Data auditing intelligent learning method and system based on big data

Abstract

The invention relates to the technical field of data auditing, in particular to a data auditing intelligent learning method and system based on big data. The method comprises the following steps: acquiring and analyzing a project auditing signal, and determining auditing content; according to the audit content, calling a material warehouse-in and warehouse-out record; analyzing the material warehouse-in and warehouse-out records, and determining whether abnormal warehouse refund exists or not; and if yes, according to the suspicious record corresponding to the abnormal refund, determining the review content, and sending the review content to the reviewer. Project auditing signals are automatically obtained and analyzed, manual data processing is reduced, and auditing efficiency is improved. The influence of human errors and subjective judgment on the audit result is reduced, and the objectivity and accuracy of the audit result are improved. The big data analysis technology is utilized to identify the mode and trend of abnormal refund, discover suspicious behaviors, prevent risks such as inventory false increase and false profit, and can be used in risk control fields such as internal control auditing, construction project auditing, project material auditing and the like.

Description

Technical Field

[0001] The present application relates to the field of data auditing technology, and in particular to a data auditing intelligent learning method and system based on big data. Background Art

[0002] With the rapid development of information technology, big data technology is increasingly being used in various fields. In the production management process of enterprises, data auditing is an important means to ensure the rational use of enterprise resources and prevent corruption and waste. Its efficiency and accuracy have become the focus of enterprises.

[0003] False returns refer to the process of returning materials or products that have been used but not actually consumed at the end of the month or during inventory counting, in order to correctly calculate the current production cost or sales cost, and then re-using the same amount at the beginning of the next month. Traditional data auditing methods often rely on manual work, which is not only inefficient, but also easily affected by subjective factors, resulting in inaccurate audit results, and failure to detect abnormal false returns. There are risks such as inflated inventory, false profits, and impact on decision-making. Summary of the invention

[0004] The present application provides a data audit intelligent learning method and system based on big data to solve the above problems.

[0005] In a first aspect, the present application provides a data audit intelligent learning method based on big data, the method comprising: Obtain and analyze project audit signals and determine audit content; According to the audit content, retrieve the material in and out records; analyze the material in and out records to determine whether there is any abnormal return; If so, the review content is determined based on the suspicious records corresponding to the abnormal return, and the review content is sent to the reviewer.

[0006] Through this solution, project audit signals are automatically acquired and analyzed, reducing the need for manual data processing, thereby greatly improving the efficiency of the audit. By analyzing historical data and real-time data, the system can accurately identify false warehouse return behaviors at the end of the month or during inventory counting, and even complex false warehouse return patterns can be detected. By analyzing data through algorithms, the impact of human errors and subjective judgments on audit results is reduced, and the objectivity and accuracy of audit results are improved. Using big data analysis technology, patterns and trends of abnormal warehouse returns can be identified more quickly, so as to timely detect suspicious behaviors and prevent risks such as inflated inventory and false profits. Timely detection and correction of abnormal warehouse returns can help maintain inventory accuracy, avoid excessive inventory or inventory shortages, and thus reduce costs and operational risks. Problems discovered through audits can prompt companies to strengthen internal controls, improve relevant management processes, and improve overall management levels. Through problems discovered through audits and improvement measures, companies can allocate resources more effectively, optimize production processes, and improve overall operational efficiency.

[0007] Optionally, analyzing the material in-and-out records to determine whether there is an abnormal return includes: Determine the return period of materials based on the audit content; According to the return cycle, analyze the material in and out records to determine whether there is an abnormally high frequency of returns; If there is an abnormally high frequency of returns, obtain the factory's processing plan; analyze the processing plan to determine the frequency of material consumption; According to the material consumption frequency, determining whether the abnormally high frequency of rejection is in compliance with the processing plan; If it does not meet the requirements, it is determined that there is an abnormal return.

[0008] Through this solution, automated analysis of material in and out records can significantly improve audit efficiency and reduce the time and resources required for manual review. Through algorithms and data analysis, abnormal return behavior can be more accurately identified, reducing the impact of human errors and subjective judgments. Timely detection and correction of abnormal return behavior can help prevent false inventory and ensure the accuracy of financial statements. Through precise auditing, false profits caused by abnormal returns can be avoided, ensuring the authenticity and transparency of financial reports.

[0009] Optionally, analyzing the material in-and-out records to determine whether there is an abnormal return includes: If there is no abnormally high frequency of withdrawal, determine whether the withdrawal time requirement is met according to the withdrawal cycle; If the return time requirement is not met, the processing records within the preset time period with the current time as the deadline are obtained; Analyze the processing record to determine whether the current processing task is different from the processing plan; If it is determined that there is a difference, the current processing amount is determined based on the processing record; According to the current processing volume, determine whether there is any abnormal return to stock.

[0010] This solution can automate the processing of large amounts of data, reducing the time and labor intensity required for manual audits. The efficiency of audits is improved by automating tasks through algorithms and systems. Inaccurate audit results caused by human errors or negligence are reduced. The influence of subjective factors is reduced, and the objectivity of audits is improved. The material in and out of the warehouse can be monitored in real time to detect problems in a timely manner. Regular analysis can prevent the occurrence of abnormal warehouse withdrawal behavior. The audit process is data-driven, reducing human intervention and improving accuracy.

[0011] Optionally, the content of the review is determined based on the suspicious records corresponding to the abnormal return, including: Determine the operator of the abnormal return based on the suspicious records; Obtain and analyze the operator's historical in-and-out records to determine the false judgment results of each historical in-and-out record; The review content is determined according to the false determination result.

[0012] Through this solution, by analyzing suspicious records, operators who may have abnormal returns can be accurately located. Based on the false judgment results, all inbound and outbound records of relevant operators can be comprehensively reviewed. By reviewing and correcting abnormal returns, similar risks can be prevented from occurring in the future. Inventory inflation and cost errors caused by abnormal returns can be reduced, thereby reducing the company's financial costs. Improve the transparency and accuracy of audits and enhance stakeholders' trust in corporate financial reports.

