Flight auditing system and method

Abstract

The invention provides a flight auditing system and method, and relates to the technical field of big data. According to the system, through cooperation of modules, a data acquisition module acquires original data, and a risk early warning model is obtained through training of the original data and is synchronized to a flight auditing recognition module; the flight auditing identification module is used for processing the original data of the to-be-audited project acquired by the data acquisition module to obtain a risk prediction result of whether to start a flight auditing process for the to-be-audited project, and when the risk prediction result is started, the pre-auditing preparation module is triggered; the pre-auditing preparation module generates an auditing data directory including a first to-be-audited object for the to-be-audited project according to a data source of original data of the to-be-audited project, and transmits the auditing data directory to the team recommendation module; and the team recommendation module determines a flight auditing team used for auditing the first to-be-audited object according to the collaboration degree among the auditing members and the first to-be-audited object so as to realize automatic and high-efficiency flight auditing of big data.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a flight audit system and method. Background Art

[0002] The review mechanism plays a key role in the audit process. It is the key basis for ensuring that the audit process is clear and transparent, the audit results are accurate and reliable, and the audit conclusions are objective and fair. It is an effective means to improve the effectiveness of internal control, provide reliable decision-making basis, promote risk management and control measures, improve the quality of audit reports, and reduce the time of on-site audits. Therefore, a sound and reliable review mechanism is an important guarantee for ensuring that audits are carried out in an orderly and efficient manner.

[0003] At present, the traditional audit mechanism still relies heavily on manual on-site spot checks. This model can no longer meet the requirements of full risk coverage and continuous risk control, and it greatly increases the time auditors spend on-site, increases audit expenses, and reduces audit efficiency. Traditional audit verification relies heavily on manual collection, sorting, and comparison, and requires manual processing of a large amount of data. Due to the huge manual workload, usually only a small amount of data is checked in detail.

[0004] Therefore, there is an urgent need for an efficient, automated review solution that can review big data. Summary of the invention

[0005] The present application provides a flight audit system and method for efficiently and automatically reviewing large amounts of data.

[0006] In a first aspect, the present application provides a flight audit system, comprising:

[0007] The data collection module is used to collect the original data of the target project and provide data for the flight audit identification module, pre-audit preparation module and team recommendation module. The target project includes the project to be audited and the historical audit project. The risk warning model is trained based on the original data corresponding to the historical audit project. The risk warning model is used to predict the risk of the original data.

[0008] The flight audit identification module is used to input the original data of the project to be audited into the risk warning model for risk prediction and obtain the risk prediction result. The risk prediction result includes whether to start the flight audit process for the project to be audited; when it is determined to start the flight audit process for the project to be audited, the pre-audit preparation module is triggered;

[0009] A pre-audit preparation module, used to determine the data source of the original data of the project to be audited, and generate an audit data directory for the project to be audited according to the data source, and transmit the audit data directory to the team recommendation module, wherein the audit data directory includes the first audit object corresponding to the project to be audited;

[0010] The team recommendation module is used to determine the flight audit team based on the degree of collaboration among audit members and the first object to be audited. The flight audit team is used to audit the first object to be audited.

[0011] Optionally, the risk prediction result also includes early warning risks, where:

[0012] The flight audit identification module is also used to transmit the warning risk and the model characteristics of the risk warning model to the pre-audit preparation module when determining to initiate the flight audit process for the audited project;

[0013] The pre-audit preparation module is also used to analyze the contribution of different model characteristics of the risk warning model to the warning risk and the source of risk according to the warning risk; generate a list of objects to be audited according to the risk source and contribution, and transmit the list of objects to be audited to the team recommendation module. The list of objects to be audited includes the second object to be audited and the importance of the corresponding second object to be audited;

[0014] Correspondingly, the team recommendation module is specifically used to determine the flight audit team based on the degree of collaboration among audit members, the first audit object, the second audit object and the importance of the second audit object. The flight audit team is used to audit the first audit object and the second audit object.

[0015] Optionally, the flight audit identification module includes a hyperparameter tuning unit and a feature engineering unit, wherein:

[0016] A hyperparameter tuning unit is used to optimize all hyperparameters of the risk warning model according to a preset hyperparameter optimization algorithm;

[0017] The feature engineering unit is used to associate the original data of the audit project with the original data of the historical audit projects according to a preset deep feature synthesis algorithm to obtain a target feature vector, and update the risk warning model according to the target feature vector.

[0018] Optionally, the pre-trial preparation module includes a data catalog preparation unit and an interview list preparation unit, wherein:

[0019] A data directory preparation unit is used to determine the data source of the original data of the project to be audited, and generate an audit data directory for the project to be audited based on the data source;

[0020] The interview list preparation unit is used to analyze the contribution of different model characteristics of the risk warning model to the warning risk and the source of risk; and generate a list of objects to be audited based on the source of risk and the contribution, the list of objects to be audited includes the second object to be audited and the importance of the corresponding second object to be audited.

[0021] Optionally, the interview checklist preparation unit is specifically used to:

[0022] According to a preset sorting method, the importance of the second audit object is sorted to obtain a review order of the second audit object corresponding to the importance of the second audit object; wherein the preset sorting method is one of from large to small or from small to large;

[0023] A list of objects to be audited is obtained according to the second objects to be audited, the review order of the second objects to be audited, and the importance of the second objects to be audited.

[0024] Optionally, the team recommendation module includes a member portrait unit and a member recommendation unit, wherein:

[0025] A member portrait unit is used to construct a member portrait based on the audit member's historical audit projects and basic information of the audit member; wherein the basic information includes the skill information of the audit member;

[0026] The member recommendation unit is used to obtain a flight audit team based on member portraits, the degree of collaboration between audit members and the first object to be audited, including: determining the flight audit team based on the degree of collaboration between audit members, the first object to be audited, the second object to be audited and the importance of the second object to be audited, and the flight audit team is used to audit the first object to be audited and the second object to be audited.

