Collaborative process intelligent auxiliary approval method and system based on dynamic risk modeling

Abstract

The application relates to the technical field of intelligent auxiliary examination and approval, in particular to a collaborative process intelligent auxiliary examination and approval method and system based on dynamic risk modeling, which comprises the following steps: obtaining a delay factor of each examination and approval role type node according to the difference between all examination and approval time lengths of each examination and approval role type node in all structure trees and reference examination and approval time lengths; obtaining an overall delay factor of each examination and approval task according to the delay factor and a weight coefficient; mapping all examination and approval tasks in history in a risk feature space to obtain a plurality of data points; clustering all data points in the risk feature space to obtain a plurality of class clusters; obtaining an abnormal risk factor of each to-be-analyzed point according to the distance between each to-be-analyzed point and all class cluster center points and the distance between the class cluster center points and the original point, so as to obtain a priority, and performing examination and approval through the priority. The application improves the efficiency and fluency of intelligent auxiliary examination and approval.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent auxiliary approval, and in particular to a collaborative process intelligent auxiliary approval method and system based on dynamic risk modeling. BACKGROUND

[0002] At present, with the increase of enterprise scale and business complexity, traditional approval processes have significant defects in efficiency and risk management. Many enterprises still rely on manual approval, resulting in cumbersome and time-consuming approval processes, poor information transmission, and even errors due to human factors. In addition, existing approval systems are mostly based on rule engines, which have improved automation but lack real-time dynamic risk assessment and response mechanisms, and cannot flexibly respond to external environmental changes (such as policy changes, market fluctuations, etc.) and complex situations that occur during the approval process.

[0003] Especially in multi-department collaborative approval processes, coordination and communication between each link still rely on manual operation, which is prone to poor communication and delayed approval. While traditional intelligent approval systems can improve efficiency, their fixed approval rules and processes are difficult to adapt to rapidly changing risk environments, resulting in a lack of flexibility and accuracy in decision-making. Therefore, developing an intelligent auxiliary approval method based on dynamic risk modeling can optimize the approval process according to real-time risk changes and effectively coordinate the collaborative work between departments, which is an urgent need to improve approval efficiency and reduce approval risks. SUMMARY

[0004] The present application provides a collaborative process intelligent auxiliary approval method and system based on dynamic risk modeling to solve the problems of low efficiency, lack of dynamic risk assessment, and poor coordination in existing approval processes.

[0005] The purpose of the present application can be achieved by the following technical solutions:

[0006] The first aspect of the present application is to provide a collaborative process intelligent auxiliary approval method based on dynamic risk modeling, comprising:

[0007] A tree structure is constructed for each approval task to obtain a structure tree for each approval task; all historical data of the approval task are obtained; the historical data includes approval duration, number of modifications;

[0008] A delay factor is obtained for each approval role type node based on the difference between all approval durations of each approval role type node in all structure trees and a reference approval duration; the type of each task in each approval task is determined, and the importance of each node is determined through the type of each task; a weight coefficient of each node is obtained based on the importance of each node and the level corresponding to each node; an overall delay factor of each approval task is obtained based on the delay factor and the weight coefficient;

[0009] obtaining compliance risk and legal risk of each approval task in the structure tree; performing linear normalization on the amount in the financial risk, the compliance risk, the legal risk and the modification times to obtain a financial risk factor, a compliance risk factor, a legal risk factor and a modification factor, constructing a risk feature space through the overall delay factor, the financial risk factor, the compliance risk factor, the legal risk factor and the modification factor; mapping all approval tasks in the history in the risk feature space to obtain a plurality of data points; clustering all data points in the risk feature space to obtain a plurality of clusters; mapping all current approval tasks in the risk feature space to obtain a plurality of to-be-analyzed points; obtaining an abnormal risk factor of each to-be-analyzed point according to the distance between each to-be-analyzed point and all cluster center points and the distance between the cluster center points and the origin;

[0010] determining the priority of all current approval tasks through the abnormal risk factor, and approving through the priority.