[0013] Optionally, determining the review content according to the false determination result includes: Determining the credibility of the operator according to the false determination result; Obtaining the position information of the operator; Determining the operable space of the operator according to the position information and the reputation value; Determining whether the operator has the authority to withdraw the payment based on the operable space; If it is determined that it does not exist, the management scope is determined based on the position information, and the review content is determined based on the management scope.

[0014] Through this solution, the key areas and contents of the review can be accurately determined based on the false judgment results. Combined with position information, reputation value and management scope, all relevant behaviors of operators can be comprehensively reviewed. Through the review, similar risks can be prevented from occurring in the future. The audit results can be used to continuously monitor the behavior of operators to ensure their compliance. Stakeholders' trust in the company's financial reports and internal control processes is improved.

[0015] Optionally, determining the review content according to the management scope includes: Based on the material in and out records, obtain material information; Analyze the material information and determine the material return method based on the information analysis results; The material return method includes a false return method. If the material return method is the material return method, the material value is determined according to the material information; The review content shall be determined based on the value of the materials and the scope of management.

[0016] Through this solution, by combining the management scope, the review content is more focused, unnecessary reviews are avoided, and the pertinence of the review is improved. The review focuses on key areas, reducing the time and resource consumption of the review and improving the efficiency of the review. Through the review, deficiencies in internal control can be discovered and corrected, and the effectiveness of internal control can be enhanced. According to the management scope, the key areas and contents of the review can be accurately determined. Combined with the management scope, all relevant behaviors of operators can be comprehensively reviewed. Through the review, similar risks can be prevented from occurring in the future. The review results can be used to continuously monitor the behavior of operators to ensure their compliance. Stakeholders' trust in the company's financial reporting and internal control processes is improved.

[0017] Optionally, also include: Determine the return credit points based on the value of the materials; According to the warehouse return credit score, the credit value of the operator is adjusted to obtain an adjusted credit value; updating the withdrawal authority according to the adjusted credit value; An operation monitoring plan within a preset time period is determined based on the withdrawal authority.

[0018] Through this solution, by linking with reputation points and reputation values, operators will be more cautious in handling material returns to maintain their reputation. Adjust permissions based on reputation values ​​to ensure that permissions match responsibilities. A strict reputation point and reputation value adjustment mechanism can effectively curb false returns and reduce risks such as false inventory increases and false profits. The automated reputation point and permission update process reduces the workload of manual audits and improves audit efficiency.

[0019] Optionally, determining the reputation value of the operator according to the false determination result includes: Obtaining a preset personnel evaluation rule, and determining the initial credibility value of the operator according to the preset personnel evaluation rule; Determine the number of false determinations and the number of historical in-and-out stockings according to the false determination results in the historical in-and-out stocking record set; The credibility value of the operator is determined based on the initial credibility value, the number of false determinations and the historical number of in-and-out warehouse times.

[0020] Through this solution, the transparency and fairness of the evaluation process are ensured through quantified reputation values ​​and clear evaluation rules, and each operator can clearly understand his or her reputation status and improvement direction. The increase or decrease of reputation values ​​is directly related to the daily behavior of operators, which can motivate operators to handle inbound and outbound transactions more responsibly and avoid violations. Clear penalties (such as reduced reputation values) have a deterrent effect on reducing violations such as false returns, and help maintain the authenticity and accuracy of inventory. The automated reputation value calculation and adjustment process reduces the need for manual audits, improves audit efficiency, and reduces audit costs.

[0021] Optionally, the credibility value of the operator is determined based on the initial credibility value, the number of false determinations and the historical number of in-and-out warehouses, and is calculated according to the following formula: ; in, Indicates the reputation value of the operator; represents the initial reputation value; Indicates the number of false judgments; Indicates the historical in / out times in the historical in / out records.

[0022] Through this scheme, the use of a logarithmic function means that the impact of the number of false judgments on the reputation value is nonlinear. As the number of false positives increases, the rate of decrease of reputation value will accelerate, which helps to punish serious violations more effectively. The logarithmic function smooths the impact of the number of false positives on the reputation value, avoiding the unfair punishment that may be caused by a simple linear relationship, especially in In smaller cases. By introducing As the denominator, the formula takes into account the operator's historical workload and experience, avoiding large adjustments to reputation based on just a few violations. The formula provides a transparent and predictable reputation adjustment mechanism, and operators can understand how their reputation changes based on their behavior and experience. The dynamic nature of the formula allows the reputation to adjust in real time based on the operator's behavior, which helps to reflect the operator's work performance in a timely manner.

[0023] In a second aspect, the present application provides a data audit intelligent learning system based on big data, comprising: Signal analysis module, used to obtain and analyze project audit signals and determine audit content; A record analysis module is used to retrieve material in and out records according to the audit content; analyze the material in and out records to determine whether there is an abnormal return; The content sending module is used to determine the review content according to the suspicious records corresponding to the abnormal return, if any, and send the review content to the reviewer.

[0024] Optionally, when the record analysis module analyzes the material in-and-out records to determine whether there is an abnormal return, it is used to: Determine the return period of materials based on the audit content; According to the return cycle, analyze the material in and out records to determine whether there is an abnormally high frequency of returns; If there is an abnormally high frequency of returns, obtain the factory's processing plan; analyze the processing plan to determine the frequency of material consumption; According to the material consumption frequency, determining whether the abnormally high frequency of rejection is in compliance with the processing plan; If it does not meet the requirements, it is determined that there is an abnormal return.

[0025] Optionally, when the record analysis module analyzes the material in-and-out records to determine whether there is an abnormal return, it is used to: If there is no abnormally high frequency of withdrawal, determine whether the withdrawal time requirement is met according to the withdrawal cycle; If the return time requirement is not met, the processing records within the preset time period with the current time as the deadline are obtained; Analyze the processing record to determine whether the current processing task is different from the processing plan; If it is determined that there is a difference, the current processing amount is determined based on the processing record; According to the current processing volume, determine whether there is any abnormal return to stock.

