Audit scheme generation method and device, product, equipment and storage medium

Abstract

The invention discloses an audit scheme generation method and device, a product, equipment and a storage medium, and relates to the technical field of artificial intelligence, and the audit scheme generation method comprises the steps: inputting user question information to an audit planning agent for analysis, and obtaining route planning information; calling at least one agent to process the user problem information through the audit planning agent according to the route planning information to obtain an initial audit scheme; and calling an auditing evaluation agent through the auditing planning agent to evaluate the initial auditing scheme to obtain a target auditing scheme. The auditing implementation scheme is generated through the large-model multi-agent, limitation of a fixed template can be avoided, and the auditing implementation scheme conforming to the scene-related problem can be given according to the scene and the problem set by the user. The audit scheme generation efficiency is improved, and meanwhile the audit scheme generation flexibility is guaranteed.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to audit solution generation methods, devices, products, equipment and storage media. Background Art

[0002] The audit implementation plan in audit work is a systematic tool or structure used to guide all aspects of the audit process, including planning, execution and follow-up. It helps auditors ensure the comprehensiveness, consistency and effectiveness of the audit. The audit implementation plan usually includes the following components: audit matters, which define the matters and directions of the audit; audit content, which defines the specific content of the audit matters; audit objectives, which determine the specific purpose of the audit matters and content; audit procedures, which are specific steps and methods for collecting and analyzing audit evidence. The methods for generating audit implementation plans can vary according to different needs and technical means. The following are several common methods: 1. Manual generation: auditors manually write audit plans, procedures and report templates based on audit objectives and standards. 2. Template generation: Use pre-designed audit templates to quickly generate audit implementation plans. 3. Rule-based generation: Define a series of generation rules and automatically generate audit implementation plans based on these rules. 4. Machine algorithm recommendation: Use machine learning algorithms and artificial intelligence technology to learn from historical audit data and recommend the generation of audit implementation plans. The existing methods for generating audit implementation plans have the following technical problems: the manual generation of audit implementation plans is inefficient, poor in quality and consistency; the audit template generation is not flexible enough and difficult to update; the rule set generation is complex and has poor adaptability; the machine algorithm recommends audit implementation plans with strong data dependence and poor interpretability. Therefore, how to solve the low efficiency of the existing audit plan generation method while ensuring the flexibility of audit plan generation has become an urgent problem to be solved. Summary of the invention

[0003] The main purpose of this application is to provide an audit plan generation method, device, product, equipment and storage medium, aiming to solve the technical problem of low efficiency of existing audit plan generation methods while ensuring the flexibility of audit plan generation.

[0004] To achieve the above purpose, the present application proposes an audit scheme generation method, the audit scheme generation method comprising: Input user question information to the audit planning agent for analysis to obtain routing planning information; According to the routing planning information, multiple agents are called through the audit planning agent to process the user problem information to obtain an initial audit plan; The audit planning agent calls the audit evaluation agent to evaluate the initial audit plan to obtain the target audit plan.

[0005] In one embodiment, the process of processing the user question information according to the routing planning information and calling at least one agent through the audit planning agent to obtain an initial audit solution includes: Calling the audit matter agent and the audit scheme agent through the audit planning agent according to the routing planning information; Input the user question information into the audit matter agent to obtain the target audit matter; Input the user question information and the target audit items into the audit program intelligent body to obtain the target audit purpose, target audit content and target audit procedure; An initial audit plan is determined based on the target audit matters, the target audit objectives, the target audit content and the target audit procedures.

[0006] In one embodiment, the step of inputting the user question information into the audit matter agent to obtain the target audit matter includes: Perform information retrieval on the user problem information according to the audit solution database to obtain historical audit items; The user problem information is processed according to the historical audit matters and through the audit matter intelligent agent to obtain the target audit matters.

[0007] In one embodiment, before the step of performing information retrieval based on the user question information and the audit solution database to obtain historical audit items, the step further includes: Extract information from historical audit data based on audit system information to obtain audit extraction information; Determine corresponding audit vector information and audit keywords according to the audit extraction information; Establish a keyword index and a vector index respectively according to the audit vector information and the audit keyword; An audit solution database is obtained according to the historical audit data, the keyword index and the vector index.

[0008] In one embodiment, the step of evaluating the initial audit plan by the audit planning agent calling the audit evaluation agent to obtain the target audit plan includes: The audit planning agent calls the audit evaluation agent to evaluate the initial audit plan to obtain an audit plan evaluation result; In response to the audit scheme evaluation result being a passed evaluation result, the initial audit scheme is determined to be a target audit scheme.