[0011] Further, the delay factor of each approval role type node is obtained according to the difference between all approval durations of each approval role type node in all structure trees and reference approval durations, comprising:

[0012]

[0013] wherein, the mean of all approval durations of each approval role type in all structure trees is represented by the reference approval duration of each approval role type is represented by the activation function is represented by the activation function is used for data normalization; the delay factor of each approval role type node is represented by

[0014] Further, the task type in each approval task is determined, and the importance of each node is determined through the task type; the weight coefficient of each node is obtained according to the importance of each node and the level corresponding to each node, comprising:

[0015] the task type includes the task type corresponding to the financial risk, the legal risk and the compliance risk;

[0016] the importance of the node corresponding to the financial risk is recorded as a preset first importance, the importance of the node corresponding to the legal risk is recorded as a preset second importance, and the importance of the node corresponding to the compliance risk is recorded as a preset third importance;

[0017] the weight coefficient of each node is specifically represented by the formula:

[0018]

[0019] wherein, represents the importance of each node, represents the number of levels corresponding to each node, represents the weight coefficient of each node, represents a linear normalization function.

[0020] Further, the obtaining of the overall delay factor of each approval task according to the delay factor and the weight coefficient comprises:

[0021] obtaining the delay factor of each node in each approval task through the delay factor of each approval role type node;

[0022] The overall delay factor of each approval task is specifically represented by a formula as:

[0023]

[0024] wherein, represents the delay factor of the th node in each approval task, represents the weight coefficient of the th node in each approval task, represents the number of all nodes in each approval task, represents the overall delay factor of each approval task.

[0025] Further, the clustering of all data points in the risk feature space to obtain a plurality of class clusters comprises:

[0026] The clustering of all data points in the risk feature space through a K-means clustering algorithm to obtain a plurality of class clusters; wherein, the number of class clusters is obtained through an elbow method.

[0027] Further, the obtaining of the abnormal risk factor of each to-be-analyzed point according to the distance between each to-be-analyzed point and all class cluster center points and the distance between the class cluster center points and the origin comprises:

[0028]

[0029]

[0030] wherein, represents the distance between each to-be-analyzed point and the th class cluster center point, represents the distance between the th class cluster center point and the origin of the risk feature space, represents the number of all clusters, represents the risk reference weight of the first cluster to each point to be analyzed; represents the abnormal risk factor of each point to be analyzed, represents an exponential function with a natural constant as the base.

[0031] Further, the priority of all current approval tasks is determined by the abnormal risk factor, and the approval is performed by the priority, comprising:

[0032] The abnormal risk factors of all current approval tasks are normalized, and the normalization result is taken as the priority of all current approval tasks, and the approval is performed in the order from large to small according to the priority of all current approval tasks.

[0033] The second aspect of the application is to provide a collaborative process intelligent auxiliary approval system based on dynamic risk modeling, comprising:

[0034] The historical data acquisition module is used to construct a tree structure for each approval task, obtain a structure tree of each approval task, and acquire all historical data of the approval task; the historical data includes approval duration, modification times;

[0035] The delay analysis module is used to obtain a delay factor of each approval role type node according to the difference between all approval durations of each approval role type node in all structure trees and a reference approval duration, determine a task type in each approval task, and determine the importance of each node through the task type; obtain a weight coefficient of each node according to the importance of each node and the level corresponding to each node; and obtain an overall delay factor of each approval task according to the delay factor and the weight coefficient;

[0036] The approval task risk determination module is used to acquire compliance risk and legal risk of each approval task in the structure tree; linearly normalize the amount of money in the financial risk, the compliance risk, the legal risk and the modification times, respectively, to obtain a financial risk factor, a compliance risk factor, a legal risk factor and a modification factor, construct a risk feature space through the overall delay factor, the financial risk factor, the compliance risk factor, the legal risk factor and the modification factor; map all approval tasks in history in the risk feature space to obtain a plurality of data points; cluster all data points in the risk feature space to obtain a plurality of clusters; map all current approval tasks in the risk feature space to obtain a plurality of points to be analyzed; obtain an abnormal risk factor of each point to be analyzed according to the distance between each point to be analyzed and all cluster center points and the distance between the cluster center points and the origin;

[0037] Priority determination and approval module: used for determining the priority of all current approval tasks through the abnormal risk factor, and approving through the priority.

[0038] The third aspect of the present application provides an electronic device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the intelligent auxiliary approval method based on dynamic risk modeling when executing the computer program.

[0039] The fourth aspect of the present application provides a computer readable storage medium, which stores a computer program, wherein the computer program is executed by a processor to implement the intelligent auxiliary approval method based on dynamic risk modeling.

