Audit doubtful point identification method and device based on data analysis, equipment and medium

Abstract

The invention discloses an audit doubtful point identification method, device and equipment based on data analysis and a medium, and relates to the field of audit doubtful point identification, and the method comprises the steps: carrying out the preprocessing of to-be-audited data, and determining the preprocessed audit data; identifying differences between the preprocessed audit data, and determining abnormal data; predicting a data change trend according to the abnormal data; identifying abnormal data deviating from the data change trend and an abnormal mode, and determining potential auditing doubtful points; determining risk levels according to the potential auditing doubtful points, performing priority ranking on the risk levels, and allocating auditing resources; and auditing the corresponding potential auditing doubtful points based on the allocated auditing resources, and generating an auditing report according to the auditing result, so that the auditing work efficiency can be remarkably improved, and optimal configuration of the auditing resources is realized.

Description

Technical Field

[0001] The present application relates to the field of audit doubt point identification, and in particular to a method, device, equipment and medium for audit doubt point identification based on data analysis. Background Art

[0002] In today's complex and ever-changing business environment, auditing is an important means to ensure the financial health of enterprises and promote compliance operations, and its importance is increasing day by day. With the expansion of enterprise scale and the increase in business complexity, traditional auditing methods can no longer meet the needs of efficiency and accuracy. Existing auditing technologies mainly rely on manual review of financial statements, verification of accounts, and sampling surveys. These methods have obvious technical bottlenecks in processing massive data and identifying hidden risks.

[0003] Specifically, existing auditing technologies have the following main problems:

[0004] Inefficiency: Faced with huge amounts of data, manual auditing requires a lot of time and manpower costs, and it is difficult to complete a comprehensive and detailed review in a short period of time. This leads to a longer audit cycle and an inability to detect and respond to potential risks in a timely manner.

[0005] Insufficient accuracy: Manual audits are limited by the professional level and experience of auditors, and are prone to misjudgment or omission. Especially when dealing with complex data and hidden risks, the accuracy of manual audits is often difficult to guarantee.

[0006] Uneven allocation of resources: Traditional audit methods often adopt a "one-size-fits-all" approach, conducting indiscriminate reviews of all audit objects, resulting in excessive investment of audit resources in low-risk areas and insufficient resources in high-risk areas. Summary of the invention

[0007] The purpose of this application is to provide a method, device, equipment and medium for identifying audit doubts based on data analysis to solve the problems of low efficiency, low accuracy and uneven resource allocation of existing manual audit methods.

[0008] To achieve the above objectives, this application provides the following solutions:

[0009] In a first aspect, the present application provides a method for identifying audit doubts based on data analysis, comprising:

[0010] Preprocessing the audit data to be audited and determining the preprocessed audit data; the audit data to be audited includes sales orders, inventory change records, financial statements, bank statements, employee salary details, purchase contract records, supplier invoices and tax returns;

[0011] Identifying differences between the pre-processed audit data and determining abnormal data;

[0012] Predicting data change trends based on the abnormal data;

[0013] Identify abnormal data and abnormal patterns that deviate from the data change trend and determine potential audit doubts;

[0014] Determine the risk level based on the potential audit doubts, prioritize the risk levels, and allocate audit resources;

[0015] Audit the corresponding potential audit doubts based on the allocated audit resources, and generate an audit report based on the audit results.

[0016] In a second aspect, the present application provides an audit doubt identification device based on data analysis, comprising:

[0017] A preprocessing module is used to preprocess the audit data and determine the preprocessed audit data; the audit data includes sales orders, inventory change records, financial statements, bank statements, employee salary details, purchase contract records, supplier invoices and tax returns;

[0018] An abnormal data determination module, used to identify the differences between the pre-processed audit data and determine the abnormal data;

[0019] A data change trend prediction module, used to predict the data change trend based on the abnormal data;

[0020] A potential audit doubt point determination module is used to identify abnormal data and abnormal patterns that deviate from the data change trend and determine potential audit doubt points;

[0021] An audit resource allocation module, used to determine the risk level according to the potential audit doubts, prioritize the risk levels, and allocate audit resources;

[0022] The audit report generation module is used to audit the corresponding potential audit doubts based on the allocated audit resources, and generate an audit report based on the audit results.