[0009] In one embodiment, after the step of evaluating the initial audit plan by the audit planning agent calling the audit evaluation agent to obtain the audit plan evaluation result, the step further includes: In response to the audit plan evaluation result being an evaluation failure result, the audit planning agent calls the audit modification agent to modify the initial audit plan to obtain an updated audit plan; Input the updated audit plan to the audit evaluation agent for plan evaluation to obtain an updated plan evaluation result; When the evaluation result of the update plan is a passing result, the update audit plan is determined to be a target audit plan.

[0010] In addition, to achieve the above-mentioned purpose, the present application also proposes an audit plan generation system, characterized in that the system includes: a problem input end, a plan generation end and a plan output end, the problem input end is used to receive the problem information input by the user; the plan generation end is used to generate a target audit plan based on the user's problem information and through the audit planning agent to call the audit matter agent, the audit plan agent, the audit evaluation agent and the audit modification agent; the plan output end is used to output the generated target audit plan.

[0011] In addition, to achieve the above purpose, the present application also proposes an audit scheme generating device, the audit scheme generating device comprising: The processing module is used to input user question information to the audit planning agent for analysis to obtain routing planning information; A generation module, configured to process the user problem information according to the routing planning information and by calling multiple agents through the audit planning agent to obtain an initial audit plan; The evaluation module is used to evaluate the initial audit plan through the audit planning agent calling the audit evaluation agent to obtain the target audit plan.

[0012] In addition, to achieve the above-mentioned purpose, the present application also proposes an audit scheme generating device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and the computer program is configured to implement the steps of the audit scheme generating method described above.

[0013] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the audit scheme generation method described above are implemented.

[0014] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the audit scheme generation method described above.

[0015] This application obtains routing planning information by inputting user problem information into the audit planning agent for analysis; based on the routing planning information, at least one agent is called through the audit planning agent to process the user problem information to obtain an initial audit plan; the audit planning agent calls the audit evaluation agent to evaluate the initial audit plan to obtain a target audit plan. By generating audit implementation plans through large-model multi-agents, it is not restricted by fixed templates and can provide audit implementation plans that match scenario-related issues based on the scenarios and problems set by the user. This improves the efficiency of audit plan generation while ensuring the flexibility of audit plan generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0018] Figure 1 A flowchart diagram of some embodiments of the audit scheme generation method of the present application is provided; Figure 2 A schematic diagram of the audit implementation plan generation process provided for some embodiments of the audit plan generation method of the present application; Figure 3 Another flowchart diagram provided for some embodiments of the audit scheme generation method of the present application; Figure 4 A brief flowchart of a method for generating an audit plan provided in some embodiments of the present application; Figure 5 This is a schematic diagram of the module structure of the audit scheme generating device according to the embodiment of the present application; Figure 6 This is a schematic diagram of the device structure of the hardware operating environment involved in the audit scheme generation method in the embodiment of the present application.

[0019] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0020] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0021] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0022] The main solution of the embodiment of the present application is: input user problem information to the audit planning agent for analysis to obtain routing planning information; process the user problem information according to the routing planning information and call at least one agent through the audit planning agent to obtain an initial audit plan; evaluate the initial audit plan through the audit planning agent calling the audit evaluation agent to obtain a target audit plan.

[0023] The audit implementation plan in audit work is a systematic tool or structure used to guide all aspects of the audit process, including planning, execution and follow-up. It helps auditors ensure the comprehensiveness, consistency and effectiveness of the audit. The audit implementation plan usually includes the following components: audit matters, which define the matters and directions of the audit; audit content, which defines the specific content of the audit matters; audit objectives, which determine the specific purpose of the audit matters and content; audit procedures, which are specific steps and methods for collecting and analyzing audit evidence. The methods for generating audit implementation plans can vary according to different needs and technical means. The following are several common methods: 1. Manual generation: auditors manually write audit plans, procedures and report templates based on audit objectives and standards. 2. Template generation: Use pre-designed audit templates to quickly generate audit implementation plans. 3. Rule-based generation: Define a series of generation rules and automatically generate audit implementation plans based on these rules. 4. Machine algorithm recommendation: Use machine learning algorithms and artificial intelligence technology to learn from historical audit data and recommend the generation of audit implementation plans. The existing methods for generating audit implementation plans have the following technical problems: the manual generation of audit implementation plans is inefficient, poor in quality and consistency; the audit template generation is not flexible enough and difficult to update; the rule set generation is complex and has poor adaptability; the machine algorithm recommends audit implementation plans with strong data dependence and poor interpretability. Therefore, how to solve the low efficiency of the existing audit plan generation method while ensuring the flexibility of audit plan generation has become an urgent problem to be solved.