[0040] Compared with the prior art, the beneficial effects of the present application are as follows: according to the difference between all approval durations of each approval role type node in all structure trees and the reference approval duration, a delay factor of each approval role type node is obtained; according to the importance of each node and the level corresponding to each node, a weight coefficient of each node is obtained; according to the delay factor and the weight coefficient, an overall delay factor of each approval task is obtained; the accuracy of delay influence analysis between different levels is improved; all approval tasks in the history are mapped in a risk feature space to obtain a plurality of data points; all data points in the risk feature space are clustered to obtain a plurality of clusters; all current approval tasks are mapped in the risk feature space to obtain a plurality of to-be-analyzed points; according to the distance between each to-be-analyzed point and all cluster center points and the distance between the cluster center points and the origin, an abnormal risk factor of each to-be-analyzed point is obtained; the accuracy of dynamic risk analysis is improved, and the influence of risk on the approval process is reduced; the priority of all current approval tasks is determined through the abnormal risk factor, and the approval is performed through the priority, thereby improving the efficiency and fluency of intelligent auxiliary approval. BRIEF DESCRIPTION OF DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0042] Figure 1 The step flowchart of the intelligent auxiliary approval method based on dynamic risk modeling of the present application is shown in the following figure:

[0043] Figure 2A module flow diagram of the intelligent auxiliary examination and approval system based on dynamic risk modeling and collaborative process is provided for the present application. DETAILED DESCRIPTION

[0044] In order for those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0045] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not necessarily limit to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0046] In view of the problems in the background art, a collaborative process intelligent auxiliary examination and approval method and system based on dynamic risk modeling are designed, which has important practical significance.

[0047] As shown in Figure 1 The first aspect of the present application is to provide a collaborative process intelligent auxiliary examination and approval method based on dynamic risk modeling, comprising the following steps:

[0048] Step S001: Construct a tree structure for each examination and approval task, obtain the structure tree of each examination and approval task, and obtain all historical data of the examination and approval task.

[0049] It should be noted that in order to clearly understand the relationship between different levels and the dependency relationship in the examination and approval task, the key nodes and potential risks of each link are identified by visual means. The tree structure can clearly show the order of task execution and the association of each examination and approval link, facilitate monitoring the progress of the task, and ensure that there is no omission or inefficient link. At the same time, it is also helpful to optimize resource allocation, improve processing efficiency, prioritize high-risk or delayed nodes, avoid bottlenecks or errors in task execution, and thus improve the standardization and accuracy of the overall examination and approval process.

[0050] Further need to explain is, in order to improve the flexibility and efficiency of system approval, then through the acquisition of historical data, through historical data and the specific circumstances of the approval task to analyze and determine the priority of each approval task.

[0051] Specifically, a tree structure is constructed for each approval task, and a structure tree of each approval task is obtained. All historical data of the approval task is obtained; the historical data includes the approval duration, the number of modifications of each approval task.

[0052] So far, the construction of the approval task structure tree and all historical data of all approval tasks in the approval process are completed.

[0053] Step S002: According to the difference between all approval durations of each approval role type node in all structure trees and the reference approval duration, obtain the delay factor of each approval role type node; according to the importance of each node and the level corresponding to each node, obtain the weight coefficient of each node; according to the delay factor and the weight coefficient, obtain the overall delay factor of each approval task.

[0054] It should be noted that in an approval task, there will be multiple people of different levels to approve, and these approvers correspond to different nodes in the structure tree, that is, one node corresponds to one approver.

[0055] Further need to explain is, because the approval person of different nodes corresponds to different positions, therefore in the process of an approval task, the points of different nodes of the approval person are different; For example, in an approval task, there may be general manager approval, financial approval and department manager approval, etc.; Therefore, the risks of different nodes in the approval are different, so it is necessary to analyze the different risk situations by combining the duty type of each node of the approval person.

[0056] Further need to explain is, in order to determine the risk situation of each node, it is necessary to determine the approval role type of each node, and the approval delay degree corresponding to each node is analyzed according to the historical data of the same approval role type.

[0057] Specifically, first, determine the approval role type corresponding to the node; wherein, the approval role type includes financial approval, leadership approval, department manager approval, senior management approval, specialist approval and administrative approval, etc.; That is, the approval role type node is divided into financial approval node, leadership approval node, department manager approval node, senior management approval node, specialist approval node and administrative approval node, etc.

