical or other forms.

[0079] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or they can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0080] In addition, each functional unit in the embodiments provided in this application can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0081] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of this application. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs and other various media that can store program codes.

[0082] Finally, it should be noted that the above embodiments are only specific implementation manners of the present application, used to illustrate the technical solutions of the present application, rather than limiting it. The protection scope of the present application is not limited thereto. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that any person skilled in the art within the technical scope disclosed by the present application can still modify the technical solutions recorded in the foregoing embodiments, or can easily think of changes, or perform equivalent replacements on some of the technical features; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application. All should be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. An intelligent guidance method for handling documents, characterized in that: Applied to a consulting and guidance robot interaction system with a built-in large language model LLM, the method comprises the following steps: Extracting the guide words and data parsing rules of the corresponding matters from the business management platform according to the acquired matter names; Injecting the guide words into the prompt words of the large language model LLM, and conducting a closed domain dialogue with the user based on the large language model LLM, guiding the user to provide necessary information until the necessary information is complete; According to the data analysis rules, the necessary handling information is converted into a data format consistent with the business processing system, and automatically filled into the business processing system, and the handling information filling result is displayed to the user.

2. The intelligent guidance method for handling documents according to claim 1, characterized in that: in, The guide words are configured on the business management platform according to the information filling requirements of the business processing system for each matter, and are used to indicate the necessary handling information of each matter.

3. The intelligent guidance processing method according to claim 2, characterized in that: The closed domain dialogue with the user based on the large language model LLM includes the following steps: Obtain the user's historical conversation information from the cache data, and inject the historical conversation information into the prompt words of the large language model LLM; Based on the large language model (LLM), the user's current conversation information is logically completed, and the completed results are analyzed to summarize the known necessary information, and the completed results are stored as historical conversation information in the cache data; The known necessary information is compared with the guide words to determine whether the necessary information is complete. If so, switch to the processing interface of the business processing system; if not, generate dialogue content based on the guide words based on the large language model LLM to guide the user to provide the missing necessary information.

4. The intelligent guidance processing method according to claim 3, characterized in that: After the large language model LLM is used to generate the dialogue content according to the guide words, the following steps are included: The text data of the conversation content is obtained, and the text data is converted into audio, and the lip shape, body movement or facial expression corresponding to the digital human is generated according to the text data, and outputted synchronously through audio and video.

5. The intelligent guidance method for handling documents according to claim 3, characterized in that: The closed domain dialogue with the user based on the large language model LLM also includes the following steps: Based on the large language model (LLM), semantic analysis is performed on the current conversation information of each round of users to identify whether the user requires the termination of the current case. If so, the case is stopped.

6. The intelligent guidance processing method according to claim 1, characterized in that: The data parsing rules are configured on the business management platform according to the data format requirements of the business processing system for each matter, and are used to guide the translation of necessary processing information provided by the user in natural language into code representations that can be recognized by the business processing system.

7. The intelligent guidance processing method according to claim 1, characterized in that: The method further comprises the following steps: Respond to the user's instruction that the filled-in information is correct and submit the document.

8. An intelligent document handling guidance device, characterized in that: Applicable to a consulting robot interaction system with a built-in large language model LLM, the device comprises: An extraction module, used to extract the guide words and data parsing rules of the corresponding items from the business management platform according to the acquired item names; A guiding module, used for injecting the guiding words into the prompt words of the large language model LLM, and conducting a closed domain dialogue with the user based on the large language model LLM, guiding the user to provide necessary information until the necessary information is complete; The processing module is used to convert the necessary processing information into a data format consistent with the business processing system according to the data analysis rules, automatically fill it into the business processing system, and display the processing information filling results to the user.

9. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the intelligent guidance processing method as described in any one of claims 1 to 7 is executed.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the intelligent guidance processing method according to any one of claims 1 to 7 is executed.

Images