[0027] Optionally, the member recommendation unit is specifically used for:

[0028] Input the first audit object and the second audit object into the Bert model for labeling to obtain labels;

[0029] Substitute the target features in the labels and member portraits into the preset formula to obtain the experience matching score of the audit members; where the target feature is the historical audit department in the historical audit project;

[0030] Obtain the skill matching score of the audit member according to the cosine similarity algorithm, the member skill information in the member portrait, and the first audit object;

[0031] According to the idleness information of the audit members, the availability scores of the audit members are obtained;

[0032] The audit member's availability score, the audit member's experience matching score, and the audit skill matching score are weighted and summed to obtain the audit member's personal contribution score;

[0033] The flight audit team is determined based on the individual contribution scores of audit members, the degree of collaboration among audit members, and the importance of the second audit object.

[0034] Optionally, the member recommendation unit is specifically used for:

[0035] Determine the status of jointly completed projects among audit members based on the historical audit projects in all member portraits;

[0036] Based on the Pearson correlation coefficient and the joint completion of projects among audit members, the collaboration scores among audit members are obtained; among them, the collaboration score represents the degree of collaboration;

[0037] According to the importance of the second audit object, the audit members are randomly combined to obtain a candidate audit team;

[0038] Determine the total score of the candidate audit team based on the individual contribution scores of all audit members in the candidate audit team and the collaboration scores among audit members;

[0039] The team with the highest total score among all candidate audit teams will be determined as the flight audit team.

[0040] In a second aspect, the present application provides a flight audit method, comprising:

[0041] Obtain the original data of the project to be audited;

[0042] Input the original data into the risk early warning model for risk prediction and obtain the risk prediction results, which include whether to initiate the flight audit process for the audited items;

[0043] If it is determined to start the flight audit process for the project to be audited, an audit data directory for the project to be audited is generated according to the data source of the original data, wherein the audit data directory includes the first audit object corresponding to the project to be audited;

[0044] A flight audit team is determined based on the degree of collaboration among audit members and the first audit object, and the flight audit team is used to audit the first audit object.

[0045] Optionally, the risk prediction result also includes early warning risks, and the method further includes:

[0046] According to the early warning risk, analyze the contribution of different model characteristics of the risk early warning model to the early warning risk and the source of risk;

[0047] Generate a list of objects to be audited based on the risk sources and contribution, which includes the second objects to be audited and the importance of the corresponding second objects to be audited;

[0048] Correspondingly, a flight audit team is determined based on the degree of collaboration between audit members and the first object to be audited, including: determining the flight audit team based on the degree of collaboration between audit members, the first object to be audited, the second object to be audited, and the importance of the second object to be audited, and the flight audit team is used to audit the first object to be audited and the second object to be audited.

[0049] In a third aspect, an embodiment of the present application provides an electronic device, including: a memory, a processor;

[0050] Memory stores computer-executable instructions;

[0051] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementations of the first aspect.

[0052] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the first aspect above and / or various possible implementations of the first aspect.

[0053] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above first aspect and / or various possible implementation methods of the first aspect.

[0054] The present application provides a flight audit system and method, the system includes: a data acquisition module, a flight audit identification module, a pre-audit preparation module and a team recommendation module, through the cooperation between these modules, wherein the data acquisition module collects original data, and uses the original data to train to obtain a risk warning model, and synchronizes the risk warning model to the flight audit identification module; the flight audit identification module uses the prediction of the flight audit identification module to predict the risk of the original data of the project to be audited collected by the data acquisition module, and obtains the risk prediction result of whether to start the flight audit process for the project to be audited, thereby accurately predicting the risk of the original data and discovering the risk, and when starting the flight audit process for the project to be audited, triggering the pre-audit preparation module, the pre-audit preparation module determines the data source of the original data of the project to be audited, and generates an audit data directory for the project to be audited according to the data source, and transmits the audit data directory to the team recommendation module, wherein the audit data directory includes the first audit object corresponding to the project to be audited; the team recommendation module further determines the flight audit team according to the degree of collaboration between the audit members and the first audit object, and the flight audit team is used to audit the first audit object, thereby realizing automated and efficient flight audits that can review big data. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] 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.

[0056] Figure 1 A schematic diagram of the structure of a flight audit system provided in an embodiment of the present application;

[0057] Figure 2 A flowchart of a flight audit method provided in an embodiment of the present application;

[0058] Figure 3 A flowchart of another flight audit method provided in an embodiment of the present application;

[0059] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0060] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0061] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0062] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0063] At present, the traditional audit verification mechanism needs to manually collect, organize and analyze a large number of paper or electronic documents, which requires a lot of manpower and resources. The audit process is time-consuming and costly. Due to time and cost constraints, the traditional audit verification mechanism usually only uses a small amount of data for spot checks. This method may cause omissions of important anomalies and problems and fail to fully assess risks. The traditional audit verification mechanism is usually a periodic, point-to-point audit method. Problems and anomalies may not be discovered until the end of the audit. Problems and rectifications cannot be discovered in time, and the opportunity for timely corrective measures is missed, causing greater losses. The traditional audit verification mechanism faces difficulties in processing and analyzing large-scale data. Manual operations are prone to business trips, and comprehensive analysis and mining of data requires a lot of time and energy. The traditional audit verification mechanism often needs to collect materials on site for review, resulting in a long on-site audit cycle, increased time for remote office work, and increased auditors' energy and audit costs. The traditional audit mechanism often presents audit results in text form, lacks intuitive visualization such as charts and reports, and limits the in-depth understanding and analysis of data by auditors and related interests. In the traditional audit mechanism, auditors usually communicate with each other orally to inform each other of the current progress. The progress control and clue exchange are not timely, and the auditors do not have timely control of the current progress. In summary, the traditional audit mechanism has the disadvantages of high time cost, limited coverage, large subjective factors, lack of real-time, difficulty in data processing and analysis, lack of visualization and interactivity, etc. These disadvantages limit the efficiency and accuracy of the audit, making it impossible for the organization to discover and solve problems in a timely manner, reducing the significance and value of the audit.

