Complex business project auditing system based on multi-dimensional historical data

Abstract

The application discloses a complex business project auditing system based on multidimensional historical data and belongs to the technical field of business project auditing, comprising: a data acquisition preprocessing module, which is used for acquiring multi-source data of business project auditing and preprocessing the same; a risk assessment anomaly detection module, which is used for identifying abnormal patterns and potential risks in auditing tasks and determining risk assessment results of business project auditing; a task sorting intelligent distribution module, which is used for prioritizing tasks and intelligently distributing the same; and a real-time monitoring dynamic adjustment module, which is used for dynamically adjusting a task distribution scheduling scheme according to real-time monitoring data feedback. The application solves the problem that the existing system cannot effectively identify abnormal patterns and potential risks in auditing tasks, cannot optimize resource distribution and leads to low auditing efficiency. The application can effectively identify abnormal patterns and potential risks in auditing tasks, can optimize resource distribution and reduce risks and can improve the efficiency of complex business project auditing.

Description

Technical Field

[0001] This invention relates to the field of business project review technology, specifically a complex business project review system based on multi-dimensional historical data. Background Technology

[0002] Business project review refers to a comprehensive review of a project's activities based on relevant laws and regulations, corporate policies, and management standards to determine whether the project meets its expected goals, identify potential problems, and propose improvement suggestions.

[0003] The importance of business project review is reflected in the following aspects: 1) Ensuring compliance: ensuring that the project complies with national laws and regulations and corporate policies; 2) Optimizing resource allocation: allocating resources reasonably and avoiding waste; 3) Improving project efficiency: improving project execution efficiency by identifying and correcting problems; 4) Supporting decision-making: providing management with scientific decision-making basis.

[0004] The existing review process for complex business projects cannot effectively identify abnormal patterns and potential risks in the review tasks, nor can it optimize resource allocation and reduce risks, resulting in low efficiency in the review of complex business projects. Summary of the Invention

[0005] The purpose of this invention is to provide a complex business project review system based on multi-dimensional historical data, which can effectively identify abnormal patterns and potential risks in review tasks, optimize resource allocation and reduce risks, improve the efficiency of complex business project review, and solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A complex business project review system based on multi-dimensional historical data includes:

[0008] The data acquisition and preprocessing module is used to collect multi-source data on business project review from internal enterprise systems and external data sources, and to preprocess the multi-source data on business project review to determine the characteristic data of business project review.

[0009] The risk assessment anomaly detection module is used to build a business project audit risk assessment model using machine learning algorithms, identify abnormal patterns and potential risks in audit tasks, and determine the business project audit risk assessment results.

[0010] The task sorting and intelligent allocation module is used to prioritize tasks and intelligently allocate tasks to ensure efficient use of resources.

[0011] The real-time monitoring and dynamic adjustment module is used to monitor the task execution status in real time and dynamically adjust the task allocation and scheduling scheme based on the feedback of real-time monitoring data to deal with emergencies or resource changes.

[0012] The data visualization decision support module is used to display the review progress, risk distribution, and resource usage using data visualization tools, which facilitates management decision-making.

[0013] Preferably, to identify abnormal patterns and potential risks in audit tasks and determine the results of business project audit risk assessments, the following operations are performed:

[0014] Deploy the business project review risk assessment model and place it in the actual business project review risk assessment environment;

[0015] The characteristic data of business project audits are input into the business project audit risk assessment model. The model is used to analyze the characteristic data of business project audits and automatically identify abnormal patterns and potential risks in business project audit tasks. The risk level of the task is assessed and the business project audit risk assessment result is determined.

[0016] Preferably, assess the task risk level and perform the following operations:

[0017] Extract the number of associated business nodes and the probability of anomalies corresponding to the associated business nodes from the abnormal patterns obtained by the business project audit risk assessment model analysis;

[0018] Retrieve the weight value corresponding to each associated business node;

[0019] By performing a weighted average of the weight value and the anomaly probability of each associated business node, a comprehensive parameter for the anomaly probability of associated businesses can be obtained. Where 'a' represents the number of associated business nodes; w yi P represents the weight value corresponding to the i-th associated business node; yi This represents the probability of an anomaly occurring for the i-th associated business node;

[0020] Extract the number of business process steps involved in the potential risks obtained from the business project review risk assessment model analysis, and the probability of risk occurrence corresponding to each business process step;

[0021] Retrieve the weight value corresponding to each business process step;

[0022] By performing a weighted average of the weight value corresponding to each business process step and the probability of risk occurrence corresponding to each business process step, a comprehensive parameter of business process risk probability is obtained. Where b represents the number of steps in the business process; w i P represents the weight value corresponding to the i-th business process step; xiThis represents the probability of risk occurring at the i-th business process stage;

[0023] The comprehensive parameters of the probability of anomalies in related businesses and the comprehensive parameters of the probability of risks in business processes are summed to obtain the initial risk score corresponding to the audit feature data of business projects.