[0026] Optionally, when determining the review content according to the suspicious records corresponding to the abnormal return, the content sending module is used to: Determine the operator of the abnormal return based on the suspicious records; Obtain and analyze the operator's historical in-and-out records to determine the false judgment results of each historical in-and-out record; The review content is determined according to the false determination result.

[0027] Optionally, when the content sending module determines the review content according to the false determination result, it is used to: Determining the credibility of the operator according to the false determination result; Obtaining the position information of the operator; Determining the operable space of the operator according to the position information and the reputation value; Determining whether the operator has the authority to withdraw the payment based on the operable space; If it is determined that it does not exist, the management scope is determined based on the position information, and the review content is determined based on the management scope.

[0028] Optionally, when the content sending module determines the review content according to the management scope, it is used to: Based on the material in and out records, obtain material information; Analyze the material information and determine the material return method based on the information analysis results; The material return method includes a false return method. If the material return method is the material return method, the material value is determined according to the material information; The review content shall be determined based on the value of the materials and the scope of management.

[0029] Optionally, the data audit intelligent learning system further includes a program formulation module, which is used to: Determine the return credit points based on the value of the materials; According to the warehouse return credit score, the credit value of the operator is adjusted to obtain an adjusted credit value; updating the withdrawal authority according to the adjusted credit value; An operation monitoring plan within a preset time period is determined based on the withdrawal authority.

[0030] Optionally, when the content sending module determines the reputation value of the operator according to the false determination result, it is used to: Obtaining a preset personnel evaluation rule, and determining the initial credibility value of the operator according to the preset personnel evaluation rule; Determine the number of false determinations and the number of historical in-and-out stockings according to the false determination results in the historical in-and-out stocking record set; The credibility value of the operator is determined based on the initial credibility value, the number of false determinations and the historical number of in-and-out warehouse times.

[0031] Optionally, the content sending module determines the credibility of the operator according to the initial credibility value, the number of false determinations and the number of historical in-and-out times, and calculates it according to the following formula: ; in, Indicates the reputation value of the operator; represents the initial reputation value; Indicates the number of false judgments; Indicates the historical in / out times in the historical in / out records. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0033] Figure 1 A schematic diagram of an application scenario provided for an embodiment of the present application; Figure 2 A flowchart of a data audit intelligent learning method based on big data provided in one embodiment of the present application; Figure 3 A schematic diagram of the structure of a data audit intelligent learning system based on big data is provided for one embodiment of the present application. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0035] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0036] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0037] False returns refer to the process of returning materials or products that have been used but not actually consumed at the end of the month or during inventory counting, in order to correctly calculate the current production cost or sales cost, and then re-using the same amount at the beginning of the next month. Traditional data auditing methods often rely on manual work, which is not only inefficient, but also easily affected by subjective factors, resulting in inaccurate audit results, and failure to detect abnormal false returns. There are risks such as inflated inventory, false profits, and impact on decision-making.

[0038] Based on this, the present application provides a data audit intelligent learning method and system based on big data, which acquires and analyzes project audit signals and determines the audit content; retrieves material in and out records according to the audit content; analyzes material in and out records to determine whether there is abnormal withdrawal; if so, determines the review content according to the suspicious records corresponding to the abnormal withdrawal, and sends the review content to the reviewer. Automatic acquisition and analysis of project audit signals reduces the need for manual data processing, thereby greatly improving the efficiency of auditing. By analyzing historical data and real-time data, the system can accurately identify false withdrawal behaviors at the end of the month or at inventory, and even complex false withdrawal patterns can be detected. By analyzing data through algorithms, the impact of human errors and subjective judgments on audit results is reduced, and the objectivity and accuracy of audit results are improved. Using big data analysis technology, patterns and trends of abnormal withdrawal can be identified more quickly, so as to timely discover suspicious behaviors and prevent risks such as false inventory and false profits. Timely discovery and correction of abnormal withdrawals can help maintain inventory accuracy, avoid excessive inventory or inventory shortages, and thus reduce costs and operational risks. Problems found through audits can prompt companies to strengthen internal control, improve relevant management processes, and improve overall management levels. Through problems found through audits and improvement measures, companies can allocate resources more effectively, optimize production processes, and improve overall operational efficiency. This technology can be used in risk control areas such as internal control audits, construction project audits, and project material audits.

[0039] Figure 1A schematic diagram of an application scenario provided by the present application is provided. During the production cycle audit, the method provided by the present application is applied. Specifically, the method provided by the present application is applied to any server, and the server, the production management system and the terminal of the auditor interact with each other. The project audit signal is automatically obtained and analyzed from the production management system, reducing the need for manual data processing, thereby greatly improving the efficiency of the audit. By analyzing historical data and real-time data, the system can accurately identify false warehouse withdrawal behaviors at the end of the month or at inventory, and even complex false warehouse withdrawal patterns can be detected. By analyzing data through algorithms, the impact of human errors and subjective judgments on the audit results is reduced, and the objectivity and accuracy of the audit results are improved. Using big data analysis technology, the patterns and trends of abnormal warehouse withdrawals can be identified more quickly, so as to timely discover suspicious behaviors and prevent risks such as false inventory increases and false profits. Timely discovery and correction of abnormal warehouse withdrawals can help maintain inventory accuracy, avoid excessive inventory or inventory shortages, and thus reduce costs and operational risks. The problems found by the audit are sent to the auditor's terminal, which can prompt the company to strengthen internal control, improve relevant management processes, and improve the overall management level. Through the problems found by the audit and improvement measures, the company can allocate resources more effectively, optimize production processes, and improve overall operational efficiency. This technology can be used in risk control areas such as internal control audits, construction project audits, and project material audits.

[0040] For specific implementation methods, please refer to the following embodiments.

[0041] Figure 2 This is a flowchart of a data audit intelligent learning method based on big data provided by an embodiment of the present application. The method of this embodiment can be applied to the server in the above scenario. Figure 2 As shown, the method includes: S201. Obtain and analyze project audit signals and determine audit content.