[0064] The present application provides a flight audit system and method based on big data. The flight audit system automates the flight audit process through collaboration between modules, including a data acquisition module that automatically collects raw data, thereby improving data collection efficiency compared to manual collection; a flight audit identification module facilitates the risk warning model to accurately obtain risk prediction results for a large amount of raw data; a pre-audit preparation module generates an audit data directory for the audited project based on the risk prediction results and the source of the raw data, can locate the source of risk, and automatically form the steps of the audit data directory; finally, a team recommendation module determines a flight audit team for the first audited object in the audit data directory through the degree of collaboration between the first audited object and each review member, thereby enabling the flight review team to review the first audited object more efficiently; and at the same time, combined with the degree of collaboration of each review member, the running-in period of the review members is reduced, thereby making the review more efficient and faster.

[0065] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0066] Among them, the flight audit system refers to the audit process that includes the data collection module, flight audit identification module, pre-audit preparation module, and team recommendation module. Specifically, it relies on the perfect risk identification ability to complete the remote and rapid location of risks, and realize the traceability of problems. With the help of the personnel recommendation of the intelligent audit team, it ultimately promotes the timely resolution of major risks. The flight audit system can improve the accuracy, efficiency, and timeliness of modern audits.

[0067] Figure 1 The structure diagram of a flight audit system 10 provided in the embodiment of the present application is shown in FIG. Figure 1 As shown, the flight audit system 10 includes: a data collection module 101, a flight audit identification module 102, a pre-audit preparation module 103 and a team recommendation module 104.

[0068] The data collection module 101 is used to collect the original data of the target project and provide data for the flight audit identification module 102, the pre-audit preparation module 103 and the team recommendation module 104. The target project includes the project to be audited and the historical audit project. The risk warning model is trained based on the original data corresponding to the historical audit project, and the risk warning model is used to predict the risk of the original data.

[0069] Among them, the original data sources of the target project include: (1) Internal enterprise data: This type of data directly reflects the enterprise's operating status, financial status and specific business details, including but not limited to contract information, project information, engineering information, transaction fund flow information, receivables and arrears information and internal process documents, such as contract signing time, contract execution time, contract counterparty, contract receivable (payable) amount, project establishment time, project construction time, project acceptance time, project acceptance payment time, material transfer category, material transfer time, online flow, offline flow, payment account, item classification, material coding, procurement method, market category, etc. (2) External market data: In order to comprehensively evaluate the market environment in which the enterprise is located, it is necessary to collect industry trend reports, competitor analysis, macroeconomic indicators and relevant policies and regulations. (3) Historical audit information: including audit model result data over the years, such as risk classification, risk level, risk involved amount, risk affected province; flight audit problem ledgers, working papers, audit reports, accountability handling methods over the years, such as audit category, audit data list, audit problem list, key department conversation records, etc.

[0070] The methods for collecting raw data are as follows: (1) Automatic capture through API (Application Programming Interface): Using the pre-set API interface, relevant information from the enterprise's internal system and external data sources can be captured in real time. This method is efficient and has good data accuracy. (2) Extraction from ERP system and other enterprise management software: Through the enterprise resource planning (ERP) system and other enterprise management software, structured data can be extracted in batches to facilitate subsequent data integration and analysis. (3) Screening of public information sources: Filter out important information related to the enterprise from public information sources such as news reports and industry research reports.

[0071] The risk warning model can be a random forest model or a neural network.

[0072] The original data of the target project is collected through the data acquisition module 101, and the original data of the target project is preprocessed. Then, the preprocessed original data of the historical audit project is used as training data to train a risk warning model, wherein the preprocessing can be missing value processing, outlier processing, and data format unification.

[0073] The flight audit identification module 102 is used to input the original data of the project to be audited into the risk warning model for risk prediction to obtain the risk prediction result. The risk prediction result includes whether to start the flight audit process for the project to be audited; when it is determined to start the flight audit process for the project to be audited, the pre-audit preparation module 103 is triggered.

[0074] Among them, the original data of the project to be audited is input into the risk warning model, and the risk warning model predicts the risk of high-risk data in the original data of the project to be audited. If high-risk data exists, then the flight audit process for the project to be audited is initiated, and when it is determined to initiate the flight audit process for the project to be audited, it is necessary to trigger the pre-audit preparation module 103 to start the subsequent review process.

[0075] The pre-audit preparation module 103 is used to determine the data source of the original data of the project to be audited, and generate an audit data directory for the project to be audited based on the data source, and transmit the audit data directory to the team recommendation module 104, wherein the audit data directory includes the first audit object corresponding to the project to be audited.

[0076] Among them, the first objects to be reviewed include high-risk businesses, departments, projects, personnel, and amounts involved in projects.

[0077] The pre-audit preparation module 103 traces back the source of high-risk data (i.e., warning risks) in the risk prediction results according to the risk prediction results, and generates an audit data directory.

[0078] Audit data catalog refers to a detailed plan and guide developed to ensure the comprehensiveness and systematicness of the audit when conducting audit work. This catalog usually includes various aspects and steps of the audit to ensure the orderly progress of the audit process. Specifically, it includes: audit objectives, audit scope, audit standards, audit plans, audit procedures, audit data, etc.

[0079] In the embodiment of the present application, the risk prediction result also includes early warning risk. In this case:

[0080] The flight audit identification module 102 is also used to transmit the warning risk and the model characteristics of the risk warning model to the pre-audit preparation module 103 when determining to initiate the flight audit process for the project to be audited;

[0081] The pre-audit preparation module 103 is also used to analyze the contribution of different model features of the risk warning model to the warning risk and the risk source according to the warning risk; generate a list of objects to be audited according to the risk source and contribution, and transmit the list of objects to be audited to the team recommendation module 104, the list of objects to be audited includes the second object to be audited and the importance of the corresponding second object to be audited;

[0082] Correspondingly, the team recommendation module 104 is specifically used to determine the flight audit team based on the degree of collaboration among audit members, the first audit object, the second audit object and the importance of the second audit object. The flight audit team is used to audit the first audit object and the second audit object.