[0058] According to the difference between the reference approval time length and the approval time length of each approval role type node in all the structure trees, a delay factor of each approval role type node is obtained; wherein the delay factor of each approval role type node is specifically expressed by a formula:

[0059]

[0060] In the formula, represents the average of the approval time length of each approval role type in all the structure trees, represents the reference approval time length of each approval role type, represents an activation function, represents an activation function for data normalization; represents the delay factor of each approval role type node.

[0061] wherein, represents the difference between the average of the approval time length of each approval role type in all the structure trees and the reference approval time length, and the greater the difference, the longer the delay time of the approval role type in the historical approval process; when the difference is smaller and positive, it means that there is a delay but the delay time is shorter; when the difference is negative, it means that the approval role type in the historical approval process does not exist delay, and the whole is completed within the approved time.

[0062] At this point, the delay factor of each approval role type node is obtained by the above method; and the delay factor of each node in each approval task is also obtained.

[0063] It should be noted that since the importance of different nodes in the approval process is different, the importance of each node also needs to be analyzed.

[0064] Further, since the higher the level, the greater the responsibility, the importance of the node with the higher level should be greater. In the approval process, the same level may bear different types of approval, such as approval of data amount, approval of legal risk, etc.; therefore, the importance of each node also needs to be analyzed according to different types.

[0065] Specifically, the task type in each approval task is determined, and in this embodiment, the task type corresponding to the financial risk, the legal risk and the compliance risk is taken as an example for illustration, but the task type is not specifically limited, and the implementer can determine it according to the actual situation of the specific task type.

[0066] The importance of the financial risk corresponding node is recorded as a preset first importance, the importance of the legal risk corresponding node is recorded as a preset second importance, and the importance of the compliance risk corresponding node is recorded as a preset third importance. In this embodiment, the preset first importance is 0.8, the preset second importance is 0.7, and the preset third importance is 0.5. In this embodiment, the preset first importance, the preset second importance, and the preset third importance are not specifically limited, and can be determined according to specific conditions.

[0067] According to the importance of each node and the level corresponding to each node, a weight coefficient of each node is obtained. The weight coefficient of each node is specifically represented by a formula as follows:

[0068]

[0069] In the formula, importance represents the importance of each node,

[0070] When the importance of each node and the level corresponding to each node are greater, the weight coefficient of each node is greater. Conversely, the weight coefficient of each node is smaller.

[0071] According to the delay factor of each node and the weight coefficient in the history, a total delay factor of each approval task is obtained. The total delay factor of each approval task is specifically represented by a formula as follows:

[0072]

[0073] In the formula, delay factor represents the delay factor of the i th node in each approval task,

[0074] In the formula, delay contribution represents the delay contribution of each node, and the total delay factor of each approval task is calculated by weighted average of all nodes.

[0075]

[0076] ​​​​​​​​​​​​Step S003: constructing a risk feature space; mapping all the historical approval tasks in the risk feature space to obtain a plurality of data points; clustering all the data points in the risk feature space to obtain a plurality of clusters; mapping all the current approval tasks in the risk feature space to obtain a plurality of to-be-analyzed points; and obtaining an abnormal risk factor of each to-be-analyzed point according to the distance between each to-be-analyzed point and all cluster center points and the distance between the cluster center points and the origin.

[0077] It should be noted that the greater the amount in the approval task, the greater the financial risk; and the greater the overall delay factor of each approval task in all the structure trees, the greater the financial risk, the greater the compliance risk, the greater the legal risk, and the greater the risk of the approval task in the history, and in combination with the number of modifications of each approval task in all the structure trees, the greater the number of modifications and the greater the risk, the greater priority should be given to the approval task in the approval process for review.

[0078] Specifically, the compliance risk and the legal risk of each approval task in all the structure trees are obtained by analyzing the approval material files submitted by each approval task in all the structure trees through a large language model; wherein the large language model in this embodiment is OpenAI GPT (OpenAI Generative Pre-trained Transformer) or LegalBERT (Legal Bidirectional Encoder Representations from Transformers); wherein the large language model is not specifically limited in this embodiment, and the implementer can determine according to the specific circumstances.