[0024] This application obtains routing planning information by inputting user problem information into the audit planning agent for analysis; based on the routing planning information, at least one agent is called through the audit planning agent to process the user problem information to obtain an initial audit plan; the audit planning agent calls the audit evaluation agent to evaluate the initial audit plan to obtain a target audit plan. By generating audit implementation plans through large-model multi-agents, it is not restricted by fixed templates and can provide audit implementation plans that match scenario-related issues based on the scenarios and problems set by the user. This improves the efficiency of audit plan generation while ensuring the flexibility of audit plan generation.

[0025] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an audit plan generation device that can achieve the above functions. The following takes the audit plan generation as the execution subject as an example to illustrate this embodiment and the following embodiments.

[0026] Based on this, the present application embodiment provides a method for generating an audit scheme, referring to Figure 1 , Figure 1 A flowchart diagram of some embodiments of the audit plan generation method of the present application.

[0027] In this embodiment, the audit scheme generation method includes steps S10 to S30: Step S10, input user question information to the audit planning agent for analysis to obtain routing planning information; It should be noted that if Figure 2 As shown, in this embodiment, the user (auditor) inputs a specific description of the field and related issues. The issues will be planned by the audit implementation plan tree planning agent to generate specific planning ideas for the audit implementation plan, including the generation of audit matters, audit purposes, audit content, audit procedures, etc., and the post-evaluation of the audit implementation plan after the generation is completed, and the modification of the audit implementation plan. Then the routing call logic of the agent platform calls the corresponding agents respectively to gradually generate audit matters, audit purposes, audit contents, and audit procedures. After the generation is completed, the audit implementation plan post-evaluation agent is called again to give a list of modifications and analysis explanations. Finally, the audit implementation plan modification agent is called again to regenerate a new audit implementation plan based on the modification opinions of the previous step. The last two steps are iterated to finally complete the generation of the audit implementation plan. Among them, the logic of the specific agent in each step is to write prompts according to the generation logic of each module, and for the context of matters, purposes, contents, etc. generated in the previous step, the most relevant cases and data in the knowledge base are recalled as the context of the specific problem, and then the output of the corresponding agent module is completed.

[0028] It can be understood that user question information refers to the question information input by the user at the front end, and the user question information is a description of the generated audit implementation plan, including requirements, scenarios, generation angles and other information. The audit planning agent refers to the audit implementation plan planning intelligence, and the routing planning information refers to the order in which the audit implementation plan planning intelligence calls other agents.

[0029] In the specific implementation, the problem information input by the user at the front end is input into the audit implementation plan planning intelligence, and then the complex structure of the audit implementation plan is disassembled by the audit implementation plan planning intelligence based on the problem information input by the user at the front end, including the steps, times, content and routing planning of generating audit matters, generating audit objectives, outputting audit procedures, etc., that is, obtaining the order in which the audit implementation plan planning intelligence calls other intelligent bodies.

[0030] It should be noted that this embodiment can build multiple intelligent agents based on the basic capabilities of retrieval enhancement generation technology and large language models to form a collection of intelligent agents that cooperate with each other, including an audit implementation plan planning agent based on a tree structure and with process information, an audit matter agent, an audit content agent, an audit purpose agent, an audit procedure agent, an audit implementation plan modification agent, and an audit implementation plan evaluation agent.

[0031] Step S20, according to the routing planning information and by invoking at least one agent through the audit planning agent to process the user question information, to obtain an initial audit plan; It is to be understood that the initial audit plan refers to the initial audit implementation plan that has not been evaluated by the audit implementation plan evaluation agent.

[0032] In the specific implementation, the audit plan planning agent also bears the function of agent calling. Its calling function will list the capabilities and inputs of other writing agents as capability list parameters, and select appropriate agents for calling in turn according to the above output structure settings and process settings. Then, the audit implementation plan planning agent calls the audit matter agent, audit purpose agent, audit content agent and audit procedure agent respectively to generate the corresponding audit matter, audit purpose, audit content and audit procedure module content, and finally obtains the initial audit implementation plan that has not passed the evaluation of the audit implementation plan evaluation agent, that is, the initial audit plan.

[0033] Step S30, the audit planning agent calls the audit evaluation agent to evaluate the initial audit plan to obtain a target audit plan.

[0034] It can be understood that the target audit plan refers to the audit implementation plan that is evaluated and passed by the audit implementation plan evaluation agent.