[0023] In a third aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any of the above-mentioned audit doubt identification methods based on data analysis.

[0024] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-mentioned audit doubt identification methods based on data analysis.

[0025] According to the specific embodiments provided by this application, this application discloses the following technical effects: through automated and intelligent data analysis tools, the audit data is pre-processed, abnormal data is determined, and data change trends are predicted to determine potential audit doubts, greatly reducing the time for manual screening and comparison of data. Especially when faced with massive amounts of data, automated processing can quickly lock in potential audit doubts, significantly improve audit work efficiency, reduce manual intervention, and speed up the audit process.

[0026] In addition, this application determines the risk level based on potential audit doubts and prioritizes the risk level to allocate audit resources, so that the audit team can give priority to high-risk, high-impact potential audit doubts, avoid ineffective allocation of audit resources and waste of time, and achieve optimal configuration of audit resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0028] Figure 1 This is a flow chart of a method for identifying audit doubts based on data analysis in one embodiment of the present application;

[0029] Figure 2 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only 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.

[0031] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0032] like Figure 1 As shown, the embodiment of the present application provides a method for identifying audit doubts based on data analysis, including:

[0033] Step 101: pre-process the audit data to determine the pre-processed audit data; the audit data includes sales orders, inventory change records, financial statements, bank statements, employee salary details, purchase contract records, supplier invoices and tax returns.

[0034] Step 102: Identify the differences between the pre-processed audit data and determine abnormal data.

[0035] Step 103: predicting data change trends based on the abnormal data.

[0036] Step 104: Identify abnormal data and abnormal patterns that deviate from the data change trend, and determine potential audit doubts.

[0037] Step 105: Determine the risk level according to the potential audit doubts, prioritize the risk levels, and allocate audit resources.

[0038] Step 106: Audit the corresponding potential audit doubts based on the allocated audit resources, and generate an audit report based on the audit results.

[0039] In an exemplary embodiment, step 101 may also be replaced by steps 201 and 202.

[0040] Step 201: Clean the audit data from different sources and determine the pre-processed audit data; the sources include various business systems and external data sources.

[0041] Furthermore, before cleaning the data, it also includes collecting the data to be audited, where the data to be audited is collected from the audited unit's various business systems (such as ERP, CRM, financial systems, etc.) and external data sources (such as banks, taxes, etc.), including sales orders, inventory change records, financial statements, bank statements, employee salary details, purchase contract records, supplier invoices, tax return forms and other data.

[0042] The purpose of data cleaning is to remove noise, errors, duplicates and inconsistent data. In addition, it also includes missing value processing, outlier detection and correction, data type conversion and other processing.

[0043] Step 202: Integrate the pre-processed audit data from different sources into the same database.

[0044] Furthermore, data from different sources are integrated into a unified data warehouse or data lake to facilitate subsequent analysis and processing.

[0045] In an exemplary embodiment, step 102 may also be replaced by steps 301 to 304 .

[0046] Step 301: Based on the preprocessed audit data, the audit data to be compared is determined according to the audit objectives and scope; the audit data to be compared is audit data of the same period (such as data of the same month of this year and last year), audit data of the same industry (such as comparison with data of other companies in the industry), audit data of different departments, audit data of different business units (such as comparison of data of different departments or business units) and audit data of different regions (such as comparison of data of different regions or markets).

[0047] Step 302: Standardize the audit data to be compared to determine standardized audit data.

[0048] Furthermore, data standardization and preparation specifically include the following steps.

[0049] The data standardization steps specifically include:

[0050] Normalization: If the data has different magnitudes, normalization methods such as the minimum-maximum scaling method can be used to convert the data to the [0,1] interval.