[0083] Among them, the audit object list is a tool used to guide the audit process to ensure that auditors do not miss important steps or areas during the audit process.

[0084] The second audit objects include high-risk businesses, departments, projects, personnel, amounts involved in projects, etc. The second audit objects may be the same as the first audit objects.

[0085] The flight audit identification module 102 passes the key risk model information (i.e., the warning risk and the model characteristics of the risk warning model) to the pre-audit preparation module 103 through an interface call, and provides the pre-audit preparation module 103 with necessary analysis data for determination.

[0086] The team recommendation module 104 is used to determine a flight audit team according to the degree of collaboration among audit members and the first object to be audited. The flight audit team is used to audit the first object to be audited.

[0087] Specifically, the risk warning model determines that risk point A is not within the scope of key business and the amount involved is too small, then the flight audit identification module 102 will determine that the flight audit conditions are not met, and the flight audit process will not be initiated. Risk point B is the current key business and the amount involved is too large. The current risk point has not been audited recently, so the flight audit process will be initiated and risk point B will be sent to the pre-audit preparation module 103. The pre-audit preparation module 103 then determines the source and contribution of risk point B, and then adds risk point B and its contribution to the list of objects to be audited. The list of objects to be audited is then sent to the team recommendation module 104, which is used to determine the flight audit team based on the degree of collaboration between audit members, the first object to be audited, the second object to be audited, and the importance of the second object to be audited.

[0088] The advantage of this setting is that through automated and intelligent module design, the efficiency, quality and cost-effectiveness of flight audits can be significantly improved, while risk identification and management capabilities can be enhanced.

[0089] In an embodiment of the present application, the flight audit system includes: a data acquisition module, a flight audit identification module, a pre-audit preparation module and a team recommendation module. Through the cooperation between these modules, the data acquisition module collects original data, and uses the original data to train a risk warning model, and synchronizes the risk warning model to the flight audit identification module; the flight audit identification module uses the prediction of the flight audit identification module to perform risk prediction on the original data of the project to be audited collected by the data acquisition module, and obtains a risk prediction result of whether to start the flight audit process for the project to be audited, and when starting the flight audit process for the project to be audited, triggers the pre-audit preparation module, and the pre-audit preparation module determines the data source of the original data of the project to be audited, and generates an audit data directory for the project to be audited based on the data source, and transmits the audit data directory to the team recommendation module, wherein the audit data directory includes the first audit object corresponding to the project to be audited; the team recommendation module further determines the flight audit team based on the degree of collaboration between the audit members and the first audit object, and the flight audit team is used to audit the first audit object, thereby realizing automated and efficient flight audit of big data.

[0090] In the above embodiment of the present application, the flight audit identification module 102 includes a hyperparameter tuning unit and a feature engineering unit, wherein:

[0091] A hyperparameter tuning unit is used to optimize all hyperparameters of the risk warning model according to a preset hyperparameter optimization algorithm;

[0092] The feature engineering unit is used to associate the original data of the audit project with the original data of the historical audit projects according to a preset deep feature synthesis algorithm to obtain a target feature vector, and update the risk warning model according to the target feature vector.

[0093] Among them, the hyperparameter tuning unit: This unit will use the hyperparameter optimization (HPO) algorithm according to the incremental data of the data acquisition module 101 to automatically update the hyperparameters of the risk warning model contained in the module to ensure that the model parameters are in line with the current situation, thereby automatically improving the performance and generalization ability of the model. It can automatically find the optimal parameters that adapt to the current model and reduce the cost of manual attempts by modelers. Use the hyperparameter tuning unit to automatically explore the hyperparameters of the model to avoid missing the combination of the best model hyperparameters.

[0094] Feature Engineering Unit: This unit can thoroughly scan all available clues without missing any possible risk points, which is difficult to achieve in traditional manual risk identification methods. This unit uses the Deep Feature Synthesis (DFS) algorithm to collect and process structured data according to the data acquisition module 101, and associates the original data of the audited project with the original data of the historical audited project, and applies mathematical logic in a certain order to create new cross-features. Whenever the data acquisition module 101 collects the original data of the new audited project, this unit will automatically call this algorithm to associate the new data with the original data and produce a new feature vector (i.e., target feature vector). These new feature vectors (i.e., target feature vectors) will be reused for model training to ensure that the risk identification model can cover all data and realize the tracking and analysis of all data obtained by the audit identification module.

[0095] The benefit of this setting is that the combination of hyperparameter tuning and feature engineering units can significantly improve the performance and practicality of the risk warning model, providing strong technical support for auditing and risk management.

[0096] In the above embodiment of the present application, the pre-trial preparation module 103 includes a data directory preparation unit and an interview list preparation unit, wherein:

[0097] A data directory preparation unit is used to determine the data source of the original data of the project to be audited, and generate an audit data directory for the project to be audited based on the data source;

[0098] The interview list preparation unit is used to analyze the contribution of different model characteristics of the risk warning model to the warning risk and the source of risk; and generate a list of objects to be audited based on the source of risk and the contribution, the list of objects to be audited includes the second object to be audited and the importance of the corresponding second object to be audited.

[0099] Among them, the data directory preparation unit: The data directory unit is responsible for preparing the data that needs to be paid attention to during the audit in advance in the pre-audit preparation module 103. It helps auditors confirm the source of the data by identifying the key feature vectors in the risk model in the flight audit identification module 102. At the same time, this unit will retrieve the modification history of the recorded data from the data acquisition module 101, including the person who modified the problem data, so that auditors can quickly locate and solve data problems. In short, the data directory unit provides a clear source of risk data and background modification records for the audit, ensuring the accuracy and efficiency of the audit work.