[0035] In the specific implementation, the agent evaluation rules are set in the audit implementation plan evaluation agent, and the audit implementation plan evaluation agent evaluates the rules of the initial audit implementation plan one by one and outputs the evaluation results. The audit modification agent modifies the audit implementation plan according to the evaluation results and modification opinions, and then outputs the new audit implementation plan. This process is iterated until the audit implementation plan evaluation agent outputs that the audit implementation plan is evaluated as qualified, and finally the audit implementation plan that has passed the evaluation of the audit implementation plan evaluation agent is obtained, that is, the target audit plan.

[0036] In a feasible implementation, step S30 may include steps A31-A32: Step A31, the audit planning agent calls the audit evaluation agent to evaluate the initial audit plan to obtain the audit plan evaluation result; It should be noted that the audit plan evaluation results refer to the results of the audit implementation plan evaluation agent evaluating the initial audit plan, including the evaluation pass results and evaluation fail results.

[0037] In the specific implementation, the initial audit plan is input into the audit implementation plan evaluation intelligent agent for plan evaluation, that is, according to the agent evaluation rules set in the audit implementation plan evaluation intelligent agent, the audit implementation plan evaluation intelligent agent evaluates the rules of the initial audit implementation plan one by one and outputs the evaluation results.

[0038] In a feasible implementation, step A31 may include steps B311 to B313: Step B311, when the audit plan evaluation result is an evaluation failure result, the audit planning agent calls the audit modification agent to modify the initial audit plan to obtain an updated audit plan; It is understood that the updated audit plan refers to the updated audit implementation plan.

[0039] In the specific implementation, when the audit plan evaluation result is an evaluation failure, it indicates that the initial audit plan does not meet the agent evaluation rules set in the audit implementation plan evaluation agent, and then the audit plan evaluation result and the initial audit plan are input into the audit plan modification agent for plan modification, and finally the updated audit implementation plan, that is, the updated audit plan, is obtained.

[0040] Step B312, input the updated audit plan to the audit evaluation agent for plan evaluation, and obtain the updated plan evaluation result; It can be understood that the updated plan evaluation result refers to the result of the audit implementation plan evaluation agent evaluating the updated audit plan, including the evaluation pass result and the evaluation fail result.

[0041] In the specific implementation, the updated audit implementation plan is input into the audit implementation plan evaluation agent, and the audit implementation plan is evaluated again to obtain the result of the audit implementation plan evaluation agent evaluating the updated audit plan, that is, the updated plan evaluation result.

[0042] Step B313, in response to the updated audit scheme evaluation result being a passed evaluation result, determining the updated audit scheme as a target audit scheme.

[0043] In the specific implementation, when the evaluation result of the updated plan is a passing result, it indicates that the audit implementation plan evaluation agent evaluates that the updated audit plan satisfies the evaluation rules, and then determines that the updated audit plan is the final audit implementation plan; when the audit implementation plan evaluation agent fails to evaluate the updated audit plan, iterative calculations are performed through the audit implementation plan evaluation agent and the audit implementation plan modification agent to obtain the final audit plan that meets the evaluation rules.

[0044] Step A32, in response to the audit scheme evaluation result being a passing result, determining the initial audit scheme as a target audit scheme.

[0045] In the specific implementation, when the audit plan evaluation result is a passed result, it indicates that the audit implementation plan evaluation agent evaluates that the initial audit plan meets the evaluation rules, and then determines that the initial audit plan is the final audit implementation plan, and outputs the target audit plan to the front end.

[0046] This embodiment obtains routing planning information by inputting user problem information into the audit planning agent for analysis; based on the routing planning information, multiple agents are called by the audit planning agent to process the user problem information to obtain an initial audit plan; the audit planning agent calls the audit evaluation agent to evaluate the initial audit plan to obtain a target audit plan. The generation of audit implementation plans by large-model multi-agents is not restricted by fixed templates, and can provide audit implementation plans that match scenario-related issues based on the scenarios and issues set by the user. This improves the efficiency of audit plan generation while ensuring the flexibility of audit plan generation.

[0047] In some embodiments of the present application, the same or similar contents as those in the above embodiments can be referred to the above description and will not be described in detail later. Figure 3 , step S20 in the audit scheme generation method further includes steps S21 to S24: Step S21, calling the audit matter agent and the audit scheme agent through the audit planning agent according to the routing planning information; It can be understood that the audit matter agent is used to generate audit matters corresponding to the audit implementation plan, and the audit plan agent includes but is not limited to the audit purpose agent, the audit content agent and the audit procedure agent, wherein the audit purpose agent is used to generate the audit purpose corresponding to the audit implementation plan, the audit content agent is used to generate the audit content corresponding to the audit implementation plan, and the audit procedure agent is used to generate the audit procedure corresponding to the audit implementation plan.