[0051] Standardization: If the data distribution is close to normal distribution, you can use standardization methods, such as z-score standardization, to convert the data into a standard normal distribution with a mean of 0 and a standard deviation of 1.

[0052] Discretization processing: For certain types of continuous variables, discretization processing can be performed, such as binning, to divide the continuous values ​​into several interval segments.

[0053] The data preparation steps include:

[0054] Ensure data consistency: Check data format consistency, such as date format, currency unit, etc.

[0055] Handling missing values: For missing data, you can use interpolation methods such as filling with mean, median or prediction.

[0056] Data transformation: If necessary, some variables can be transformed into a form more suitable for comparison, such as logarithmic transformation.

[0057] Step 303: Compare and count the differences between the standardized audit data to generate statistical analysis results.

[0058] Furthermore, statistical methods or data mining techniques are applied to conduct comparative analysis and identify significant differences or deviations between the data.

[0059] Compute differences, ratios, or other statistics to measure differences between different data objects.

[0060] Statistical tests (such as t-test, ANOVA, etc.) were used to determine whether the differences were statistically significant.

[0061] Step 304: Based on the statistical analysis results, according to the data characteristics and analyzing the audit objectives, an outlier detection method is selected to determine the abnormal data; the data characteristics include data type, data size, data distribution, data quality, data update frequency, data correlation and data sparsity; the analysis process of the audit objectives includes identifying potential fraud, evaluating the effectiveness of internal control, and discovering inconsistencies in financial statements; the outlier detection methods include outlier detection methods based on statistical distribution, outlier detection methods based on distance, and outlier detection methods based on density.

[0062] Further, step 304 includes data exploration and understanding, outlier detection, and implementation detection.

[0063] Among them, data exploration and understanding include descriptive statistical analysis and data visualization.

[0064] Descriptive statistical analysis: Perform basic statistical descriptions, such as calculating the mean, median, standard deviation, maximum value, minimum value, etc., to understand the overall distribution and characteristics of the data.

[0065] Data visualization: Use charts, graphs, and other methods to display the results of statistical analysis, helping users to intuitively understand the trends, patterns, and relationships of the data.

[0066] Outlier detection includes selecting detection methods: Select appropriate outlier detection methods based on data characteristics and analysis audit objectives.

[0067] Among them, data type: data can be numerical, categorical, time series, etc.

[0068] Data size: The size of the dataset, including the number of records and the number of features.

[0069] Data distribution: The distribution characteristics of data, such as whether it conforms to a normal distribution or another specific distribution.

[0070] Data quality: data completeness, accuracy, consistency, etc.

[0071] Data update frequency: the frequency of data updates, such as real-time updates, daily updates, monthly updates, etc.

[0072] Data correlation: The correlation between data, including the degree of correlation between different data items.

[0073] Data sparsity: The proportion of missing values ​​or zero values ​​in a dataset.

[0074] Common methods include:

[0075] Outlier detection methods based on statistical distribution (such as Z-score detection): suitable for situations where the data roughly conforms to the normal distribution.

[0076] Distance-based outlier detection methods (such as K-nearest neighbor algorithm): suitable for situations where the amount of data is moderate and the dimension is not too high.

[0077] Density-based outlier detection methods (such as the local outlier factor algorithm): suitable for data with complex distribution patterns, especially when the data set contains multiple clusters.

[0078] Perform detection: Apply the selected method to perform outlier detection on the data and mark potential abnormal points or abnormal clusters.

[0079] Verify outliers: Manually verify detected outliers to confirm whether they truly represent anomalies or are simply caused by data errors, noise, etc.

[0080] In an exemplary embodiment, step 102 further includes: verifying the abnormal data and analyzing the causes of the abnormal data; conducting risk assessment based on the causes, formulating solutions, and compiling them into a preliminary report; the preliminary report includes the abnormal data discovery process, cause analysis, risk assessment and solutions.