[0079] It should be further noted that the amount involved in the financial risk of each approval task can be analyzed when analyzing the financial risk of each approval task. In order to eliminate the influence of different dimensions, the financial risk, the compliance risk, the legal risk, and the number of modifications are processed to eliminate the influence of dimensions.

[0080] Specifically, the amount in the financial risk, the compliance risk, the legal risk, and the number of modifications are linearly normalized to obtain a financial risk factor, a compliance risk factor, a legal risk factor, and a modification factor.

[0081] The overall delay factor of each approval task, the financial risk factor, the compliance risk factor, the legal risk factor, and the modification factor are respectively used as dimension features to construct a risk feature space; each approval task in all the structure trees is mapped in the risk feature space to obtain a data point; and all the historical approval tasks are mapped in the risk feature space to obtain a plurality of data points.

[0082] All data points in the risk feature space are clustered by a K-means clustering algorithm to obtain a plurality of clusters; the number of clusters is obtained by an elbow method; the K-means clustering algorithm and the elbow method are both known technologies and will not be described in detail here.

[0083] It should be noted that the closer each cluster is to the origin of the risk feature space, the smaller the risk of the corresponding approval task in the cluster; the farther each cluster is to the origin of the risk feature space, the greater the risk of the corresponding approval task in the cluster. Therefore, the degree of risk of each approval task is determined by analyzing the positional relationship between the current all approval tasks and all clusters by mapping the current all approval tasks in the risk feature space, and the priority of each approval task is determined by the degree of risk.

[0084] Specifically, a plurality of to-be-analyzed points are obtained by mapping the current all approval tasks in the risk feature space; an abnormal risk factor of each to-be-analyzed point is obtained according to the distance between each to-be-analyzed point and all cluster center points and the distance between the cluster center points and the origin; wherein the abnormal risk factor of each to-be-analyzed point is specifically expressed by the following formula:

[0085]

[0086]

[0087] In the formula, represents the distance between each to-be-analyzed point and the i-th cluster center point, represents the distance between the i-th cluster center point and the origin of the risk feature space (used to represent the abnormal risk of each cluster), represents the number of all clusters, represents the risk reference weight of the i-th cluster for each to-be-analyzed point; represents the abnormal risk factor of each to-be-analyzed point, represents an exponential function with a natural constant as the base. Wherein, the closer the distance between each to-be-analyzed point and each cluster center point, the more valuable the abnormal risk of the cluster as a reference abnormal risk of the to-be-analyzed point, that is, the greater the risk reference weight of the cluster for each to-be-analyzed point should be; otherwise, the smaller the risk reference weight of the cluster for each to-be-analyzed point should be. Then, the abnormal risk factor of each to-be-analyzed point is obtained by combining all risk reference weights with the abnormal risk of all clusters.

[0088] It should be noted that the closer each cluster is to the origin of the risk feature space, the smaller the risk of the corresponding approval task in the cluster; the farther each cluster is to the origin of the risk feature space, the greater the risk of the corresponding approval task in the cluster. Therefore, the degree of risk of each approval task is determined by analyzing the positional relationship between the current all approval tasks and all clusters by mapping the current all approval tasks in the risk feature space, and the priority of each approval task is determined by the degree of risk.

[0089] ​​Thus, the abnormal risk factors for each point to be analyzed are obtained through the above methods.

[0090] Step S004: Determine the priority of all current approval tasks based on the abnormal risk factors, and approve them according to the priority.

[0091] It should be noted that by analyzing all current approval tasks, the priority of each approval task is determined through its abnormal risk factors; that is, the higher the abnormal risk factor of each approval task, the higher the priority of that approval task should be, meaning that the task should be approved with higher priority.

[0092] Specifically, by obtaining the abnormal risk factors of each point to be analyzed, the abnormal risk factors of all current approval tasks are obtained; the abnormal risk factors of all current approval tasks are normalized, and the normalization result is used as the priority of all current approval tasks. The approval tasks are then processed in descending order of priority.

[0093] This concludes the embodiment.

[0094] like Figure 2 As shown, a second aspect of the present invention is to provide a collaborative process intelligent auxiliary approval system based on dynamic risk modeling, comprising:

[0095] Historical data acquisition module 101: used to construct a tree structure for each approval task, obtain the tree structure of each approval task; acquire all historical data of the approval task; the historical data includes approval duration and number of modifications;

[0096] Delay analysis module 102: is used to obtain the delay factor of each approval role type node based on the difference between all approval times and the reference approval time of each approval role type node in all the structure trees; determine the task type in each approval task, and determine the importance of each node through the task type; obtain the weight coefficient of each node based on the importance of each node and the level corresponding to each node; and obtain the overall delay factor of each approval task based on the delay factor and the weight coefficient.