[0048] In the specific implementation, the audit plan planning agent also bears the function of agent calling. Its calling function will list the capabilities and inputs of other writing agents as capability list parameters, and select appropriate agents for calling in turn according to the above output structure settings and process settings. That is, the audit implementation plan planning agent calls the audit matter agent, audit purpose agent, audit content agent and audit procedure agent respectively.

[0049] Step S22, inputting the user question information into the audit matter agent to obtain the target audit matter; It can be understood that the target audit matters refer to the audit matters corresponding to the audit implementation plan.

[0050] In a specific implementation, the problem information input by the user at the front end is input into the audit matter intelligent body, and then the audit matter intelligent body generates corresponding audit matters, namely, target audit matters, according to the problem information input by the user at the front end.

[0051] In a feasible implementation, step S22 may include steps A221-A222: Step A221, performing information retrieval on the user problem information according to the audit solution database to obtain historical audit items; It can be understood that the audit plan database refers to the enterprise's local audit implementation plan knowledge base, and historical audit matters refer to the audit matters in the enterprise's local audit implementation plan knowledge base.

[0052] In the specific implementation, based on the problem information entered by the user at the front end, combined with the enterprise's local audit implementation plan knowledge base, and using mixed indexing methods such as vector recall and keyword recall, information retrieval is performed to obtain the corresponding audit items in the enterprise's local audit implementation plan knowledge base, that is, historical audit items.

[0053] In a feasible implementation, step A221 may include steps B2211 to B2214: Step B2211, extracting information from historical audit data according to the audit system information to obtain audit extraction information; It can be understood that audit system information refers to the collection of all policies, procedures, standards, regulations and guidelines related to audit activities, historical audit data includes structured and unstructured data such as historical audit implementation plan information and historical audit project information, and audit extraction information refers to information extracted based on audit system information and historical audit data.

[0054] In the specific implementation, information extraction is performed based on structured and unstructured data such as the collection of all policies, procedures, standards, regulations and guidelines related to audit activities, historical audit implementation plan information, historical audit project information, etc. to obtain audit extraction information.

[0055] Step B2212, determining corresponding audit vector information and audit keywords according to the audit extraction information; It can be understood that audit vector information refers to vectorized information, and audit keywords refer to keyword information.

[0056] In a specific implementation, vectorization and keyword extraction are performed based on information extracted based on audit system information and historical audit data to obtain vectorized information and keyword information, namely, audit vector information and audit keywords.

[0057] Step B2213, establishing a keyword index and a vector index respectively according to the audit vector information and the audit keyword; It can be understood that, according to the vectorization information and the keyword information, a keyword index and a vector index are respectively established to obtain a keyword index and a vector index.

[0058] Step B2214, obtaining an audit solution database based on the historical audit data, the keyword index and the vector index.

[0059] In the specific implementation, structured and unstructured data such as historical audit implementation plan information, historical audit project information, and corresponding keyword indexes and vector indexes are summarized to obtain the enterprise's local audit implementation plan knowledge base, namely the audit plan database.

[0060] It should be noted that before writing the audit implementation plan, the historical audit data needs to be processed and indexed into the database. The specific method is to first extract information from structured and unstructured data such as audit system information, historical audit implementation plan information, and historical audit project information, store them as information fragments, vectorize and extract keywords, and establish keyword indexes and vector indexes according to keyword and vector information, and finally obtain the enterprise's local audit implementation plan knowledge base.

[0061] It should be noted that in this embodiment, before each audit agent generates questions, it must perform information retrieval, recall the most relevant data for audit-related data and information stored in the local database, and sort and select the most relevant knowledge fragments from the information. This process generally includes mixed indexes such as vector recall and keyword recall, and the sorting logic is a similarity model trained with semantically related text pairs.

[0062] Step A222, based on the historical audit matters and through the audit matter agent, the user problem information is processed to obtain the target audit matters.

[0063] It can be understood that the retrieved historical audit matters are input into the audit matter intelligent body, and combined with the problem information input by the user at the front end, the audit matters corresponding to the audit implementation plan, i.e., the target audit matters, are generated through the audit matter intelligent body.

[0064] It should be noted that the audit matters corresponding to the audit implementation plan are generated through the audit matter agent, and the audit implementation plan direction is input into the audit matter agent. Based on the local audit matter library of the enterprise's local audit implementation plan knowledge base, the relevant historical audit matters are queried and recalled to generate multiple audit matters. For each audit matter, the audit content, audit purpose and audit procedure agents are called. The audit content agent is similar. Based on the enterprise's local audit implementation plan knowledge base, the audit content related information is queried and recalled to generate the audit content.