[0081] Furthermore, the abnormal data is analyzed and interpreted, including in-depth analysis, risk assessment, and formulation of countermeasures.

[0082] In-depth analysis: Conduct in-depth analysis of confirmed abnormal data to explore the causes and background of its occurrence. This may involve communication and confirmation with relevant personnel of the audited unit.

[0083] The analysis process is as follows:

[0084] 1) Data review and verification:

[0085] Data review: Re-review confirmed abnormal data to ensure the accuracy and completeness of the data.

[0086] Data verification: Use additional data sources or methods to verify the authenticity of abnormal data.

[0087] 2) Root Cause Analysis:

[0088] Background investigation: Investigate the time, location and relevant business background of abnormal data.

[0089] Interviews with relevant personnel: Communicate with relevant personnel of the audited unit to understand the causes of abnormal data.

[0090] Business process review: Review relevant business processes and operations to identify factors that may lead to abnormal data.

[0091] 3) Impact Assessment:

[0092] Business impact analysis: Evaluate the impact of abnormal data on business operations.

[0093] Financial impact analysis: Evaluate the impact of abnormal data on financial status.

[0094] Compliance assessment: Evaluate whether abnormal data violates relevant regulations or policies.

[0095] 4) Solution formulation:

[0096] Improvement measures: Develop specific improvement measures based on the analysis results.

[0097] Preventive measures: Develop measures to prevent similar abnormal data from happening again.

[0098] Remediation: For abnormal data that has already been generated, formulate remediation measures to mitigate the negative impact.

[0099] 5) Report preparation and communication:

[0100] Report preparation: Organize the results of the in-depth analysis into a preliminary report, which includes the discovery process of abnormal data, cause analysis, risk assessment and recommended improvement measures.

[0101] Feedback communication: Communicate analysis results with the audited unit, put forward improvement suggestions, and track subsequent rectification.

[0102] Risk assessment: Assess the impact and risks that abnormal data may have on business operations, financial status, compliance, etc.

[0103] The evaluation process is as follows:

[0104] 1) Identification and assessment:

[0105] Direct Identification: Directly identify potential suspicious points during the audit process.

[0106] Risk assessment: Assess the risk level of suspicious points, considering the impact and the likelihood of occurrence.

[0107] 2) Prioritize:

[0108] Based on the risk assessment results: prioritize audit issues, with high-risk, high-impact or high-urgency issues at the front.

[0109] 3) Develop an action plan:

[0110] Detailed Audit Plan: Develop a detailed audit plan for each priority issue, including audit steps, timeline, and expected outcomes.

[0111] Execute Audits: Perform audits as planned while continuously monitoring progress and results to ensure effectiveness and timeliness of audit work.

[0112] Formulate countermeasures: formulate corresponding countermeasures and suggestions based on the analysis results to solve or alleviate the problems represented by the abnormal data. The analysis results mainly include an overview of the abnormal data, root cause analysis, and risk assessment. These contents provide an important basis for the subsequent formulation of countermeasures. When formulating countermeasures, corresponding improvement measures, preventive measures, and remedial measures will be proposed based on these analysis results to solve or alleviate the problems represented by the abnormal data.

[0113] 1) Business operation issues:

[0114] Irregular process: When operations that do not comply with regulations are found in the business process.

[0115] Operational error: When it is discovered that an employee’s operational error has caused data anomalies.

[0116] 2) Technical issues:

[0117] System failure: When data anomalies caused by system failure are discovered.

[0118] Data quality issues: When data entry errors or missing data are discovered.

[0119] 3) Compliance issues:

[0120] Non-compliance with regulations: When data or business operations are found to be non-compliant with laws and regulations.

[0121] Non-compliance with internal regulations: When non-compliance with internal operating procedures is found.

[0122] 4) Risk management issues:

[0123] Inadequate risk control: When risk management measures are found to be inadequate.