[0097] The approval task risk determination module 103 is configured to obtain the compliance risk and the legal risk of each approval task in the structure tree; linearly normalize the amount in the financial risk, the compliance risk, the legal risk, and the modification times to obtain a financial risk factor, a compliance risk factor, a legal risk factor, and a modification factor; construct a risk feature space by using the overall delay factor, the financial risk factor, the compliance risk factor, the legal risk factor, and the modification factor; map all the approval tasks in the history in the risk feature space to obtain a plurality of data points; cluster all the data points in the risk feature space to obtain a plurality of clusters; map all the current approval tasks in the risk feature space to obtain a plurality of to-be-analyzed points; and obtain an abnormal risk factor of each to-be-analyzed point according to the distance between each to-be-analyzed point and all the cluster center points and the distance between the cluster center points and the origin.

[0098] The priority determination and approval module 104 is configured to determine the priorities of all the current approval tasks by using the abnormal risk factors and to perform approval by using the priorities.

[0099] The third aspect of the present application provides an electronic device, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the intelligent auxiliary approval method for collaborative processes based on dynamic risk modeling when executing the computer program.

[0100] The fourth aspect of the present application provides a computer readable storage medium, which stores a computer program, and the computer program implements the intelligent auxiliary approval method for collaborative processes based on dynamic risk modeling when executed by a processor.

[0101] Those skilled in the art should understand that embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, etc.) containing computer-usable program code.

[0102] The present application is described with reference to flowcharts and / or block diagrams of the method, the system, and the computer program product according to embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus generate a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1apparatuses that implement the functions specified in the flowchart or flowcharts and / or blocks. Figure 1

[0103] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the flowchart or flowcharts and / or blocks. Figure 1 apparatuses that implement the functions specified in the flowchart or flowcharts and / or blocks. Figure 1

[0104] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the flowchart or flowcharts and / or blocks. Figure 1 apparatuses that implement the functions specified in the flowchart or flowcharts and / or blocks. Figure 1

[0105] Finally, it should be noted that the above-mentioned embodiments are merely used to illustrate the technical solutions of the present application, rather than limit the technical solutions of the present application. Although the present application has been described in detail with reference to the above-mentioned embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalent replaced without departing from the spirit and scope of the present application, and any modification or equivalent replacement should be covered in the protection scope of the present application.​​​

Claims

1. A method for intelligent assistant approval based on dynamic risk modeling and collaborative process, characterized in that, The application relates to a method for determining the priority of an approval task. The method comprises the following steps: constructing a tree structure for each approval task and obtaining a structure tree of each approval task; obtaining all historical data of the approval task; the historical data comprises an approval duration and a modification frequency; obtaining a delay factor of each approval role type node in all the structure trees according to the difference between the approval duration of each approval role type node and a reference approval duration; determining the task type in each approval task and determining the importance of each node according to the task type; obtaining a weight coefficient of each node according to the importance of each node and the level corresponding to each node; and obtaining an overall delay factor of each approval task according to the delay factor and the weight coefficient; obtaining the compliance risk and the legal risk of each approval task in the structure tree; linearly normalizing the amount of financial risk, the compliance risk, the legal risk and the modification frequency to obtain a financial risk factor, a compliance risk factor, a legal risk factor and a modification factor; and constructing a risk feature space by using the overall delay factor, the financial risk factor, the compliance risk factor, the legal risk factor and the modification factor; mapping all the approval tasks in the history in the risk feature space to obtain a plurality of data points; clustering all the data points in the risk feature space to obtain a plurality of clusters; mapping all the current approval tasks in the risk feature space to obtain a plurality of to-be-analyzed points; and obtaining an abnormal risk factor of each to-be-analyzed point according to the distance between each to-be-analyzed point and all the cluster center points and the distance between the cluster center points and the origin; 2. The method of claim 1, wherein, determining the priority of all the current approval tasks according to the abnormal risk factor and approving the approval tasks according to the priority. In the formula, represents the mean value of all approval durations of each approval role type in all structure trees, represents the reference approval duration of each approval role type, represents an activation function, represents an activation function for data normalization; represents the delay factor of each approval role type node. 3.The method of claim 1, wherein, The method comprises the following steps: determining the task type in each approval task and determining the importance of each node according to the task type; obtaining a weight coefficient of each node according to the importance of each node and the level corresponding to each node; and obtaining an overall delay factor of each approval task according to the delay factor and the weight coefficient; the task type comprises the task type corresponding to the financial risk, the legal risk and the compliance risk; the importance of the node corresponding to the financial risk is denoted as a preset first importance, the importance of the node corresponding to the legal risk is denoted as a preset second importance, and the importance of the node corresponding to the compliance risk is denoted as a preset third importance; In the formula, represents the importance of each node, represents the number of levels corresponding to each node, represents the weight coefficient of each node, represents a linear normalization function.