[0024] Compare the initial risk score corresponding to the business project review feature data with the preset risk score threshold;

[0025] When the initial risk score corresponding to the business project review feature data does not exceed the preset risk score threshold, the business project review task is determined to be of low risk level.

[0026] When the initial risk score corresponding to the business project review feature data exceeds the preset risk score threshold, the business project review feature data will be classified into a secondary risk level.

[0027] Preferably, when the initial risk score corresponding to the business project review feature data exceeds a preset risk score threshold, a secondary risk level classification is performed on the business project review feature data, and the following operations are performed:

[0028] Retrieve the number of associated business nodes and the probability of anomalies occurring for each associated business node;

[0029] The standard deviation of the node anomaly probability is obtained based on the number of associated business nodes and the anomaly occurrence probability corresponding to the associated business nodes.

[0030] Retrieve the probability of risk occurrence for each step of the business process;

[0031] The probability of risk occurrence for each business process step is compared with a preset risk probability threshold, where the preset risk probability threshold is set to 0.4.

[0032] Select business process steps whose probability of occurrence is not lower than the preset risk probability threshold as target business process steps;

[0033] Determine whether there are non-target business process steps interspersed between the target business process steps, and obtain each group of target business process steps that are interspersed with non-target business process steps;

[0034] Set a risk impact coefficient for each group of target business process steps that includes non-target business process steps;

[0035] The secondary risk assessment parameters are obtained by combining the risk impact coefficient with the standard deviation of the node anomaly probability.

[0036] The secondary risk assessment parameters are compared with preset risk assessment parameter thresholds;

[0037] When the secondary risk assessment parameter is lower than the preset risk assessment parameter threshold, the business project review task is determined to be of medium risk level.

[0038] When the secondary risk assessment parameter is not lower than the preset risk assessment parameter threshold, the business project review task is determined to be of a high-risk level.

[0039] Preferably, tasks are prioritized and intelligently assigned, and the following operations are performed:

[0040] Based on the risk assessment results of business project audits, the audit tasks of complex business projects are prioritized, with the priority determined by combining risk level factors, task complexity factors, and urgency factors.

[0041] Tasks are dynamically allocated based on task complexity, reviewer capabilities and work status, and task priority to optimize resource allocation.

[0042] Preferably, the multi-source data for business project review is preprocessed by performing the following operations:

[0043] Clean the multi-source data for business project review, remove noise that is irrelevant to the review of complex business projects, and handle missing and outlier values ​​in the multi-source data for business project review.

[0044] The multi-source data for business project audits is transformed into a unified data format, eliminating dimensional differences and forming standardized multi-source data for business project audits.

[0045] Preferably, in addition to preprocessing multi-source data for business project audits, the following operations are performed:

[0046] Integrate multi-source data for business project review, merge multi-source data for business project review from different sources into a unified data view, and store the integrated multi-source data for business project review.

[0047] Feature extraction is performed on multi-source data of business project review to extract features related to complex business project review and determine the feature data of business project review.

[0048] Preferably, a business project review risk assessment model is constructed using machine learning algorithms, and the following operations are performed:

[0049] Collect historical data on business project review and divide the historical data into training set and test set.

[0050] Based on machine learning algorithms, a training set is used to train the machine learning model, enabling the machine learning model to autonomously learn complex business project review behaviors from the training set, identify abnormal patterns and potential risks in the review tasks, and determine a business project review risk assessment model based on machine learning.

[0051] Input the test set into the business project review risk assessment model, test the business project review risk assessment model based on the test set, evaluate the performance of the business project review risk assessment model, and determine the model test evaluation results;

[0052] Based on the model testing and evaluation results, the business project audit risk assessment model is adjusted and optimized to determine the optimal business project audit risk assessment model.

[0053] Preferably, to evaluate the performance of the business project audit risk assessment model, the following operations are performed:

[0054] The business project review risk assessment model was tested using a test set. The accuracy, precision, and recall were used to determine whether the business project review risk assessment model could achieve the expected effect of identifying abnormal patterns and potential risks in the review task.

[0055] When the business project audit risk assessment model fails to achieve the expected effect of identifying abnormal patterns and potential risks in the audit task, the parameters of the business project audit risk assessment model are adjusted and optimized until the business project audit risk assessment model can achieve the expected effect of identifying abnormal patterns and potential risks in the audit task, thereby determining the optimal business project audit risk assessment model.

[0056] Preferably, data visualization tools are used to display the review progress, risk distribution, and resource usage, and the following operations are performed:

[0057] Data visualization tools are used to visually display the review progress, risk distribution, and resource usage through various charts and interactive dashboards;

[0058] Based on the review progress, risk distribution, and resource usage, we regularly generate review reports for complex business projects. These reports provide management with detailed information on task allocation and resource utilization, helping them quickly identify risk hotspots, optimize scheduling plans, and provide data support for strategic decision-making.