[0042] Project audit signals can be data or information generated during the construction and production process that can reflect the status, efficiency, cost, etc. of production activities.

[0043] The audit content can be production activities, processes or data that need to be focused on and reviewed during the audit process.

[0044] Specifically, the audit signals in the production process, such as production orders, bills of materials, production schedules, inventory changes, etc., are automatically acquired through the enterprise resource planning system (ERP), material requirement planning system (MRP) or other production management systems. The acquired audit signals are de-noised, normalized and standardized using data cleaning and preprocessing techniques to ensure data quality. The preprocessed data are subjected to pattern recognition and trend analysis using data mining and machine learning techniques to identify possible abnormal signals. Based on the analysis results, the key areas and specific contents of the audit are determined, such as abnormal consumption of specific materials or production lines.

[0045] S202. According to the audit content, retrieve the material in and out records; analyze the material in and out records to determine whether there is any abnormal return.

[0046] Material entry and exit records can be detailed records of materials entering and leaving the warehouse, including raw material entry orders, finished product exit orders, return orders, material type, quantity, time, operator, etc.

[0047] Abnormal returns may be returns that do not comply with normal business processes, usually involving untrue inventory adjustments or bookkeeping operations, such as raw material receipts, finished product delivery orders, and material return orders.

[0048] Specifically, according to the determined audit content, the in-and-out records of relevant materials are automatically retrieved from the warehouse management system (WMS) or other inventory management systems. Statistical methods and anomaly detection algorithms, such as box plot analysis and Z-score, are used to identify outliers in the in-and-out records. By setting business rules and thresholds, it is detected whether there are abnormal return patterns in the records, such as frequent false return behaviors. The detected abnormal records are screened to determine which records are suspicious and may indicate abnormal returns.

[0049] S203. If so, determine the review content based on the suspicious records corresponding to the abnormal return, and send the review content to the reviewer.

[0050] Suspicious records may be records discovered during the audit process that may not be consistent with normal business processes or may contain abnormalities, such as frequent return operations, outbound records with abnormal quantities, unauthorized inventory adjustments, etc.

[0051] The review may be a detailed examination by the auditor of a specific return record, including the reason for the return, the personnel involved, the relevant financial records, etc.

[0052] Specifically, if there is an abnormal return, the detected abnormal return record will be marked as a suspicious record. According to the suspicious record, the specific review content is determined, such as the details of the return, specific time, quantity, operator and other information. Through the company's internal communication tools or audit management system, the review content is automatically sent to the terminal of the responsible reviewer for in-depth investigation.

[0053] Through this solution, project audit signals are automatically acquired and analyzed, reducing the need for manual data processing, thereby greatly improving the efficiency of the audit. By analyzing historical data and real-time data, the system can accurately identify false warehouse return behaviors at the end of the month or during inventory counting, and even complex false warehouse return patterns can be detected. By analyzing data through algorithms, the impact of human errors and subjective judgments on audit results is reduced, and the objectivity and accuracy of audit results are improved. Using big data analysis technology, patterns and trends of abnormal warehouse returns can be identified more quickly, so as to timely detect suspicious behaviors and prevent risks such as inflated inventory and false profits. Timely detection and correction of abnormal warehouse returns can help maintain inventory accuracy, avoid excessive inventory or inventory shortages, and thus reduce costs and operational risks. Problems discovered through audits can prompt companies to strengthen internal controls, improve relevant management processes, and improve overall management levels. Through problems discovered through audits and improvement measures, companies can allocate resources more effectively, optimize production processes, and improve overall operational efficiency.

[0054] In some embodiments, the material return cycle is determined based on the audit content; based on the return cycle, the material entry and exit records are analyzed to determine whether there is an abnormally high-frequency return; if there is an abnormally high-frequency return, the factory's processing plan is obtained; the processing plan is analyzed to determine the material consumption frequency; based on the material consumption frequency, it is determined whether the abnormally high-frequency return complies with the processing plan; if not, it is determined that an abnormal return exists.

[0055] Audit content can be specific items or areas that need to be checked, verified and analyzed during the audit process.

[0056] The return cycle can be the time interval between the return of materials to the next use.

[0057] Abnormally high frequency of returns and withdrawals may refer to the abnormally high frequency of returns and withdrawals in the material entry and exit records, including frequent returns and withdrawals of a certain material in a short period of time, or the number of returns is much greater than the number of withdrawals.

[0058] The processing plan can be formulated according to production needs, including detailed plans for material consumption, production schedule, personnel arrangements, etc., including the number of parts that need to be produced each day, the types and quantities of materials required, and the arrangements for machinery and equipment.

[0059] Material consumption frequency can be the number of times or quantity of material consumed per unit time during the production process.

[0060] Specifically, collect historical in-and-out data of materials from the ERP system or inventory management system. Use time series analysis and statistical methods to analyze the time intervals between materials in and out of the warehouse and determine the normal withdrawal cycle of materials. According to the determined withdrawal cycle, screen out the in-and-out records that conform to the cycle law. Apply statistical methods, such as Z-score or IQR (interquartile range) method, to detect outliers in the in-and-out records. Calculate the in-and-out frequency of each material and identify materials with abnormal withdrawal frequency. Set the threshold for abnormally high-frequency withdrawal according to industry standards and historical data. Compare the withdrawal frequency with the set threshold to identify materials with a withdrawal frequency higher than the normal level. Obtain the processing plan of the relevant materials from the production planning system or ERP system. Analyze the material consumption schedule and quantity in the processing plan to determine the expected consumption frequency of the materials. Determine the actual consumption frequency of the materials based on the processing plan and actual production data. Compare the actual withdrawal frequency with the expected consumption frequency. If the actual withdrawal frequency does not match the expected consumption frequency, especially if it is significantly higher than expected, it is considered that there may be abnormal withdrawal.

[0061] Through this solution, automated analysis of material in and out records can significantly improve audit efficiency and reduce the time and resources required for manual review. Through algorithms and data analysis, abnormal return behavior can be more accurately identified, reducing the impact of human errors and subjective judgments. Timely detection and correction of abnormal return behavior can help prevent false inventory and ensure the accuracy of financial statements. Through precise auditing, false profits caused by abnormal returns can be avoided, ensuring the authenticity and transparency of financial reports.