[0124] Inaccurate risk assessment: When it is found that the risk assessment results do not match the actual situation.

[0125] 5) Financial issues:

[0126] Inaccurate financial statements: When false or misleading information is discovered in financial statements.

[0127] Financial Fraud: When financial fraud is suspected.

[0128] The preliminary report focuses on the initial verification and risk assessment of abnormal data, providing the necessary prerequisites and basic information for the subsequent generation of the formal audit report.

[0129] Based on the preliminary report, conduct more in-depth risk assessment and audit verification of potential audit issues to generate a formal audit report.

[0130] In an exemplary embodiment, step 103 may also be replaced by step 401 .

[0131] Step 401: Based on historical audit data and historical abnormal data, the data to be audited and the abnormal data are input into a time series prediction model to predict data change trends.

[0132] Furthermore, time series prediction models such as ARIMA and LSTM are used to predict the trend of time series data such as financial data and business indicators. Among them, financial data and business indicators constitute the basis of time series data: 1. The financial data of an enterprise is an important part of time series data, such as monthly revenue, quarterly profit, etc.; 2. Business indicators can also constitute time series data, such as monthly sales, annual customer growth rate, etc.

[0133] Use regression models (such as linear regression, decision tree regression) or deep learning models, combined with historical data, to predict possible future changes in values ​​or states.

[0134] By comparing actual data with predicted data, outliers or abnormal patterns that deviate from the predicted trend can be identified.

[0135] In an exemplary embodiment, step 104 may also be replaced by steps 501 to 503 .

[0136] Step 501: Determine an abnormal data threshold based on business rules, industry standards or historical audit data.

[0137] Step 502: The audit data in the data change trend that exceeds or falls below the abnormal data threshold is regarded as abnormal data and determined as a potential audit doubt.

[0138] Step 503: Identify abnormal patterns in the data change trend that deviate from normal business patterns, and determine abnormal data corresponding to the abnormal patterns as potential audit doubts.

[0139] Further, the method for identifying abnormal patterns includes:

[0140] 1) Establish a baseline: Use historical data to establish a baseline of normal business patterns, such as by calculating the mean and standard deviation.

[0141] 2) Identify anomalies: Use machine learning algorithms such as Isolation Forest to identify data points that deviate from normal patterns.

[0142] 3) Data visualization: Use time series graphs, box plots and other graphics to visually display data and help identify abnormal patterns.

[0143] 4) Business rules: Define some business rules or thresholds. When data violates these rules or exceeds the threshold, it is considered to deviate from the normal pattern.

[0144] 5) Manual review: Manual review of abnormal patterns identified by the algorithm to confirm whether they are real deviations.

[0145] In an exemplary embodiment, step 105 may also be replaced by steps 601 to 603 .

[0146] Step 601: Use a risk assessment method to assess the potential audit doubts and determine the risk level; the risk assessment method includes a machine learning algorithm, a statistical analysis method, a rule system constructed using expert knowledge and experience, and a risk matrix.

[0147] Step 602: Prioritize the risk levels.

[0148] Step 603: Allocate audit resources according to the sorted priorities.

[0149] Further, step 105 includes identification and assessment, prioritization, and development of an action plan.

[0150] Among them, identification and assessment: directly identify potential doubts in the audit process, and at the same time assess their risk levels, considering the degree of impact and the possibility of occurrence.

[0151] The evaluation method is as follows:

[0152] 1) Machine Learning Algorithms:

[0153] Classification algorithms: such as logistic regression, decision tree, random forest, etc., are used to predict which risk category a potential audit point belongs to.

[0154] Clustering algorithms: such as K-means and hierarchical clustering, are used to identify groups of suspicious points with similar risk characteristics.

[0155] 2) Statistical analysis methods:

[0156] Hypothesis testing: such as t-test, ANOVA, etc., used to evaluate the statistical significance between certain factors and doubts.