[0059] Compared with the prior art, the beneficial effects of the present invention are:

[0060] 1. This invention collects multi-source data on business project audits from internal enterprise systems and external data sources, preprocesses this data to determine characteristic data, constructs a business project audit risk assessment model using machine learning algorithms, identifies abnormal patterns and potential risks in audit tasks, and determines the risk assessment results. Based on these risk assessment results, and considering risk level, task complexity, and urgency factors, the invention prioritizes complex business project audit tasks, determines task priorities, and dynamically allocates tasks based on task complexity, auditor capabilities, and work status, optimizing resource allocation and ensuring efficient resource utilization. This invention effectively identifies abnormal patterns and potential risks in audit tasks, optimizes resource allocation, reduces risks, and improves the efficiency of complex business project audits.

[0061] 2. This invention monitors task execution status in real time and dynamically adjusts task allocation and scheduling schemes based on real-time monitoring data feedback to cope with emergencies or resource changes. It also employs data visualization tools to intuitively display review progress, risk distribution, and resource usage through various charts and interactive dashboards. Based on the review progress, risk distribution, and resource usage, it regularly generates review reports for complex business projects to provide management with detailed information on task allocation and resource utilization, helping managers quickly identify risk hotspots, optimize scheduling schemes, and provide data support for strategic decision-making. Attached Figure Description

[0062] Figure 1 This is a module diagram of the complex business project review system based on multi-dimensional historical data according to the present invention. Detailed Implementation

[0063] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0064] To address the issue of inefficient complex business project review processes, which fail to effectively identify abnormal patterns and potential risks, optimize resource allocation, and mitigate risks, please refer to [link to relevant documentation]. Figure 1 This embodiment provides the following technical solution:

[0065] A complex business project review system based on multi-dimensional historical data includes:

[0066] The data acquisition and preprocessing module is used to collect multi-source data on business project audits from internal enterprise systems and external data sources, and to preprocess the multi-source data on business project audits to determine the characteristic data of business project audits.

[0067] It should be noted that internal enterprise systems such as ERP and CRM, and external data sources such as industry databases and social media.

[0068] In this embodiment, the multi-source data for business project review is preprocessed by performing the following operations:

[0069] Cleaning multi-source data for business project audits, removing noise unrelated to complex business project audits, and processing missing and outlier values ​​in the multi-source data can improve the data quality of multi-source data for business project audits.

[0070] The multi-source data for business project audits is transformed into a unified data format, eliminating dimensional differences and forming standardized multi-source data for business project audits. This improves data usability and facilitates subsequent data analysis.

[0071] In this embodiment, the preprocessing of multi-source data for business project review is performed, and the following operations are also performed:

[0072] Integrate multi-source data for business project review, merge multi-source data for business project review from different sources into a unified data view, and store the integrated multi-source data for business project review.

[0073] Feature extraction is performed on multi-source data of business project review. Features related to complex business project review are extracted from the multi-source data of business project review to determine the characteristic data of business project review. This facilitates the subsequent analysis of the characteristic data of business project review, so as to identify abnormal patterns and potential risks in the review task.

[0074] The risk assessment anomaly detection module is used to build a business project audit risk assessment model using machine learning algorithms, identify abnormal patterns and potential risks in audit tasks, and determine the business project audit risk assessment results.

[0075] In this embodiment, a business project review risk assessment model is constructed using machine learning algorithms, and the following operations are performed:

[0076] Collect historical data on business project review and divide the historical data into training set and test set.

[0077] Based on machine learning algorithms, a training set is used to train the machine learning model, enabling the machine learning model to autonomously learn complex business project review behaviors from the training set, identify abnormal patterns and potential risks in the review tasks, and determine a business project review risk assessment model based on machine learning.

[0078] Input the test set into the business project review risk assessment model, test the business project review risk assessment model based on the test set, evaluate the performance of the business project review risk assessment model, and determine the model test evaluation results;

[0079] Among them, a test set was used to test the business project review risk assessment model, and the accuracy, precision and recall were used to determine whether the business project review risk assessment model could achieve the expected effect of identifying abnormal patterns and potential risks in the review task.

[0080] When the business project audit risk assessment model fails to achieve the expected effect of identifying abnormal patterns and potential risks in the audit task, the parameters of the business project audit risk assessment model are adjusted and optimized until the business project audit risk assessment model can achieve the expected effect of identifying abnormal patterns and potential risks in the audit task, thereby determining the optimal business project audit risk assessment model.