[0062] In some embodiments, if there is no abnormally high frequency of returns, determine whether the current return time requirement is met based on the return cycle; if the return time requirement is not met, obtain the processing records within a preset time period with the current moment as the end time; analyze the processing records to determine whether the current processing task differs from the processing plan; if it is determined that there is a difference, determine the current processing volume based on the processing records; and determine whether there is an abnormal return based on the current processing volume.

[0063] The return time requirement can be the time standard or regulations set by an enterprise or organization for the return and issuance of materials.

[0064] The cut-off time can be a specific point in time determined during an audit or record analysis, after which data will not be considered.

[0065] The preset period can be a pre-set time range in auditing or data analysis, which is used to analyze data within a specific time period.

[0066] Processing records can be detailed records that record the processing of materials or products during the production process.

[0067] The current processing task can be a production activity that is being executed or is about to start, which includes specific production goals, required resources and estimated completion time.

[0068] The current processing volume can be the actual processing volume completed within a certain period of time.

[0069] Specifically, according to the preset return cycle, check whether the current time reaches the next return time point. If the current time does not match the return cycle, it indicates that there may be abnormal return behavior. Extract the processing records within the preset period calculated from the current moment from the production management system or ERP system. Select the processing records related to the audit content, such as specific material codes, production lines or production batches. Compare the obtained processing records with the preset processing plan to check whether there are any differences. Confirm whether each processing task is consistent with the planned task, including material requirements, production quantity, time schedule, etc. Calculate the total processing volume completed in the current period based on the processing records. Determine the amount of materials required to complete these processing tasks. Compare the current processing volume with the return volume in the material in and out records. If the return volume does not match the actual processing volume, especially if the return volume is significantly higher than the actual consumption, there may be abnormal return behavior. Conduct a detailed review of the records suspected of abnormal returns, including the reasons for the return, approval process, inventory adjustments, etc.

[0070] This solution can automate the processing of large amounts of data, reducing the time and labor intensity required for manual audits. The efficiency of audits is improved by automating tasks through algorithms and systems. Inaccurate audit results caused by human errors or negligence are reduced. The influence of subjective factors is reduced, and the objectivity of audits is improved. The material in and out of the warehouse can be monitored in real time to detect problems in a timely manner. Regular analysis can prevent the occurrence of abnormal warehouse withdrawal behavior. The audit process is data-driven, reducing human intervention and improving accuracy.

[0071] In some embodiments, based on suspicious records, operators who perform abnormal warehouse returns are identified; historical warehouse entry and exit records of operators are obtained and analyzed to determine the false judgment results of each historical warehouse entry and exit; and based on the false judgment results, the review content is determined.

[0072] Operators can be people who perform specific tasks in a particular organization or institution, such as factory workers, warehouse managers, employees in the finance department, etc.

[0073] Historical in-and-out records can be detailed records of items entering and leaving the warehouse or production line, including material collection, return, production use, sales and delivery, etc.

[0074] The false determination result can be a determination of whether a certain record or behavior is false through data analysis or auditing process. For example, if a certain warehouse entry and exit record shows that the material was not actually used after being issued, and it is determined to be false after analysis, then the false determination result of the record is "yes".

[0075] Specifically, filter out suspicious return records from the audit system. These records may be marked because of abnormal return frequency, quantity or time point. Analyze suspicious records and determine the personnel who performed these return operations. Extract the historical in and out records of relevant operators from the ERP system or inventory management system. Sort and classify historical records for subsequent analysis. Analyze the historical in and out records of operators to observe whether there are regular or abnormal patterns in their in and out behaviors. According to the set standards and rules, make false judgments on historical records, that is, identify which in and out records may be false or abnormal. Apply logical rules and statistical analysis methods to make false judgments on each historical in and out record. Record the judgment results of each record, including normal, suspicious or false. Based on the false judgment results, determine the key areas of review, such as specific time periods, specific materials, specific operators, etc. Based on the judgment results, determine the depth of the review, including detailed document review, on-site investigation or further data analysis, so as to determine the content of the review.

[0076] Through this solution, by analyzing suspicious records, operators who may have abnormal returns can be accurately located. Based on the false judgment results, all inbound and outbound records of relevant operators can be comprehensively reviewed. By reviewing and correcting abnormal returns, similar risks can be prevented from occurring in the future. Inventory inflation and cost errors caused by abnormal returns can be reduced, thereby reducing the company's financial costs. Improve the transparency and accuracy of audits and enhance stakeholders' trust in corporate financial reports.

[0077] In some embodiments, the credibility value of the operator is determined based on the false determination result; the position information of the operator is obtained; the operable space of the operator is determined based on the position information and the credibility value; based on the operable space, it is determined whether the operator has the authority to withdraw the payment; if it is determined that the authority does not exist, the management scope is determined based on the position information, and the review content is determined based on the management scope.

[0078] The reputation value may be a value derived from the operator's historical behavior and performance evaluation to measure his or her reliability or trustworthiness. The reputation value may be a score between 0 and 100, with a higher score indicating a better reputation.

[0079] Position information can be detailed information describing a person's role, responsibilities, and authority in an organization, including position title, direct superior, subordinates, job description, management area, responsible projects, etc.

[0080] The operable space may refer to the range of activities that an operator can perform within the scope permitted by the position. For example, if someone’s position is a warehouse manager, then his operable space may include warehouse management, inventory control, material distribution, etc.

[0081] Return authority can be whether the operator has the right to return or re-issue materials from inventory.

[0082] The scope of management can be the area or department that the operator is responsible for supervising, controlling and guiding, such as a production line, a department or a group of subordinate employees.