[0100] Interview list preparation unit: The interview list preparation unit helps the audit team identify the key data sources (i.e. risk sources) that have the greatest impact on the audit by calculating the contribution of different model features of the risk warning model to the warning risk and the risk source. A list of audit objects is generated based on the risk source and contribution. This allows auditors to audit the second audit object based on the contribution, or to allocate different auditors and resources to the corresponding second audit object based on the contribution.

[0101] The SHAP (Shapley Additive Explanation) model is used to calculate the contribution (marginal contribution value, Shapley Value) of different model features to the warning risk to measure the impact of each model feature.

[0102] The audit risk model we use is defined as: , the feature weight explanation model of each model is: , represents the SHAP value of the i-th model feature in the risk warning model, It represents the average value of the prediction of model f for all samples. The mathematical definition of is as follows:

[0103]

[0104]

[0105] in, Representing model features The expected value of marginal contribution under different feature combinations, each feature combination needs to be considered In the parameter combination ( ) and not in the parameter combination ( ) output result; Represents the full set of feature vectors (i.e., model features), with a length of p, i.e., there are a total of p feature vectors. Indicates The subset of features excluded from the feature vector, and finally, represents the probability of each feature combination, It represents the average value predicted by the audit risk model f for all samples (i.e., the original data of historical audit projects).

[0106] According to the classification of the affiliated departments, the SHAP value of each affiliated department (i.e. the department to be audited) is calculated, that is, the weight of each affiliated department for a single audit sample:

[0107]

[0108] in, It represents the importance of department k in the entire risk warning model. This importance value is obtained by summing up the SHAP values ​​of all model features belonging to department k. Representation contains model features A subset of features , Refers to the model features feature subset.

[0109] Specifically, a list of audit objects can be generated based on the departments and their contributions. At the same time, departments correspond to different projects, personnel, businesses, and data sources.

[0110] The benefits of this setup are: supporting an efficient, flexible and data-driven pre-audit preparation process, ensuring that audit work can accurately identify and assess risks, and reasonably allocate resources and priorities.

[0111] In the above embodiment of the present application, the interview list preparation unit is specifically used for:

[0112] According to a preset sorting method, the importance of the second audit object is sorted to obtain a review order of the second audit object corresponding to the importance of the second audit object; wherein the preset sorting method is one of from large to small or from small to large;

[0113] A list of objects to be audited is obtained according to the second objects to be audited, the review order of the second objects to be audited, and the importance of the second objects to be audited.

[0114] Among them, the greater the corresponding contribution of the second audit object, the higher the corresponding importance.

[0115] Specifically, the second audit objects are ranked from large to small according to their importance. The second audit objects with higher importance have higher review orders, and more resources and personnel are allocated to them. Correspondingly, the second audit objects with higher importance have lower review orders, and fewer resources and personnel are allocated to them.

[0116] The benefit of this setup is that it ensures that the audit team can give priority to the second review objects with higher importance, which helps the audit team allocate resources and attention more effectively, thereby improving audit efficiency.

[0117] In the above embodiment of the present application, the team recommendation module 104 includes a member portrait unit and a member recommendation unit, wherein:

[0118] A member portrait unit is used to construct a member portrait based on the audit member's historical audit projects and basic information of the audit member; wherein the basic information includes the skill information of the audit member;

[0119] The member recommendation unit is used to obtain a flight audit team based on member portraits, the degree of collaboration between audit members and the first object to be audited, including: determining the flight audit team based on the degree of collaboration between audit members, the first object to be audited, the second object to be audited and the importance of the second object to be audited, and the flight audit team is used to audit the first object to be audited and the second object to be audited.

[0120] Among them, historical review items correspond to historical review departments, businesses and other information.

[0121] The member portrait module is used to build user portraits based on the audit members' skill information, historical review projects, historical review departments, and historical review businesses.

[0122] The member recommendation module obtains the flight audit team through member portraits, collaboration between audit members, and calculation with the first audit object. At the same time, the higher the importance of the second audit object, the more likely it is to affect the number of members of the audit team, requiring audit members with high experience and more matching skill information.

[0123] The advantage of this setting is that the team recommendation module 104 can significantly improve the efficiency and effectiveness of audit team formation and ensure high-quality completion of audit tasks through intelligent member portraits and recommendation mechanisms.

[0124] In the above embodiment of the present application, the member recommendation unit is specifically used for:

[0125] Input the first audit object and the second audit object into the Bert model for labeling to obtain labels;

[0126] Substitute the target features in the labels and member portraits into the preset formula to obtain the experience matching score of the audit members; where the target feature is the historical audit department in the historical audit project;

[0127] Obtain the skill matching score of the audit member according to the cosine similarity algorithm, the member skill information in the member portrait, and the first audit object;

[0128] According to the idleness information of the audit members, the availability scores of the audit members are obtained;

[0129] The audit member's availability score, the audit member's experience matching score, and the audit skill matching score are weighted and summed to obtain the audit member's personal contribution score;

[0130] The flight audit team is determined based on the individual contribution scores of audit members, the degree of collaboration among audit members, and the importance of the second audit object.

[0131] The first audit object and the second audit object may be departments, and the labels are labels representing the departments.

[0132] The BERT model (Bidirectional Encoder Representations from Transformers) uses a bidirectional Transformer architecture to generate contextual representations of words.