[0081] In this embodiment, abnormal patterns and potential risks in the audit task are identified, and the business project audit risk assessment results are determined. The following operations are performed:

[0082] Deploy the business project review risk assessment model and place it in the actual business project review risk assessment environment;

[0083] The characteristic data of business project audits are input into the business project audit risk assessment model. The model is used to analyze the characteristic data of business project audits and automatically identify abnormal patterns and potential risks in business project audit tasks. The risk level of the task is assessed and the business project audit risk assessment result is determined.

[0084] The task sorting and intelligent allocation module is used to prioritize tasks and intelligently allocate them to ensure efficient resource utilization.

[0085] Specifically, assess the task risk level and perform the following operations:

[0086] Extract the number of associated business nodes and the probability of anomalies corresponding to the associated business nodes from the abnormal patterns obtained by the business project audit risk assessment model analysis;

[0087] Retrieve the weight value corresponding to each associated business node;

[0088] By performing a weighted average of the weight value and the anomaly probability of each associated business node, a comprehensive parameter for the anomaly probability of associated businesses can be obtained. Where 'a' represents the number of associated business nodes; w yi This represents the weight value corresponding to the i-th associated business node. The weight ratio is set empirically based on the importance of the associated business nodes, for example, 0.1 or 0.18. Furthermore, the sum of the weight values ​​of all associated business nodes is 1. yi This represents the probability of an anomaly occurring for the i-th associated business node;

[0089] Extract the number of business process steps involved in the potential risks obtained from the business project review risk assessment model analysis, and the probability of risk occurrence corresponding to each business process step;

[0090] Retrieve the weight value corresponding to each business process step;

[0091] By performing a weighted average of the weight value corresponding to each business process step and the probability of risk occurrence corresponding to each business process step, a comprehensive parameter of business process risk probability is obtained. Where b represents the number of steps in the business process; w i This represents the weight value corresponding to the i-th business process step. The weight ratio is set based on experience, taking into account the importance of each business process step; for example, 0.2 or 0.21. Furthermore, the sum of the weight values ​​of all related business nodes is 1. xi This represents the probability of risk occurring at the i-th business process stage;

[0092] The comprehensive parameters of the probability of anomalies in related businesses and the comprehensive parameters of the probability of risks in business processes are summed to obtain the initial risk score corresponding to the audit feature data of business projects.

[0093] Compare the initial risk score corresponding to the business project review feature data with the preset risk score threshold;

[0094] When the initial risk score corresponding to the business project review feature data does not exceed the preset risk score threshold, the business project review task is determined to be of low risk level.

[0095] When the initial risk score corresponding to the business project review feature data exceeds the preset risk score threshold, the business project review feature data will be classified into a secondary risk level.

[0096] The technical effects of the above solution are as follows: By extracting the number of associated business nodes and the probability of anomalies corresponding to abnormal patterns, the number of business process steps involved in potential risks and the probability of risk occurrence, and combining the weights of nodes and steps for quantitative calculation, abstract anomalies and risks are transformed into measurable comprehensive parameters of associated business anomaly probability, comprehensive parameters of business process risk probability, and initial risk scores. This achieves accurate capture and quantitative presentation of highly concealed and correlated abnormal patterns and potential risks in complex business projects, avoiding the subjectivity and oversight of manual identification. Simultaneously, based on the initial risk score and preset threshold division logic, low-risk tasks can be allocated basic review resources for rapid processing, while tasks exceeding the threshold undergo secondary division to match more precise resource allocation. This avoids redundant resource consumption on low-risk tasks while ensuring that high-risk tasks receive sufficient review resources, achieving a reasonable allocation of resources to high-risk, high-priority review tasks. Furthermore, by quantitatively assessing risks and clarifying them into measurable scores, the potential risks of low-risk tasks are monitored in a standardized manner, and high-risk tasks exceeding the threshold are subject to key control through secondary division. This reduces review oversights caused by untimely risk identification or inadequate control, and reduces the possibility of risks escalating into actual problems from a process perspective. At the same time, by using quantitative parameters and scoring mechanisms, risk levels can be classified in an orderly manner, avoiding the inefficiency caused by using a uniform review process for all tasks. This allows low-risk tasks to be approved quickly and high-risk tasks to be accurately focused, shortening the overall review cycle, reducing duplicate reviews and invalid operations, and improving the efficiency of the review process in complex business scenarios.

[0097] Specifically, when the initial risk score corresponding to the business project review feature data exceeds a preset risk score threshold, a secondary risk level classification is performed on the business project review feature data, and the following operations are performed:

[0098] Retrieve the number of associated business nodes and the probability of anomalies occurring for each associated business node;

[0099] The standard deviation of the node anomaly probability is obtained based on the number of associated business nodes and the anomaly occurrence probability corresponding to the associated business nodes.