[0083] Specifically, a reputation value is calculated based on the operator's historical inventory records and false judgment results to reflect the compliance and reliability of his / her operations. The reputation value is compared with the preset standard to evaluate the operator's reputation level. The operator's position information is extracted from the human resources management system or the enterprise organizational structure. The operator's position information is sorted and classified for subsequent analysis. Based on the position information, the operator's responsibilities and authority are analyzed to determine their operational space. The operator's actual operation is compared with the operational space to check whether there is any unauthorized behavior. Based on the position information and reputation value, it is determined whether the operator has the authority to return the materials. Confirm whether the operator's authority to return the materials is consistent with his / her position and reputation value. Based on the position information, the operator's management scope is determined. Confirm whether the operator's management scope is reasonable and whether it matches his / her position. Based on the management scope, the key areas of review are determined, such as specific subordinates, specific areas, specific projects, etc. Based on the management scope, the review content is determined.

[0084] Through this solution, the key areas and contents of the review can be accurately determined based on the false judgment results. Combined with position information, reputation value and management scope, all relevant behaviors of operators can be comprehensively reviewed. Through the review, similar risks can be prevented from occurring in the future. The audit results can be used to continuously monitor the behavior of operators to ensure their compliance. Stakeholders' trust in the company's financial reports and internal control processes is improved.

[0085] In some embodiments, material information is obtained based on material in and out records; the material information is analyzed, and the material return method is determined based on the information analysis results; the material return method includes a false return method, and if the material return method is a material return method, the material value is determined based on the material information; the review content is determined based on the material value and the management scope.

[0086] Material information can be a detailed record of materials, including but not limited to the name, specification, model, quantity, storage date, supplier information, price, etc. of the material.

[0087] The information analysis results can be the conclusions drawn after processing and analyzing the collected data. For example, by analyzing the material in and out records, information analysis results such as material consumption rate and inventory turnover rate can be obtained.

[0088] The material return method can be the specific operation method and process of returning materials from the using department to the inventory.

[0089] A false return can be a violation of the rules in which materials are returned on the books, but in reality the materials are not actually returned to inventory; for example, in order to falsify costs or inventory, an operator may record the return of materials on the books, but in reality the materials are still in use.

[0090] Material value can be the financial cost or price of the material, which usually includes the purchase cost, transportation cost, taxes, etc. of the material.

[0091] Specifically, extract material in and out records from the inventory management system or ERP system. Extract material information, including material name, quantity, in and out date, price, etc. Clean the collected material information to ensure the accuracy and completeness of the data. Analyze the trend of material in and out and identify abnormalities or abnormal patterns. Identify possible false return patterns by analyzing material in and out records. Determine whether the material return method is a false return based on the definition of false return and the identified pattern. Calculate the cost value of false return materials, including direct costs and indirect costs. Record the cost value of materials for subsequent review and risk assessment. Based on the value of the materials, evaluate the impact of false return behavior on financial statements and decision-making. Combined with the management scope, determine the specific review content, including but not limited to the following aspects: Inventory accuracy: Check whether the inventory records are consistent with the actual inventory. Cost calculation: Review whether the cost calculation method complies with accounting standards. Approval process: Check whether the return approval process complies with company policies. Internal control: Evaluate the effectiveness of internal control, including separation of duties and supervision mechanisms. Risk report: Prepare a risk report based on the review results and put forward improvement suggestions.

[0092] Through this solution, by combining the management scope, the review content is more focused, unnecessary reviews are avoided, and the pertinence of the review is improved. The review focuses on key areas, reducing the time and resource consumption of the review and improving the efficiency of the review. Through the review, deficiencies in internal control can be discovered and corrected, and the effectiveness of internal control can be enhanced. According to the management scope, the key areas and contents of the review can be accurately determined. Combined with the management scope, all relevant behaviors of operators can be comprehensively reviewed. Through the review, similar risks can be prevented from occurring in the future. The review results can be used to continuously monitor the behavior of operators to ensure their compliance. Stakeholders' trust in the company's financial reporting and internal control processes is improved.

[0093] In some embodiments, a return credit score is determined based on the material value; based on the return credit score, the operator's credit value is adjusted to obtain an adjusted credit value; based on the adjusted credit value, the return authority is updated; based on the return authority, an operation monitoring plan within a preset time period is determined.

[0094] The return credit score can be an evaluation mechanism to measure the compliance and reliability of operators in the material return process.

[0095] An operations monitoring program can be a set of systematic methods and measures to monitor and record operators' inbound and outbound activities over a specific period of time to ensure that these activities comply with company policies and regulatory requirements.

[0096] Specifically, operators with higher reputation values ​​can be trusted more, so that the number and frequency of reviews of operators can be reduced, thereby reducing audit costs. Standards for credit points for return of materials can be formulated based on the value of materials. For example, the return of high-value materials may receive lower credit points. The return credit points of operators are calculated based on the actual value of materials returned. Rules for credit value adjustment are formulated based on credit points. For example, the higher the credit points, the more credit value is improved. The credit value of operators is adjusted based on the calculated credit points. Rules for updating return rights are formulated based on the adjusted credit values. For example, higher return rights can be obtained when the credit value reaches a certain level. The return rights of operators are updated based on the adjustment of credit values. Based on the return rights of operators, an operation monitoring plan within a preset period is designed. The monitoring plan is implemented to monitor the in-and-out behavior of operators.

[0097] Through this solution, by linking with reputation points and reputation values, operators will be more cautious in handling material returns to maintain their reputation. Adjust permissions based on reputation values ​​to ensure that permissions match responsibilities. A strict reputation point and reputation value adjustment mechanism can effectively curb false returns and reduce risks such as false inventory increases and false profits. The automated reputation point and permission update process reduces the workload of manual audits and improves audit efficiency.

[0098] In some embodiments, preset personnel assessment rules are obtained, and the initial credibility value of the operator is determined according to the preset personnel assessment rules; the number of false determinations and the historical number of in-and-out records are determined according to the false determination results of the historical in-and-out records; the credibility value of the operator is determined according to the initial credibility value, the number of false determinations and the historical number of in-and-out records.