[0133] The Bert model is used to label the audit working papers (i.e. the first audit object and the second audit object), and the labels are the departments involved in the input audit working papers. The departments involved are :

[0134]

[0135] Analyze historical flight audit cases (i.e., audit members’ historical audit projects), especially those similar to the current audit assignment, to identify team members who have performed well in similar assignments. The experience and expertise of these team members are highly applicable to the current assignment. Audit staff About Department The level of familiarity is the historical audit projects that the auditor has participated in ( ), that is:

[0136]

[0137] Auditors Experience matching score , defined as the sum of its familiarity with all departments (N), that is:

[0138]

[0139] For availability scores, when auditors If there is no free time for the first audit object or the second audit object, the availability score is a negative value, for example: -10. If there is free time, the availability score is a positive number, for example 10. The specific free time information can be assigned based on whether there is free time when the review plan is generated (i.e., the review time is officially determined), whether the project has just been completed before the review plan is generated, or whether there are long-term special plans in the future.

[0140] Matching with the professional skills of auditors, screening out auditors with higher relevance scores, auditor i for audit project j skill matching score , calculated by cosine similarity:

[0141]

[0142] For the audited project j (i.e. the first audit object or the second audit object), the personal contribution of audit member i is: project-person-score (pps):

[0143]

[0144] Among them, α, β, and δ are the weights of each score, which can be set according to the actual needs of the enterprise; is the experience matching score of audit member i, is the skill matching score of team member i for project j to be audited, is the availability score of team member i during the estimated time of audit project j. If team member i is not available during the estimated period of audit project j, is negative.

[0145] The flight audit team is determined based on a comprehensive consideration of the individual contribution scores of audit members, the degree of collaboration among audit members, and the importance of the second audit object.

[0146] The benefit of this setup is that this systematic and intelligent team building method can significantly improve the overall effectiveness of audit work and provide strong support for the organization's risk management and compliance.

[0147] In the above embodiment of the present application, the member recommendation unit is specifically used for:

[0148] Determine the status of jointly completed projects among audit members based on the historical audit projects in all member portraits;

[0149] Based on the Pearson correlation coefficient and the joint completion of projects among audit members, the collaboration scores among audit members are obtained; among them, the collaboration score represents the degree of collaboration;

[0150] According to the importance of the second audit object, the audit members are randomly combined to obtain a candidate audit team;

[0151] Determine the total score of the candidate audit team based on the individual contribution scores of all audit members in the candidate audit team and the collaboration scores among audit members;

[0152] The team with the highest total score among all candidate audit teams will be determined as the flight audit team.

[0153] Among them, the collaborative filtering algorithm is used to recommend a combination of members who can form an efficient collaborative team based on the cooperation performance and complementarity of team members in historical tasks. This helps to optimize the cooperation between team members and improve the efficiency and quality of audit work. Specifically, the Pearson correlation coefficient is used to calculate the collaboration score between team members. Assume that the collaboration score of member i and member j in historical tasks is , then the Pearson correlation coefficient It can be calculated by the following formula:

[0154]

[0155] in, and are the scores of member i and member j in task k, respectively; m is the number of tasks completed jointly in the past (i.e., the number of projects completed jointly by each audit member); and are the average scores of member i and member j respectively.

[0156] When the importance of the second audit object is high, a larger number of flight audit teams are required, and therefore more audit members are included in the random combination.

[0157] For organizing flight audit teams , to conduct a comprehensive score of the effect of the flight audit on the audited project i Sum the individual contribution scores of all team members and the collaboration scores among team members:

[0158]

[0159] Through the application of the above recommendation algorithms and principles, the system can intelligently recommend team members suitable for the current flight audit project. For the audit project i, the final audit team of the organization Make For a high total score (i.e. ), the system will automatically generate a selectable team set ,Right now:

[0160]

[0161]

[0162] The benefit of this setup is that it can significantly improve the efficiency and effectiveness of audit work by optimizing the construction and management of the audit team in a systematic and data-driven manner.

[0163] A flight audit system provided by an embodiment of the present application has a data acquisition module capable of collecting original data, a flight audit identification module capable of performing risk prediction on the original data through a risk warning model, a pre-audit preparation module capable of backtracking the original data to obtain the data source of the original data, and then obtaining an audit data directory, and finally, according to the audit data directory, a flight audit team is obtained, which can accurately and timely discover risks, and according to the audit data directory, confirm a suitable flight audit team, thereby improving the efficiency of flight audits.

[0164] It should be noted that it should be understood that the division of the various modules of the above system is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated. And these modules can all be implemented in the form of software called by processing elements; they can also be all implemented in the form of hardware; some modules can also be implemented in the form of software called by processing elements, and some modules can be implemented in the form of hardware. For example, the processing module can be a separately established processing element, or it can be integrated in a certain chip of the above system for implementation. In addition, it can also be stored in the memory of the above system in the form of program code, and called and executed by a certain processing element of the above system. The implementation of other modules is similar. In addition, these modules can be fully or partially integrated together, or they can be implemented independently. The processing element here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each module above can be completed by an integrated logic circuit of hardware in the processor element or instructions in the form of software.

[0165] For example, the above modules may be one or more integrated circuits configured to implement the above methods, such as one or more application specific integrated circuits (ASIC), or one or more microprocessors (DSP), or one or more field programmable gate arrays (FPGA). For another example, when a module above is implemented in the form of a processing element scheduling program code, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call program code. For another example, these modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0166] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loading and executing computer instructions on a computer, a process or function according to an embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (Digital Subscriber Line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or a data center that includes one or more available media integrations. The available media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, Digital Video Discs (DVD)), or semiconductor media (eg, solid state disks (SSD)).

[0167] Figure 2 A schematic diagram of a flight audit method provided in an embodiment of the present application. Figure 2 As shown, the method includes:

[0168] S201. Obtaining the original data of the project to be audited;

[0169] S202, inputting the original data into the risk early warning model to perform risk prediction and obtain risk prediction results, the risk prediction results including whether to initiate a flight audit process for the project to be audited;

[0170] In the embodiment of the present application, the risk prediction result also includes a warning risk, and the method further includes:

[0171] According to the early warning risk, analyze the contribution of different model characteristics of the risk early warning model to the early warning risk and the source of risk;

[0172] Generate a list of objects to be audited based on the risk sources and contribution, which includes the second objects to be audited and the importance of the corresponding second objects to be audited;

[0173] Correspondingly, a flight audit team is determined based on the degree of collaboration between audit members and the first object to be audited, including: determining the flight audit team based on the degree of collaboration between audit members, the first object to be audited, the second object to be audited, and the importance of the second object to be audited, and the flight audit team is used to audit the first object to be audited and the second object to be audited.