[0100] Retrieve the probability of risk occurrence for each step of the business process;

[0101] The probability of risk occurrence for each business process step is compared with a preset risk probability threshold, where the preset risk probability threshold is set to 0.4.

[0102] Select business process steps whose probability of occurrence is not lower than the preset risk probability threshold as target business process steps;

[0103] Determine whether there are non-target business process steps interspersed between the target business process steps, and obtain each group of target business process steps that are interspersed with non-target business process steps;

[0104] Set a risk impact coefficient for each group of target business process steps that includes non-target business process steps;

[0105] The risk impact coefficient is obtained using the following formula:

[0106]

[0107] Where S represents the risk impact coefficient; n represents the number of target business process steps that are interspersed with non-target business process steps; w 01i and w 02i Let P represent the weight values ​​of the first and second target business process steps contained in the i-th group of target business process steps, respectively; 01i and P 02i P represents the risk probabilities of the first and second target business process steps contained in the i-th group of target business process steps, respectively. fi This represents the weighted average probability value of all non-target business process links interspersed between the target business process links in the i-th group, along with their respective weight values.

[0108] The secondary risk assessment parameters are obtained by combining the risk impact coefficient with the standard deviation of the node anomaly probability.

[0109] The secondary risk assessment parameters are obtained using the following formula:

[0110]

[0111] Where F represents the secondary risk assessment parameter; S represents the risk impact coefficient; σ represents the standard deviation of the node anomaly probability; P ymax and P ymin These represent the maximum and minimum probabilities of anomalies occurring in the associated business nodes, respectively.

[0112] The secondary risk assessment parameters are compared with preset risk assessment parameter thresholds;

[0113] When the secondary risk assessment parameter is lower than the preset risk assessment parameter threshold, the business project review task is determined to be of medium risk level.

[0114] When the secondary risk assessment parameter is not lower than the preset risk assessment parameter threshold, the business project review task is determined to be of a high-risk level.

[0115] The technical effects of the above solution are as follows: Through detailed analysis of related business nodes and business process links in the secondary segmentation (such as calculating the standard deviation of node anomaly probability and screening target business process links), more hidden risk correlation patterns in business processes are uncovered (such as the risk transmission characteristics of target links interspersed with non-target links). This compensates for the shortcomings of the initial assessment in capturing risks in complex processes, making the identification of anomaly patterns and potential risks more comprehensive and in-depth, and helping the system to more accurately understand complex business risks. For example, in business processes with multiple nested links, the clustering and interval distribution characteristics of key risk links can be identified, and risk transmission chains that are easily overlooked by traditional single assessments can be discovered. The secondary segmentation determines the target links and risk impact coefficients based on risk probability, link distribution, etc., providing a more granular basis for resource allocation. High-risk characteristics (such as target link groups interspersed with non-target links) correspond to higher resource priority, while low-risk characteristics allow for reasonable resource streamlining, enabling the system's resource allocation to adapt to the complex business risk level and avoiding resource misallocation. For example, for high-impact coefficient tasks interspersed with multiple groups of non-target links, senior reviewers are allocated and review time is extended; for simple risk characteristic tasks, standardized rapid processes are used for processing. Meanwhile, a risk impact coefficient formula is introduced to quantify the risk amplification effect caused by interleaved process steps. Combined with secondary judgment parameters, this allows the management of high-risk tasks to focus more on key risk points (such as cluster risks in target stages). The system can formulate specific control strategies for these key risk points (such as multi-round review and cross-departmental collaborative verification) to reduce the actual probability of risk occurrence and strengthen the ability to withstand risks in complex businesses. In addition, secondary segmentation accurately identifies high-risk characteristic tasks and matches them with suitable processes, avoiding the problem of low accuracy in review due to rigid review of complex businesses. Low-risk tasks continue to use efficient initial processes, while high-risk tasks focus on the review of key risk points, reducing invalid review steps, improving the overall efficiency of the system in handling complex businesses, allowing review resources to be more rationally allocated among tasks of different risk levels, and shortening the overall business review cycle.

[0116] On the other hand, the initial risk score is obtained by weighted averaging of nodes and processes, focusing on the average impact of basic risk elements. The secondary segmentation introduces new dimensions such as the distribution characteristics of process links (the situation where target links are mixed in) and the standard deviation of node anomaly probability, supplementing the dynamic characteristics of risk transmission and fluctuation in the process. From static average assessment to dynamic fluctuation and distribution characteristic assessment, it covers the multi-dimensional attributes of complex business risks, making the risk level assessment more in line with the actual business risk landscape. For example, in project review, the initial score reflects the average level of risk in each link, while the secondary segmentation discovers the risk resonance effect caused by the cluster distribution of key risk links, correcting the bias of only looking at the average.