[0065] Step S23, input the user question information and the target audit items into the audit program agent to obtain the target audit purpose, target audit content and target audit procedure; It can be understood that the target audit objective refers to the audit objective corresponding to the audit implementation plan, the target audit content refers to the audit content corresponding to the audit implementation plan, and the target audit procedure refers to the audit procedure corresponding to the audit implementation plan.

[0066] In the specific implementation, the problem information input by the user at the front end and the audit items corresponding to the audit implementation plan are input into the audit purpose intelligent body, and then the corresponding audit purpose, i.e., the target audit purpose, is generated through the audit purpose intelligent body according to the problem information input by the user at the front end and the audit items corresponding to the audit implementation plan; the problem information input by the user at the front end, the audit items and the audit purpose corresponding to the audit implementation plan are input into the audit content intelligent body, and then the corresponding audit content, i.e., the target audit content, is generated through the audit content intelligent body according to the problem information input by the user at the front end, the audit items and the audit purpose corresponding to the audit implementation plan; the problem information input by the user at the front end, the audit items, the audit purpose and the audit content corresponding to the audit implementation plan are input into the audit procedure intelligent body, and then the corresponding audit procedure, i.e., the target audit procedure, is generated through the audit content intelligent body according to the problem information input by the user at the front end, the audit items, the audit purpose and the audit content corresponding to the audit implementation plan.

[0067] It should be noted that the audit purpose agent follows a similar process to generate the audit purpose. Finally, the audit procedure agent generates the audit procedure. The audit procedure agent needs to query the historical audit procedure library, obtain the audit procedure cases related to the issues in the historical audit projects of the enterprise from the enterprise's local audit implementation plan knowledge base, and output the audit procedures corresponding to the audit matters, audit contents, and audit purposes. The audit procedure generates detailed and executable audit steps based on historical samples.

[0068] Step S24, determining an initial audit plan according to the target audit items, the target audit objectives, the target audit contents and the target audit procedures.

[0069] In the specific implementation, the audit matters, audit objectives, audit contents and audit procedures corresponding to the audit implementation plan are summarized to obtain the audit implementation plan, that is, the initial audit plan.

[0070] It should be noted that the steps for setting up audit implementation in this embodiment are as follows: Name of Audit Implementation Plan: - Audit matter 1: Name of audit matter; - Audit objective 1: Audit objective; - Audit content 1: Audit content; - Audit Procedure 1: -- Audit Step 1: Description of Audit Steps; -- Audit Step 2: Description of Audit Steps; -- Audit Step 3: Description of Audit Steps; - Audit matter 2: Name of audit matter; - Audit objective 2: Audit objective; - Audit content 2: Audit content; - Audit Procedure 2: -- Audit Step 1: Description of Audit Steps; -- Audit Step 2: Description of Audit Steps; Through the tree structure set up above, the big model can clearly define the tree structure and generation process of the audit implementation plan and their relationships.

[0071] It should be noted that the specific calling process example for setting the audit implementation plan in this embodiment is as follows: Query is the user input question, which generates multiple audit items and their audit plans in a loop: Enter query to generate the audit item name; Enter the query and audit item name to generate the audit purpose; Enter the query, audit item name, and audit purpose to generate the audit content; Enter query, audit item name, audit purpose, and audit content to generate audit procedures in a loop: Generate audit step 1; Generate audit step 2; Generate audit step 3; Cycle to evaluate whether the audit implementation plan outputted above meets the requirements: the evaluation agent inputs the generated plan and gives a list of questions. If there are no questions, it ends directly. If it does not meet the requirements, regenerate the audit implementation plan according to the list of questions outputted by the evaluation agent. The above is the generation process of the planning agent settings.

[0072] This embodiment calls the audit matter agent and the audit scheme agent through the audit planning agent according to the routing planning information; inputs the user question information into the audit matter agent to obtain the target audit matter; inputs the user question information and the target audit matter into the audit scheme agent to obtain the target audit purpose, target audit content and target audit procedure; determines the initial audit scheme according to the target audit matter, the target audit purpose, the target audit content and the target audit procedure. By using a large model multi-agent design including an audit implementation plan tree planning agent, an audit matter agent, an audit content agent, an audit purpose agent, an audit procedure agent, an audit implementation plan modification agent and an audit implementation plan evaluation agent, the generation of a complete audit implementation plan is finally completed through the collaborative cooperation of multiple agents. The efficiency of audit plan generation is improved.