[0099] The preset personnel assessment rules can be a series of standards and criteria pre-set by the enterprise to evaluate and measure the work performance and reputation of operators, including the increase in credibility value for each correct completion of the warehouse entry and exit operation, the decrease in credibility value for each false warehouse return behavior, and the setting of upper and lower limits of credibility value.

[0100] The initial reputation value may be a starting reputation score set for an operator, which is usually assigned when the operator starts working.

[0101] The number of false judgments can be the number of times an operator is marked as non-compliant or having false behavior by the system or auditors in the historical inventory records.

[0102] The historical number of in-and-out warehouse operations may be the cumulative number of in-and-out warehouse operations completed by the operator in the past period of time.

[0103] The initial reputation value may be a dynamic score calculated by preset rules based on the operator's work performance and compliance, and is used to reflect the operator's reputation level.

[0104] Specifically, the preset personnel assessment rules are extracted from the enterprise's management system. These rules may include the initial setting of the reputation value, the standard for increasing or decreasing the reputation value, etc. According to the preset personnel assessment rules, an initial reputation value is set for each operator. The initial reputation value is assigned to the corresponding operator. The operator's historical entry and exit records and false judgment results are collected. The number of false judgments and historical entry and exit times of each operator are counted. According to the historical entry and exit times, the operator's initial reputation value is adjusted on a basic basis. According to the number of false judgments, the operator's reputation value is adjusted up or down. Generally, the more false judgments there are, the more the reputation value decreases.

[0105] Through this solution, the transparency and fairness of the evaluation process are ensured through quantified reputation values ​​and clear evaluation rules, and each operator can clearly understand his or her reputation status and improvement direction. The increase or decrease of reputation values ​​is directly related to the daily behavior of operators, which can motivate operators to handle inbound and outbound transactions more responsibly and avoid violations. Clear penalties (such as reduced reputation values) have a deterrent effect on reducing violations such as false returns, and help maintain the authenticity and accuracy of inventory. The automated reputation value calculation and adjustment process reduces the need for manual audits, improves audit efficiency, and reduces audit costs.

[0106] In some embodiments, the operator's reputation value is determined based on the initial reputation value, the number of false determinations, and the number of historical in-and-out warehouse times, and is calculated according to the following formula (1): ; in, Indicates the reputation value of the operator; represents the initial reputation value; Indicates the number of false judgments; Indicates the historical in / out times in the historical in / out records.

[0107] Specifically, the initial reputation value of each operator is extracted from the enterprise's management system . Collect and count the number of false judgments made by each operator . Count the historical in and out times of each operator . Determine the adjustment coefficient based on the actual situation and policies of the enterprise For each operator, use formula (1) to calculate its reputation value The calculated reputation value Updated to each operator's personal record. Among them, the adjustment factor is a positive number that determines the impact of the number of false judgments on the reputation value. The larger the number, the greater the impact of false judgments on the reputation value. Logarithmic function It is usually used to transform a nonlinear relationship into a linear relationship. Here, the logarithmic function can be used to smooth the impact of the number of false judgments on the reputation value. Initial reputation value It is the benchmark of reputation. In order to punish the number of false judgments , we need to subtract a certain number of points from the initial reputation value. In order to make the penalty equal to the number of false judgments Proportional to the historical number of inbound and outbound transactions , we introduce the logarithmic function. Logarithmic function It means that as the number of false judgments increases, the value of the logarithmic function will also increase, resulting in a greater penalty on the reputation value. Combining the above two parts, we get the final formula (1).

[0108] Through this scheme, the use of a logarithmic function means that the impact of the number of false judgments on the reputation value is nonlinear. As the number of false positives increases, the rate of decrease of reputation value will accelerate, which helps to punish serious violations more effectively. The logarithmic function smooths the impact of the number of false positives on the reputation value, avoiding the unfair punishment that may be caused by a simple linear relationship, especially in In smaller cases. By introducing As the denominator, the formula takes into account the operator's historical workload and experience, avoiding large adjustments to reputation based on just a few violations. The formula provides a transparent and predictable reputation adjustment mechanism, and operators can understand how their reputation changes based on their behavior and experience. The dynamic nature of the formula allows the reputation to adjust in real time based on the operator's behavior, which helps to reflect the operator's work performance in a timely manner.

[0109] Figure 3 A schematic diagram of the structure of a data audit intelligent learning system based on big data provided by an embodiment of the present application is as follows: Figure 3 As shown, the data audit intelligent learning system 300 based on big data of this embodiment includes: a signal analysis module 301, a record analysis module 302 and a content sending module 303.

[0110] The signal analysis module 301 is used to obtain and analyze the project audit signal and determine the audit content; The record analysis module 302 is used to retrieve the material in and out records according to the audit content; analyze the material in and out records to determine whether there is an abnormal return; The content sending module 303 is used to determine the review content according to the suspicious records corresponding to the abnormal return, if any, and send the review content to the reviewer.

[0111] Optionally, when the record analysis module 302 analyzes the material in-and-out records to determine whether there is an abnormal return, it is used to: Determine the return period of materials based on the audit content; According to the return cycle, analyze the material in and out records to determine whether there is an abnormally high frequency of returns; If there is an abnormally high frequency of returns, obtain the factory's processing plan; analyze the processing plan to determine the frequency of material consumption; According to the material consumption frequency, determining whether the abnormally high frequency of rejection is in compliance with the processing plan; If it does not meet the requirements, it is determined that there is an abnormal return.

[0112] Optionally, when the record analysis module 302 analyzes the material in-and-out records to determine whether there is an abnormal return, it is used to: If there is no abnormally high frequency of withdrawal, determine whether the withdrawal time requirement is met according to the withdrawal cycle; If the return time requirement is not met, the processing records within the preset time period with the current time as the deadline are obtained; Analyze the processing record to determine whether the current processing task is different from the processing plan; If it is determined that there is a difference, the current processing amount is determined based on the processing record; According to the current processing volume, determine whether there is any abnormal return to stock.

[0113] Optionally, when determining the review content according to the suspicious records corresponding to the abnormal return, the content sending module 303 is used to: Determine the operator of the abnormal return based on the suspicious records; Obtain and analyze the operator's historical in-and-out records to determine the false judgment results of each historical in-and-out record; The review content is determined according to the false determination result.