[0174] S203: If it is determined to start the flight audit process for the project to be audited, an audit data directory for the project to be audited is generated according to the data source of the original data, wherein the audit data directory includes the first audit object corresponding to the project to be audited;

[0175] S204. Determine a flight audit team based on the degree of collaboration among audit members and the first audit object. The flight audit team is used to audit the first audit object.

[0176] A flight audit method provided in an embodiment of the present application can perform risk prediction through original data to obtain risk prediction results, and determine whether to start a flight audit process based on the risk prediction results. If the flight audit process is determined, the data source of the original data is obtained, and an audit data directory is generated. According to the audit data directory, the flight audit team is determined, thereby realizing efficient and rapid flight audits.

[0177] Figure 3 A flowchart of another flight audit method provided in an embodiment of the present application is shown as follows: Figure 3 As shown, the method includes:

[0178] S301, Data Collection: Responsible for the unified collection, storage and pre-processing of all data involved in flight audit.

[0179] S302, Flight Audit Risk Identification: Use algorithms and models to conduct in-depth analysis of the collected data, treat the flight audit project according to the background characteristics of the flight audit project and the analysis results of different algorithms and models, and ultimately provide a decision on whether to initiate the flight audit, as well as risk data.

[0180] Among them, the algorithm includes: trend analysis and other methods. The model can be a random forest, a neural network, etc.

[0181] Background features refer to the basic information for risk identification that the algorithm and model need to determine, such as whether the current risk is within the scope of the group's major business, whether it is a key enterprise risk category, the amount of money involved in the risk, etc. Specifically, if risk point A is not within the scope of key business and the amount involved is too small, it will be judged as not meeting the conditions for flight audit in the flight audit identification. Risk point B is the current key business, the amount involved is too large, and the current risk point has not been audited recently, which is the risk data.

[0182] S303, Pre-audit preparation: trace back the source of risk data and generate an audit data directory; analyze the contribution of different risk data in the flight audit identification module to the audit items, and generate a list to be audited based on the risk data and the contribution of the risk data. The list to be audited includes: a list of departments to be audited and personnel to be reviewed and marked with importance. The importance is marked according to the contribution of the risk data.

[0183] S304, Team Recommendation: Based on the audit data catalog and the list of items to be audited, analyze the key issues and concerns of the items to be audited, and combine the professional background, past experience, skill matching of the audit team members, as well as the current workload and time arrangements of the audit team members to accurately match the most suitable audit team for the items to be audited.

[0184] Another flight audit method provided in an embodiment of the present application improves the efficiency and accuracy of flight audits, saves a lot of time and manpower costs, and provides data (i.e., data catalogs, lists to be audited) support for the audit team, thereby improving the limitations of traditional audit methods and playing a great role in curbing the spread and expansion of internal risks in the enterprise.

[0185] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 4 As shown, the electronic device 40 includes:

[0186] The electronic device 40 may include a processor 401 with one or more processing cores, a memory 402 with one or more computer-readable storage media, a communication component 403 and other components. The processor 401 , the memory 402 and the communication component 403 are connected via a bus 404 .

[0187] In the specific implementation process, at least one processor 401 executes the computer execution instructions stored in the memory 402, so that at least one processor 401 executes a flight audit method as described above.

[0188] The specific implementation process of the processor 401 can be found in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be repeated here.

[0189] In the above Figure 4 In the illustrated embodiment, it should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. The general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc. The steps of the method disclosed in the invention may be directly embodied as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0190] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (NVM), such as at least one disk storage.

[0191] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of this application is not limited to only one bus or one type of bus.

[0192] In some embodiments, a computer program product is also proposed, including a computer program or instructions, which implement the steps of any of the above-mentioned flight audit methods when executed by a processor.

[0193] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0194] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0195] To this end, an embodiment of the present application provides a computer-readable storage medium, in which multiple instructions are stored, and the instructions can be loaded by a processor to execute the steps of any flight audit method provided in the embodiment of the present application.

[0196] The storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0197] According to one aspect of the present application, a computer program product or a computer program is provided. The computer program product or the computer program comprises computer instructions stored in a computer-readable storage medium.

[0198] Since the instructions stored in the storage medium can execute the steps in any flight audit method provided in the embodiments of the present application, the beneficial effects that can be achieved by any flight audit method provided in the embodiments of the present application can be achieved. Please see the previous embodiments for details and will not be repeated here.

[0199] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0200] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A flight audit system, characterized in that: include: Data collection module, flight audit identification module, pre-audit preparation module and team recommendation module, including: The data acquisition module is used to collect the original data of the target project and provide data for the flight audit identification module, the pre-audit preparation module and the team recommendation module. The target project includes the project to be audited and the historical audit project. The risk warning model is trained based on the original data corresponding to the historical audit project, and the risk warning model is used to predict the risk of the original data. The flight audit identification module is used to input the original data of the project to be audited into the risk warning model to perform risk prediction and obtain a risk prediction result, wherein the risk prediction result includes whether to start the flight audit process for the project to be audited; when it is determined to start the flight audit process for the project to be audited, the pre-audit preparation module is triggered; The pre-audit preparation module is used to determine the data source of the original data of the project to be audited, and generate an audit data directory for the project to be audited according to the data source, and transmit the audit data directory to the team recommendation module, wherein the audit data directory includes the first audit object corresponding to the project to be audited; The team recommendation module is used to determine a flight audit team based on the degree of collaboration between audit members and the first object to be audited, and the flight audit team is used to audit the first object to be audited.