[0073] For example, to help understand the implementation process of the audit scheme generation method obtained by combining this embodiment with the above-mentioned embodiment 1, please refer to Figure 4 , Figure 4 A brief flow chart of an audit plan generation method is provided. Specifically, the system has a front-end page for inputting user questions. User questions are descriptions of the generated audit implementation plan, including requirements, scenarios, generation angles and other information, so as to generate personalized audit implementation plans that meet user needs. The system uses a large model multi-agent design including an audit implementation plan tree planning agent, an audit matter agent, an audit content agent, an audit purpose agent, an audit procedure agent, an audit implementation plan modification agent and an audit implementation plan evaluation agent. Through the collaborative cooperation of multiple agents, the generation of a complete audit implementation plan is finally completed, and the generated audit implementation plan is also output to the front-end page.

[0074] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the audit plan generation method of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0075] This application also provides an audit scheme generation device, please refer to Figure 5 , the audit scheme generating device comprises: Processing module 10, used to input user question information to the audit planning agent for analysis to obtain routing planning information; A generating module 20, configured to process the user question information according to the routing planning information and by invoking at least one more agent through the audit planning agent, to obtain an initial audit plan; The evaluation module 30 is used to evaluate the initial audit plan through the audit planning agent calling the audit evaluation agent to obtain a target audit plan.

[0076] Optionally, the generating module 20 is further used for: Calling the audit matter agent and the audit scheme agent through the audit planning agent according to the routing planning information; Input the user question information into the audit matter agent to obtain the target audit matter; Input the user question information and the target audit items into the audit program intelligent body to obtain the target audit purpose, target audit content and target audit procedure; An initial audit plan is determined based on the target audit matters, the target audit objectives, the target audit content and the target audit procedures.

[0077] Optionally, the generating module 20 is further used for: Perform information retrieval on the user problem information according to the audit solution database to obtain historical audit items; The user problem information is processed according to the historical audit matters and through the audit matter intelligent agent to obtain the target audit matters.

[0078] Optionally, the generating module 20 is further used for: Extract information from historical audit data based on audit system information to obtain audit extraction information; Determine corresponding audit vector information and audit keywords according to the audit extraction information; Establish a keyword index and a vector index respectively according to the audit vector information and the audit keyword; An audit solution database is obtained according to the historical audit data, the keyword index and the vector index.

[0079] Optionally, the evaluation module 30 is further configured to: The audit planning agent calls the audit evaluation agent to evaluate the initial audit plan to obtain an audit plan evaluation result; In response to the audit scheme evaluation result being a passed evaluation result, the initial audit scheme is determined to be a target audit scheme.

[0080] Optionally, the evaluation module 30 is further configured to: When the audit plan evaluation result is an evaluation failure result, the audit planning agent calls the audit modification agent to modify the initial audit plan to obtain an updated audit plan; Input the updated audit plan to the audit evaluation agent for plan evaluation to obtain an updated plan evaluation result; In response to the updated audit scheme evaluation result being a passed evaluation result, the updated audit scheme is determined to be a target audit scheme.

[0081] The audit scheme generation device provided by the present application adopts the audit scheme generation method in the above-mentioned embodiment, which can solve the technical problem that the existing audit scheme generation method is inefficient while ensuring the flexibility of audit scheme generation. Compared with the prior art, the beneficial effects of the audit scheme generation device provided by the present application are the same as the beneficial effects of the audit scheme generation method provided by the above-mentioned embodiment, and the other technical features in the audit scheme generation device are the same as the features disclosed in the above-mentioned embodiment method, which will not be repeated here.

[0082] The present application provides an audit scheme generating device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the audit scheme generating method in the above-mentioned embodiment one.

[0083] Reference below Figure 6 , which shows a schematic diagram of the structure of an audit scheme generating device suitable for implementing the embodiment of the present application. The audit scheme generating device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 6 The audit scheme generating device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0084] like Figure 6 As shown, the audit scheme generating device may include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the audit scheme generating device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the audit solution generation device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows an audit solution generation device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have alternatively.

[0085] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0086] The audit solution generation device provided by the present application adopts the audit solution generation method in the above embodiment, which can solve the technical problem that the existing audit solution generation method is inefficient while ensuring the flexibility of audit solution generation. Compared with the prior art, the beneficial effects of the audit solution generation device provided by the present application are the same as the beneficial effects of the audit solution generation method provided by the above embodiment, and the other technical features in the audit solution generation device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0087] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0088] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0089] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the audit scheme generating method in the above-mentioned embodiment.