[0114] Optionally, when determining the review content according to the false determination result, the content sending module 303 is used to: Determining the credibility of the operator according to the false determination result; Obtaining the position information of the operator; Determining the operable space of the operator according to the position information and the reputation value; Determining whether the operator has the authority to withdraw the payment based on the operable space; If it is determined that it does not exist, the management scope is determined based on the position information, and the review content is determined based on the management scope.

[0115] Optionally, when the content sending module 303 determines the review content according to the management scope, it is used to: Based on the material in and out records, obtain material information; Analyze the material information and determine the material return method based on the information analysis results; The material return method includes a false return method. If the material return method is the material return method, the material value is determined according to the material information; The review content shall be determined based on the value of the materials and the scope of management.

[0116] Optionally, the data audit intelligent learning system 300 further includes a program formulation module 304, which is used to: Determine the return credit points based on the value of the materials; According to the warehouse return credit score, the credit value of the operator is adjusted to obtain an adjusted credit value; updating the withdrawal authority according to the adjusted credit value; An operation monitoring plan within a preset time period is determined based on the withdrawal authority.

[0117] Optionally, when the content sending module 303 determines the reputation value of the operator according to the false determination result, it is used to: Obtaining a preset personnel evaluation rule, and determining the initial credibility value of the operator according to the preset personnel evaluation rule; Determine the number of false determinations and the number of historical in-and-out stockings according to the false determination results in the historical in-and-out stocking record set; The credibility value of the operator is determined based on the initial credibility value, the number of false determinations and the historical number of in-and-out warehouse times.

[0118] Optionally, the content sending module 303 determines the credibility of the operator according to the initial credibility value, the number of false determinations and the number of historical in-and-out warehouse times, and calculates it according to the following formula: ; in, Indicates the reputation value of the operator; represents the initial reputation value; Indicates the number of false judgments; Indicates the historical in / out times in the historical in / out records.

[0119] The system of this embodiment can be used to execute the method of any of the above embodiments. The implementation principles and technical effects are similar and will not be described in detail here.

Claims

1. A data audit intelligent learning method based on big data, characterized in that: include: Obtain and analyze project audit signals and determine audit content; According to the audit content, retrieve the material in and out records; Analyze the material in and out records to determine whether there is any abnormal return; If so, the review content is determined based on the suspicious records corresponding to the abnormal return, and the review content is sent to the reviewer.

2. The method according to claim 1, characterized in that The analyzing the material in and out records to determine whether there is an abnormal return includes: Determine the return period of materials based on the audit content; According to the return cycle, analyze the material in and out records to determine whether there is an abnormally high frequency of returns; If there is an abnormally high frequency of returns, obtain the factory's processing plan; analyze the processing plan to determine the frequency of material consumption; According to the material consumption frequency, determining whether the abnormally high frequency of rejection is in compliance with the processing plan; If it does not meet the requirements, it is determined that there is an abnormal return.

3. The method according to claim 2, characterized in that The analyzing the material in and out records to determine whether there is an abnormal return includes: If there is no abnormally high frequency of withdrawal, determine whether the withdrawal time requirement is met according to the withdrawal cycle; If the return time requirement is not met, the processing records within the preset time period with the current time as the deadline are obtained; Analyze the processing record to determine whether the current processing task is different from the processing plan; If it is determined that there is a difference, the current processing amount is determined based on the processing record; According to the current processing volume, determine whether there is any abnormal return to stock.

4. The method according to claim 1, characterized in that: The content of the review is determined based on the suspicious records corresponding to the abnormal return, including: Determine the operator of the abnormal return based on the suspicious records; Obtain and analyze the operator's historical in-and-out records to determine the false judgment results of each historical in-and-out record; The review content is determined according to the false determination result.

5. The method according to claim 4, characterized in that Determining the review content according to the false determination result includes: Determining the credibility of the operator according to the false determination result; Obtaining the position information of the operator; Determining the operable space of the operator according to the position information and the reputation value; Determining whether the operator has the authority to withdraw the payment based on the operable space; If it is determined that it does not exist, the management scope is determined based on the position information, and the review content is determined based on the management scope.

6. The method according to claim 5, characterized in that According to the management scope, the review content is determined, including: Based on the material in and out records, obtain material information; Analyze the material information and determine the material return method based on the information analysis results; The material return method includes a false return method. If the material return method is the material return method, the material value is determined according to the material information; The review content shall be determined based on the value of the materials and the scope of management.

7. The method according to claim 6, characterized in that Also includes: Determine the return credit points based on the value of the materials; According to the warehouse return credit score, the credit value of the operator is adjusted to obtain an adjusted credit value; updating the withdrawal authority according to the adjusted credit value; An operation monitoring plan within a preset time period is determined based on the withdrawal authority.

8. The method according to claim 5, characterized in that Determining the reputation value of the operator according to the false determination result includes: Obtaining a preset personnel evaluation rule, and determining the initial credibility value of the operator according to the preset personnel evaluation rule; Determine the number of false determinations and the number of historical in-and-out stockings according to the false determination results in the historical in-and-out stocking record set; The credibility value of the operator is determined based on the initial credibility value, the number of false determinations and the historical number of in-and-out warehouse times.

9. The method according to claim 8, characterized in that The operator's reputation value is determined based on the initial reputation value, the number of false determinations, and the historical number of in-and-out warehouses, and is calculated according to the following formula: ； in, Indicates the reputation value of the operator; represents the initial reputation value; Indicates the number of false judgments; Indicates the historical in / out times in the historical in / out records.

10. A data audit intelligent learning system based on big data, characterized in that: include: Signal analysis module, used to obtain and analyze project audit signals and determine audit content; A record analysis module is used to retrieve material in and out records according to the audit content; analyze the material in and out records to determine whether there is an abnormal return; The content sending module is used to determine the review content according to the suspicious records corresponding to the abnormal return, if any, and send the review content to the reviewer.

Images