[0117] Meanwhile, the secondary classification refines the criteria for judging high-risk tasks (e.g., distinguishing between medium and high risk by comparing F with a threshold), and combined with the initial scoring for basic screening, constructs a risk level boundary of "initial coarse screening + secondary fine screening". In complex business scenarios, this avoids the initial scoring "misplaced" high-risk tasks, and the secondary classification accurately distinguishes between medium and high risk, making the risk level boundary clearer and more accurate, reducing misjudgments. For example, in the review of bidding for large projects, after the initial scoring identifies high risk, the secondary classification accurately distinguishes between "medium risk (F slightly exceeds the threshold, risk is controllable)" and "high risk (F far exceeds the threshold, risk is likely to get out of control)," providing an accurate basis for subsequent handling. At the same time, the initial solution quickly completes basic risk stratification (low-risk / high-risk initial screening), and the secondary classification only conducts fine-grained evaluations for high-risk initial screening tasks, avoiding complex secondary analysis for all tasks and reducing ineffective calculations. When the system processes complex business scenarios, most low-risk tasks are quickly approved using an efficient initial process, and only a small portion of high-risk tasks undergo time-consuming but necessary secondary evaluations, improving overall evaluation efficiency.

[0118] On the other hand, the secondary classification formula focuses on the key characteristics of high-risk tasks (such as the number of nodes and the complexity of processes), and the parameter acquisition relies on the results of previous model analysis (such as the number of related business nodes and the probability of risk occurrence, which can reuse the initial assessment data), reducing the time cost of repeatedly collecting data. Meanwhile, the formula's calculation logic is clear and can be quickly executed through algorithm optimization (such as parallel calculation of node standard deviation and process probability comparison), ensuring assessment efficiency under complex calculations. Simultaneously, the secondary classification outputs a clear determination of medium-to-high risk levels (the comparison result of F and the threshold), which, combined with the initial score, forms a rapid mapping of "risk level - handling strategy." Based on the determination results, the system can directly trigger preset handling procedures (such as initiating standard review for medium-risk tasks and expert review for high-risk tasks), without additional decision-making steps, accelerating the closed loop from risk assessment to control execution and improving the efficiency of handling complex business risks.

[0119] In this embodiment, tasks are prioritized and intelligently assigned, and the following operations are performed:

[0120] Based on the risk assessment results of business project audits, the audit tasks of complex business projects are prioritized, with the priority determined by combining risk level factors, task complexity factors, and urgency factors.

[0121] Tasks are dynamically allocated based on task complexity, reviewer capabilities and work status, and task priority to optimize resource allocation.

[0122] It should be noted that task priority is determined by combining risk level factors, task complexity factors, and urgency factors. Among them, risk level factors refer to the degree of risk that the task may bring, such as financial loss, reputational damage, etc.; task complexity factors refer to the difficulty of the task and the professional skills required to complete it; urgency factors refer to the urgency of completing the task, such as deadlines, customer needs, etc.

[0123] Depending on the risk level, task complexity, and urgency, a rating scale (e.g., 1-5 points) is used to set scoring criteria for each key factor. For example:

[0124] Risk level: 1 (low risk) - 5 (high risk);

[0125] Task complexity: 1 (simple) - 5 (complex);

[0126] Urgency level: 1 (not urgent) - 5 (urgent);

[0127] A risk matrix is ​​constructed based on the scoring criteria set for each key factor in the rating scale. The risk matrix is ​​a two-dimensional table with risk level and urgency as rows and columns, and the intersection points represent the corresponding risk levels, as shown in Table 1.

[0128]

[0129] In this matrix, the numbers represent risk levels; the higher the value, the higher the risk. Therefore, task priorities are determined based on the risk assessment results of business project audits, combined with risk level factors, task complexity factors, and urgency factors. By prioritizing the audit tasks of complex business projects, each task is assigned a priority, allowing for the priority audit of tasks with higher risks. At the same time, task priorities may change with project progress, resource changes, or the emergence of new information. Therefore, it is necessary to update priorities regularly or in real time to ensure that resource allocation and project progress are aligned with risk control.

[0130] It should be noted that tasks are dynamically allocated based on task complexity, auditor capabilities, and work status, combined with task priority. Task complexity refers to the difficulty of the task and the professional skills required to complete it; auditor capabilities refer to the auditor's skills, including their expertise, experience level, and certification status in different fields; and work status refers to the auditor's workload, including the current number of tasks, task completion progress, and estimated completion time.

[0131] The system dynamically matches tasks and reviewers based on task priority, complexity, reviewer capabilities, and work status. It also adjusts task allocation in real time based on changes in work status and task priority.

[0132] The real-time monitoring and dynamic adjustment module is used to monitor the task execution status in real time and dynamically adjust the task allocation and scheduling scheme based on the feedback of real-time monitoring data to deal with emergencies or resource changes.

[0133] The data visualization decision support module is used to display the review progress, risk distribution, and resource usage using data visualization tools, which facilitates management decision-making.