[0090] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM: Random Access Memory), a read-only memory (ROM: Read Only Memory), an erasable programmable read-only memory (EPROM: Erasable Programmable Read Only Memory or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM: CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency: Radio Frequency), etc., or any suitable combination of the above.

[0091] The computer-readable storage medium may be included in the audit solution generation device; or may exist independently without being assembled into the audit solution generation device.

[0092] The above-mentioned computer-readable storage medium carries one or more programs. When the above-mentioned one or more programs are executed by the audit plan generating device, the audit plan generating device: inputs user problem information to the audit planning agent for analysis to obtain routing planning information; processes the user problem information according to the routing planning information and through the audit planning agent to obtain an initial audit plan; and evaluates the initial audit plan through the audit planning agent calling the audit evaluation agent to obtain a target audit plan.

[0093] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0094] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0095] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0096] The readable storage medium provided by the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned audit scheme generation method, and can solve the technical problem that the existing audit scheme generation method is inefficient while ensuring the flexibility of audit scheme generation. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present application are the same as the beneficial effects of the audit scheme generation method provided by the above-mentioned embodiment, and will not be repeated here.

[0097] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned audit scheme generation method when executed by a processor.

[0098] The computer program product provided by the present application can solve the technical problem that the existing audit scheme generation method is inefficient while ensuring the flexibility of audit scheme generation. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as the beneficial effects of the audit scheme generation method provided by the above embodiment, which will not be repeated here.

[0099] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A method for generating an audit plan, characterized in that: The audit scheme generation method comprises: Input user question information to the audit planning agent to obtain routing planning information; According to the routing planning information, at least one agent is called through the audit planning agent to process the user question information to obtain an initial audit plan; The audit planning agent calls the audit evaluation agent to evaluate the initial audit plan to obtain the target audit plan.

2. The method according to claim 1, characterized in that The process of processing the user problem information according to the routing planning information and calling at least one agent through the audit planning agent to obtain an initial audit plan includes: Calling the audit matter agent and the audit scheme agent through the audit planning agent according to the routing planning information; Input the user question information into the audit matter agent to obtain the target audit matter; Input the user question information and the target audit items into the audit program intelligent body to obtain the target audit purpose, target audit content and target audit procedure; An initial audit plan is determined based on the target audit matters, the target audit objectives, the target audit content and the target audit procedures.

3. The method according to claim 2, characterized in that The step of inputting the user question information into the audit matter agent to obtain the target audit matter includes: Perform information retrieval on the user problem information according to the audit solution database to obtain historical audit items; The user problem information is processed according to the historical audit matters and through the audit matter intelligent agent to obtain the target audit matters.

4. The method according to claim 3, characterized in that Before the step of performing information retrieval based on the user question information and the audit scheme database to obtain historical audit items, the method further includes: Extract information from historical audit data based on audit system information to obtain audit extraction information; Determine corresponding audit vector information and audit keywords according to the audit extraction information; Establish a keyword index and a vector index respectively according to the audit vector information and the audit keyword; An audit solution database is obtained according to the historical audit data, the keyword index and the vector index.

5. The method according to any one of claims 1 to 4, characterized in that: The step of evaluating the initial audit plan by the audit planning agent calling the audit evaluation agent to obtain the target audit plan includes: The audit planning agent calls the audit evaluation agent to evaluate the initial audit plan to obtain an audit plan evaluation result; In response to the audit scheme evaluation result being a passed evaluation result, the initial audit scheme is determined to be a target audit scheme.

6. The method according to claim 5, characterized in that After the step of evaluating the initial audit scheme by the audit planning agent calling the audit evaluation agent to obtain the audit scheme evaluation result, the method further includes: In response to the audit plan evaluation result being an evaluation failure result, the audit planning agent calls the audit modification agent to modify the initial audit plan to obtain an updated audit plan; Input the updated audit plan to the audit evaluation agent for plan evaluation to obtain an updated plan evaluation result; When the evaluation result of the update plan is a passing result, the update audit plan is determined to be a target audit plan.

7. An audit plan generation system, characterized in that: The system includes: a problem input terminal, a solution generation terminal and a solution output terminal. The problem input terminal is used to receive problem information input by a user; the solution generation terminal is used to generate a target audit solution based on the user's problem information and by calling an audit matter agent, an audit solution agent, an audit evaluation agent and an audit modification agent through an audit planning agent; and the solution output terminal is used to output the generated target audit solution.

8. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the audit scheme generation method as described in any one of claims 1 to 6 are implemented.

9. An audit scheme generating device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the audit scheme generating method according to any one of claims 1 to 6.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the audit scheme generation method according to any one of claims 1 to 6 are implemented.

Images