[0157] Regression analysis: such as linear regression and multiple regression, used to quantify the impact of different factors on the risk level of suspicious points.

[0158] 3) Expert system:

[0159] Use expert knowledge and experience to build a rule system to provide additional basis and support for the evaluation results.

[0160] 4) Risk Matrix:

[0161] Construct a risk matrix and classify suspicious points according to the two dimensions of impact and possibility of occurrence, so as to facilitate the intuitive identification of high-risk areas.

[0162] Determine priorities: Based on the risk assessment results, directly prioritize audit issues, with high-risk, high-impact or high-urgency issues being prioritized first.

[0163] Develop an action plan: Develop a detailed audit plan for each priority issue, including audit steps, timeline, and expected results. Perform the audit as planned while continuously monitoring progress and results to ensure the effectiveness and timeliness of the audit work.

[0164] In an exemplary embodiment, step 106 includes:

[0165] Verification of suspicious points: Further data verification and on-site inspection are carried out on the identified suspicious points to confirm their authenticity and accuracy.

[0166] Report preparation: Organize the verified doubts into an audit report, describing in detail the discovery process of the doubts, risk assessment results and recommended audit measures.

[0167] Feedback communication: Communicate the audit results with the audited unit, put forward improvement suggestions, and track the subsequent rectification.

[0168] Audit reporting focuses on more detailed audit work and report generation based on these preliminary findings.

[0169] Based on the same inventive concept, the embodiment of the present application also provides a data analysis-based audit doubt identification device for implementing the data analysis-based audit doubt identification method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more data analysis-based audit doubt identification device embodiments provided below can refer to the limitations of the data analysis-based audit doubt identification method above, and will not be repeated here.

[0170] In an exemplary embodiment, the present application provides an audit doubt point identification device based on data analysis, characterized in that the audit doubt point identification device based on data analysis includes:

[0171] The preprocessing module is used to preprocess the audit data to be audited and determine the preprocessed audit data; the audit data to be audited includes sales orders, inventory change records, financial statements, bank statements, employee salary details, purchase contract records, supplier invoices and tax returns.

[0172] The abnormal data determination module is used to identify the differences between the pre-processed audit data and determine the abnormal data.

[0173] The data change trend prediction module is used to predict the data change trend based on the abnormal data.

[0174] The potential audit doubt point determination module is used to identify abnormal data and abnormal patterns that deviate from the data change trend and determine potential audit doubt points.

[0175] The audit resource allocation module is used to determine the risk level according to the potential audit doubts, prioritize the risk levels, and allocate audit resources.

[0176] The audit report generation module is used to audit the corresponding potential audit doubts based on the allocated audit resources, and generate an audit report based on the audit results.

[0177] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 2 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store audit doubt identification data based on data analysis. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for identifying audit doubts based on data analysis is implemented.

[0178] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0179] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0180] 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 regulations.

[0181] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ReadOnlyMemory, ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (Magnetoresistive RandomAccess Memory, MRAM), ferroelectric random access memory (Ferroelectric RandomAccess Memory, FRAM), phase change memory (Phase Change Memory, PCM), graphene memory, etc. Volatile memory can include random access memory (RandomAccess Memory, RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0182] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0183] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0184] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for identifying audit doubts based on data analysis, characterized in that: The audit doubt identification method based on data analysis includes: Preprocessing the audit data to be audited and determining the preprocessed audit data; the audit data to be audited includes sales orders, inventory change records, financial statements, bank statements, employee salary details, purchase contract records, supplier invoices and tax returns; Identifying differences between the pre-processed audit data and determining abnormal data; Predicting data change trends based on the abnormal data; Identify abnormal data and abnormal patterns that deviate from the data change trend and determine potential audit doubts; Determine the risk level based on the potential audit doubts, prioritize the risk levels, and allocate audit resources; Audit the corresponding potential audit doubts based on the allocated audit resources, and generate an audit report based on the audit results.