[0134] In this embodiment, data visualization tools are used to display the review progress, risk distribution, and resource usage, and the following operations are performed:

[0135] Data visualization tools are used to visually display the review progress, risk distribution, and resource usage through various charts and interactive dashboards;

[0136] Based on the review progress, risk distribution, and resource usage, we regularly generate review reports for complex business projects. These reports provide management with detailed information on task allocation and resource utilization, helping them quickly identify risk hotspots, optimize scheduling plans, and provide data support for strategic decision-making.

[0137] In summary, by constructing a business project review risk assessment model using machine learning algorithms, abnormal patterns and potential risks in review tasks are identified, and the results of the business project review risk assessment are determined. Based on the business project review risk assessment results, combined with risk level factors, task complexity factors, and urgency factors, the review tasks of complex business projects are prioritized, and tasks are dynamically allocated based on task complexity, reviewer capabilities, and work status, combined with task priority. This optimizes resource allocation, ensures efficient resource utilization, effectively identifies abnormal patterns and potential risks in review tasks, optimizes resource allocation and reduces risks, and improves the efficiency of complex business project review.

[0138] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0139] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A complex business project review system based on multi-dimensional historical data, characterized in that: include: The data acquisition and preprocessing module is used to collect multi-source data on business project review from internal enterprise systems and external data sources, and to preprocess the multi-source data on business project review to determine the characteristic data of business project review. The risk assessment anomaly detection module is used to build a business project review risk assessment model using machine learning algorithms, identify abnormal patterns and potential risks in review tasks, assess the risk level of tasks, and determine the business project review risk assessment results. Among them, the comprehensive parameters of the probability of anomalies in related businesses and the comprehensive parameters of the probability of risks in business processes are summed to obtain the initial risk score corresponding to the business project review feature data. When the initial risk score corresponding to the business project review feature data exceeds the preset risk score threshold, the business project review feature data is then classified into secondary risk levels. The risk impact coefficient is combined with the standard deviation of the node anomaly probability to obtain secondary risk assessment parameters, and the risk level of the business project review task is determined based on the secondary risk assessment parameters. Based on the associated business nodes corresponding to the obtained abnormal patterns, the probability of abnormal occurrence of the associated business nodes corresponding to the abnormal patterns is comprehensively calculated to obtain the comprehensive parameter of the probability of abnormality of associated business. Based on the business process links involved in the identified potential risks, the probability of occurrence of the risks corresponding to each business process link is comprehensively calculated to obtain a comprehensive parameter of business process risk probability. The task sorting and intelligent allocation module is used to prioritize tasks and intelligently allocate tasks to ensure efficient use of resources. The real-time monitoring and dynamic adjustment module is used to monitor the task execution status in real time and dynamically adjust the task allocation and scheduling scheme based on the feedback of real-time monitoring data to deal with emergencies or resource changes. The data visualization decision support module is used to display the review progress, risk distribution and resource usage using data visualization tools to facilitate management decisions. When the initial risk score corresponding to the business project review feature data exceeds the preset risk score threshold, a secondary risk level classification is performed on the business project review feature data, and the following operations are performed: Retrieve the number of associated business nodes and the probability of anomalies occurring for each associated business node; The standard deviation of the node anomaly probability is obtained based on the number of associated business nodes and the anomaly occurrence probability corresponding to the associated business nodes. Retrieve the probability of risk occurrence for each step of the business process; The probability of risk occurrence for each business process step is compared with a preset risk probability threshold, where the preset risk probability threshold is set to 0.

4. Select business process steps whose probability of occurrence is not lower than the preset risk probability threshold as target business process steps; Determine whether there are non-target business process steps interspersed between the target business process steps, and obtain each group of target business process steps that are interspersed with non-target business process steps; Set a risk impact coefficient for each group of target business process steps that includes non-target business process steps; The secondary risk assessment parameters are obtained by combining the risk impact coefficient with the standard deviation of the node anomaly probability. The secondary risk assessment parameters are compared with preset risk assessment parameter thresholds; When the secondary risk assessment parameter is lower than the preset risk assessment parameter threshold, the business project review task is determined to be of medium risk level. When the secondary risk assessment parameter is not lower than the preset risk assessment parameter threshold, the business project review task is determined to be of a high-risk level.

2. The complex business project review system based on multi-dimensional historical data according to claim 1, characterized in that, Identify anomalous patterns and potential risks in audit tasks and determine the results of business project audit risk assessments, then perform the following operations: Deploy the business project review risk assessment model and place it in the actual business project review risk assessment environment; The characteristic data of business project audits are input into the business project audit risk assessment model. The model is used to analyze the characteristic data of business project audits and automatically identify abnormal patterns and potential risks in business project audit tasks. The risk level of the task is assessed and the business project audit risk assessment result is determined.