4. The method of claim 1, wherein, the weight coefficient of each node is specifically expressed by a formula; obtaining the delay factor of each node in each approval task through the delay factor of each approval role type node; the overall delay factor of each approval task is specifically expressed by a formula; In the formula, denotes the delay factor of the i-th node in each approval task, denotes the weight coefficient of the i-th node in each approval task, denotes the number of all nodes in each approval task, denotes the overall delay factor of each approval task.​​ 5. The method of claim 1, wherein, the method comprises the following steps: clustering all the data points in the risk feature space by using a K-means clustering algorithm to obtain a plurality of clusters; and the number of clusters is obtained by using an elbow method.

6. The method of claim 1, wherein the method further comprises: The abnormal risk factor of each point to be analyzed is obtained according to the distance between each point to be analyzed and all cluster center points and the distance between the cluster center points and the origin, and the abnormal risk factor comprises: In the formula, This indicates that each point to be analyzed is up to the nth The distance between the center points of each cluster Indicates the first The distance between the center point of each cluster and the origin of the risk feature space This represents the number of all clusters. Indicates the first Risk reference weights for each cluster and each point to be analyzed; This represents the anomaly risk factor for each point to be analyzed. This represents an exponential function with the natural constant as its base.

7. The method of claim 1, wherein the method further comprises: The priority of all current approval tasks is determined through the abnormal risk factor, and the approval is performed through the priority. The abnormal risk factors of all current approval tasks are normalized, and the normalization result is taken as the priority of all current approval tasks, and the approval is performed in the order from large to small according to the priority of all current approval tasks.

8. A collaborative process intelligent assistant approval system based on dynamic risk modeling, characterized in that, Comprise: The historical data acquisition module is used for constructing a tree structure for each approval task and obtaining a structure tree of each approval task; All historical data of the approval task are obtained; the historical data comprise an approval duration, a modification times; The delay analysis module is used for obtaining a delay factor of each approval role type node in all structure trees according to the difference between all approval durations of each approval role type node and a reference approval duration, determining a task type in each approval task, and determining the importance of each node through the task type; A weight coefficient of each node is obtained according to the importance of each node and the level corresponding to each node; and a whole delay factor of each approval task is obtained according to the delay factor and the weight coefficient; The approval task risk determination module is used for obtaining a compliance risk and a legal risk of each approval task in the structure tree; The amount in the financial risk, the compliance risk, the legal risk and the modification times are respectively linearly normalized to obtain a financial risk factor, a compliance risk factor, a legal risk factor and a modification factor, and a risk feature space is constructed through the whole delay factor, the financial risk factor, the compliance risk factor, the legal risk factor and the modification factor; All approval tasks in the history are mapped in the risk feature space to obtain a plurality of data points; All data points in the risk feature space are clustered to obtain a plurality of clusters; a plurality of points to be analyzed are obtained by mapping all current approval tasks in the risk feature space; and an abnormal risk factor of each point to be analyzed is obtained according to the distance between each point to be analyzed and all cluster center points and the distance between the cluster center points and the origin. The priority determination and approval module is used for determining the priority of all current approval tasks through the abnormal risk factor, and performing the approval through the priority.

9. An electronic device, comprising: The computer readable storage medium stores a computer program, and the computer program is executed by the processor to realize the collaborative process intelligent auxiliary approval method based on dynamic risk modeling according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to realize the collaborative process intelligent auxiliary approval method based on dynamic risk modeling according to any one of claims 1-7.

Images