2. The audit doubt identification method based on data analysis according to claim 1 is characterized in that: Preprocess the audit data and determine the preprocessed audit data, including: Clean the audit data from different sources and determine the pre-processed audit data; the sources include various business systems and external data sources; Consolidate pre-processed audit data from different sources into a single database.

3. The audit doubt identification method based on data analysis according to claim 1 is characterized in that: Identify the differences between the pre-processed audit data and determine abnormal data, including: Based on the pre-processed audit data, the audit data to be compared is determined according to the audit objectives and scope; the audit data to be compared is audit data of the same period, audit data of the same industry, audit data of different departments, audit data of different business units, and audit data of different regions; Performing standardization processing on the audit data to be compared to determine standardized audit data; Compare and count the differences between standardized audit data to generate statistical analysis results; Based on the statistical analysis results, according to the data characteristics and analysis of the audit objectives, an outlier detection method is selected to determine the abnormal data; the data characteristics include data type, data size, data distribution, data quality, data update frequency, data correlation and data sparsity; the analysis process of the audit objectives includes identifying potential fraud, evaluating the effectiveness of internal control, and discovering inconsistencies in financial statements; the outlier detection methods include outlier detection methods based on statistical distribution, outlier detection methods based on distance, and outlier detection methods based on density.

4. The audit doubt identification method based on data analysis according to claim 1 is characterized in that: Identifying differences between the pre-processed audit data and determining abnormal data, and then: Verify the abnormal data and analyze the causes of the abnormal data; Conduct risk assessment based on the causes, develop solutions, and compile them into a preliminary report; the preliminary report includes the abnormal data discovery process, cause analysis, risk assessment, and solutions.

5. The audit doubt identification method based on data analysis according to claim 1 is characterized in that: Predicting data change trends based on the abnormal data specifically includes: Based on historical audit data and historical abnormal data, the data to be audited and the abnormal data are input into a time series prediction model to predict the data change trend.

6. The audit doubt identification method based on data analysis according to claim 1 is characterized in that: Identify abnormal data and abnormal patterns that deviate from the data change trend and determine potential audit doubts, including: Determine abnormal data thresholds based on business rules, industry standards, or historical audit data; The audit data that exceeds or falls below the abnormal data threshold in the data change trend is regarded as abnormal data and determined as a potential audit doubt; Identify abnormal patterns in the data change trend that deviate from normal business patterns, and determine abnormal data corresponding to the abnormal patterns as potential audit doubts.

7. The audit doubt identification method based on data analysis according to claim 1 is characterized in that: Determine the risk level based on the potential audit doubts, prioritize the risk levels, and allocate audit resources, including: Using risk assessment methods to assess the potential audit doubts and determine the risk level; the risk assessment methods include machine learning algorithms, statistical analysis methods, rule systems built using expert knowledge and experience, and risk matrices; prioritizing said risk levels; Allocate audit resources according to sorted priorities.

8. An audit doubt identification device based on data analysis, characterized in that: The audit doubtful point identification device based on data analysis includes: A preprocessing module is used to preprocess the audit data and determine the preprocessed audit data; the audit data includes sales orders, inventory change records, financial statements, bank statements, employee salary details, purchase contract records, supplier invoices and tax returns; An abnormal data determination module, used to identify the differences between the pre-processed audit data and determine the abnormal data; A data change trend prediction module, used to predict the data change trend based on the abnormal data; A potential audit doubt point determination module is used to identify abnormal data and abnormal patterns that deviate from the data change trend and determine potential audit doubt points; An audit resource allocation module, used to determine the risk level according to the potential audit doubts, prioritize the risk levels, and allocate audit resources; The audit report generation module is used to audit the corresponding potential audit doubts based on the allocated audit resources, and generate an audit report based on the audit results.

9. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the audit doubt identification method based on data analysis as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for identifying audit doubts based on data analysis described in any one of claims 1 to 7 is implemented.

Images