3. The complex business project review system based on multi-dimensional historical data according to claim 2, characterized in that, Assess the risk level of the task and perform the following actions: Extract the number of associated business nodes and the probability of anomalies corresponding to the associated business nodes from the abnormal patterns obtained by the business project audit risk assessment model analysis; Retrieve the weight value corresponding to each associated business node; By performing a weighted average of the weight value and the anomaly probability of each associated business node, a comprehensive parameter for the anomaly probability of associated businesses can be obtained. Where 'a' represents the number of associated business nodes; This represents the weight value corresponding to the i-th associated business node; This represents the probability of an anomaly occurring for the i-th associated business node; Extract the number of business process steps involved in the potential risks obtained from the business project review risk assessment model analysis, and the probability of risk occurrence corresponding to each business process step; Retrieve the weight value corresponding to each business process step; By performing a weighted average of the weight value corresponding to each business process step and the probability of risk occurrence corresponding to each business process step, a comprehensive parameter of business process risk probability is obtained. Where b represents the number of steps in the business process; This represents the weight value corresponding to the i-th business process step. This represents the probability of risk occurring at the i-th business process stage; The comprehensive parameters of the probability of anomalies in related businesses and the comprehensive parameters of the probability of risks in business processes are summed to obtain the initial risk score corresponding to the audit feature data of business projects. Compare the initial risk score corresponding to the business project review feature data with the preset risk score threshold; When the initial risk score corresponding to the business project review feature data does not exceed the preset risk score threshold, the business project review task is determined to be of low risk level. When the initial risk score corresponding to the business project review feature data exceeds the preset risk score threshold, the business project review feature data will be classified into a secondary risk level.

4. The complex business project review system based on multi-dimensional historical data according to claim 2, characterized in that, Prioritize tasks and intelligently assign tasks, performing the following operations: Based on the risk assessment results of business project audits, the audit tasks of complex business projects are prioritized, with the priority determined by combining risk level factors, task complexity factors, and urgency factors. Tasks are dynamically allocated based on task complexity, reviewer capabilities and work status, and task priority to optimize resource allocation.

5. The complex business project review system based on multi-dimensional historical data according to claim 3, characterized in that, Preprocess the multi-source data for business project audits by performing the following operations: Clean the multi-source data for business project review, remove noise that is irrelevant to the review of complex business projects, and handle missing and outlier values ​​in the multi-source data for business project review. The multi-source data for business project audits is transformed into a unified data format, eliminating dimensional differences and forming standardized multi-source data for business project audits.

6. The complex business project review system based on multi-dimensional historical data according to claim 4, characterized in that, Preprocessing of multi-source data for business project audits also includes the following operations: Integrate multi-source data for business project review, merge multi-source data for business project review from different sources into a unified data view, and store the integrated multi-source data for business project review. Feature extraction is performed on multi-source data of business project review to extract features related to complex business project review and determine the feature data of business project review.

7. The complex business project review system based on multi-dimensional historical data according to claim 5, characterized in that, Utilize machine learning algorithms to build a business project review risk assessment model and perform the following operations: Collect historical data on business project review and divide the historical data into training set and test set. Based on machine learning algorithms, a training set is used to train the machine learning model, enabling the machine learning model to autonomously learn complex business project review behaviors from the training set, identify abnormal patterns and potential risks in the review tasks, and determine a business project review risk assessment model based on machine learning. Input the test set into the business project review risk assessment model, test the business project review risk assessment model based on the test set, evaluate the performance of the business project review risk assessment model, and determine the model test evaluation results; Based on the model testing and evaluation results, the business project audit risk assessment model is adjusted and optimized to determine the optimal business project audit risk assessment model.

8. The complex business project review system based on multi-dimensional historical data according to claim 7, characterized in that, To evaluate the performance of the business project audit risk assessment model, perform the following operations: The business project review risk assessment model was tested using a test set. The accuracy, precision, and recall were used to determine whether the business project review risk assessment model could achieve the expected effect of identifying abnormal patterns and potential risks in the review task. When the business project audit risk assessment model fails to achieve the expected effect of identifying abnormal patterns and potential risks in the audit task, the parameters of the business project audit risk assessment model are adjusted and optimized until the business project audit risk assessment model can achieve the expected effect of identifying abnormal patterns and potential risks in the audit task, and the optimal business project audit risk assessment model is determined.

9. The complex business project review system based on multi-dimensional historical data according to claim 7, characterized in that, Use data visualization tools to display review progress, risk distribution, and resource usage, and perform the following operations: Data visualization tools are used to visually display the review progress, risk distribution, and resource usage through various charts and interactive dashboards; Based on the review progress, risk distribution, and resource usage, we regularly generate review reports for complex business projects to provide management with detailed information on task allocation and resource utilization.

Images