2. The flight audit system according to claim 1, characterized in that: The risk prediction results also include early warning risks, in which case: The flight audit identification module is further used to transmit the warning risk and the model characteristics of the risk warning model to the pre-audit preparation module when determining to initiate the flight audit process for the project to be audited; The pre-audit preparation module is further used to analyze the contribution of different model features of the risk warning model to the warning risk and the risk source according to the warning risk; generate a list of objects to be audited according to the risk source and the contribution, and transmit the list of objects to be audited to the team recommendation module, wherein the list of objects to be audited includes a second object to be audited and the importance of the corresponding second object to be audited; Correspondingly, the team recommendation module is specifically used to determine the flight audit team based on the degree of collaboration among audit members, the first audit object, the second audit object and the importance of the second audit object, and the flight audit team is used to audit the first audit object and the second audit object.

3. The flight audit system according to claim 1 or 2, characterized in that: The flight audit identification module includes a hyperparameter tuning unit and a feature engineering unit, wherein: The hyperparameter tuning unit is used to optimize all hyperparameters of the risk warning model according to a preset hyperparameter optimization algorithm; The feature engineering unit is used to associate the original data of the project to be audited with the original data of the historical audit project according to a preset deep feature synthesis algorithm to obtain a target feature vector, and update the risk warning model according to the target feature vector.

4. The flight audit system according to claim 1 or 2, characterized in that: The pre-trial preparation module includes a data directory preparation unit and an interview list preparation unit, wherein: The data directory preparation unit is used to determine the data source of the original data of the project to be audited, and generate an audit data directory for the project to be audited according to the data source; The interview list preparation unit is used to analyze the contribution of different model characteristics of the risk warning model to the warning risk and the source of risk; and generate a list of objects to be audited based on the risk source and the contribution, and the list of objects to be audited includes the second object to be audited and the importance of the corresponding second object to be audited.

5. The flight audit system according to claim 4, characterized in that: The interview checklist preparation unit is specifically used for: Sorting the importance of the second audit object according to a preset sorting method to obtain a review order of the second audit object corresponding to the importance of the second audit object; wherein the preset sorting method is one of large to small or small to large; The list of objects to be audited is obtained according to the second objects to be audited, the review order of the second objects to be audited, and the importance of the second objects to be audited.

6. The flight audit system according to claim 2, characterized in that: The team recommendation module includes a member portrait unit and a member recommendation unit, wherein: The member portrait unit is used to construct a member portrait based on the historical audit projects of the audit member and the basic information of the audit member; wherein the basic information includes the skill information of the audit member; The member recommendation unit is used to obtain the flight audit team based on the member portraits, the degree of collaboration between the audit members and the first object to be audited, including: determining the flight audit team based on the degree of collaboration between the audit members, the first object to be audited, the second object to be audited and the importance of the second object to be audited, and the flight audit team is used to audit the first object to be audited and the second object to be audited.

7. The flight audit system according to claim 6, characterized in that: The member recommendation unit is specifically used for: Inputting the first object to be audited and the second object to be audited into the Bert model for labeling to obtain labels; Substituting the target feature in the label and the member portrait into a preset formula to obtain the experience matching score of the audit member; wherein the target feature is the historical audit department in the historical audit project; Obtaining a skill matching score of the audit member according to a cosine similarity algorithm, the member skill information in the member portrait, and the first audit object; According to the idleness information of the audit members, the availability scores of the audit members are obtained; The audit member's availability score, the audit member's experience matching score, and the audit skill matching score are weighted and summed to obtain the audit member's personal contribution score; The flight audit team is determined based on the individual contribution scores of the audit members, the degree of collaboration among the audit members, and the importance of the second audit object.

8. The flight audit system according to claim 7, characterized in that: The member recommendation unit is specifically used for: Determine the status of jointly completed projects among audit members based on the historical audit projects in all member portraits; According to the Pearson correlation coefficient and the joint completion of projects among the audit members, the collaboration scores among the audit members are obtained; wherein the collaboration scores represent the collaboration degree; According to the importance of the second audit object, randomly combine the audit members to obtain a candidate audit team; Determine the total score of the candidate audit team based on the individual contribution scores of all audit members in the candidate audit team and the collaboration scores among audit members; The team with the highest total score among all candidate audit teams is determined as the flight audit team.

9. A flight audit method, characterized in that: Applied to the flight audit system according to any one of claims 1 to 8, the flight audit method comprises: Obtain the original data of the project to be audited; Inputting the original data into a risk early warning model for risk prediction to obtain a risk prediction result, wherein the risk prediction result includes whether to initiate a flight audit process for the project to be audited; If it is determined to start the flight audit process for the item to be audited, then generating an audit data directory for the item to be audited according to the data source of the original data, wherein the audit data directory includes the first object to be audited corresponding to the item to be audited; A flight audit team is determined based on the degree of collaboration among audit members and the first object to be audited, and the flight audit team is used to audit the first object to be audited.

10. The method according to claim 9, characterized in that The risk prediction result also includes early warning risks, and the method further includes: According to the early warning risk, analyzing the contribution of different model characteristics of the risk early warning model to the early warning risk and the risk source; Generating a list of objects to be audited according to the risk source and the contribution, the list of objects to be audited including the second objects to be audited and the importance of the corresponding second objects to be audited; Correspondingly, the determining of the flight audit team based on the degree of collaboration between audit members and the first object to be audited includes: determining the flight audit team based on the degree of collaboration between audit members, the first object to be audited, the second object to be audited, and the importance of the second object to be audited, and the flight audit team is used to audit the first object to be audited and the second object to be audited.

11. An electronic device, characterized in that: include: Memory, processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 9 to 10.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 9 to 10 when executed by a processor.

13. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method according to any one of claims 9 to 10 is implemented.

Images