Cross-department collaborative examination and approval decision-making method and system based on knowledge graph reasoning

Abstract

The application provides a cross-department collaborative approval decision-making method and system based on knowledge graph reasoning, belonging to the technical field of knowledge graph. The application first constructs a knowledge graph including approval authority nodes, approval rule nodes, approval matter nodes and historical approval nodes. Then the rule nodes before and after the approval rule version update are traversed respectively, and the rule version difference node set is identified through difference set operation. For the historical approval node, the applicable rule version at that time is determined according to the approval completion time, and the approval path is extracted. Whether the approval path includes the difference node is judged through subgraph isomorphism matching, so as to identify the invalid approval node due to rule change. Finally, the depth-first traversal is performed from the invalid node as the starting point, and the target approval sequence is generated according to the path hop count. The application accurately identifies the historical approval node invalid due to the approval rule version update through the temporal knowledge graph reasoning, and generates the minimized directional cross-department supplementary approval sequence.

Description

Technical Field

[0001] This application belongs to the field of knowledge graph technology, and in particular relates to a cross-departmental collaborative approval decision-making method and system based on knowledge graph reasoning. Background Technology

[0002] In enterprise project management, cross-departmental collaborative approval processes involve approval authority and complex dependencies across multiple departments. Approval rules are frequently iterated due to organizational restructuring, management system upgrades, or changes in external regulatory requirements. Knowledge graph technology can structure and model information such as approval rules, departmental authority, and approval dependencies, providing knowledge reasoning support for cross-departmental collaborative approval decisions. It has broad application prospects in improving approval efficiency and ensuring compliance.

[0003] Existing technologies include knowledge graph-based approval decision-making methods. These methods construct approval rules and departmental authority relationships into a knowledge graph, utilize graph traversal algorithms to identify approval process paths, and achieve intelligent allocation of approval tasks. Some technical solutions employ rule-based reasoning engines to match and judge approval conditions, assisting approvers in making decisions.

[0004] However, existing technologies lack the ability to model the temporal evolution of approval rule versions, failing to distinguish the effective boundaries of approval rules across departments at different historical moments. This results in the system being forced to either re-run the entire process for all historical approvals or grant exemptions after rule updates, unable to accurately determine the compliance of each historical cross-departmental collaborative approval conclusion under the new rules based on knowledge graph reasoning. This approach leads to a large number of unnecessary duplicate approvals, resulting in inefficiency, or creates potential compliance risk exposure due to blanket exemptions. Therefore, existing technologies suffer from insufficient collaborative approval decision-making due to the inability to accurately identify failed approval nodes and minimize supplementary approvals after rule updates. Summary of the Invention

[0005] The purpose of this application is to provide a cross-departmental collaborative approval decision-making method and system based on knowledge graph reasoning, so as to solve the problem of insufficient collaborative approval decision-making in the existing technology.

[0006] To address the aforementioned technical problems, in a first aspect, this application provides a cross-departmental collaborative approval decision-making method based on knowledge graph reasoning, comprising: Obtain a knowledge graph constructed with approval authority nodes, approval rule nodes, approval item nodes, and historical approval nodes that have been completed as nodes, and permission association relationships, rule application relationships, and approval dependency relationships as edges. Historical approval nodes include the approval completion time. Taking the approval rule node before the update as the first approval node and the updated approval rule node as the second approval node, in the knowledge graph, starting from the first approval node and the second approval node respectively, we traverse along the rule application relationship edge to the approval item node and then along the permission relationship edge to each approval permission node to obtain the first node set and the second node set. The first node set and the second node set are subjected to a difference operation to obtain the third node set. The rule version corresponding to the historical approval node under the approval completion time is determined. The approval path traversed by the historical approval node under the rule version is extracted from the knowledge graph. By performing subgraph isomorphic matching on the approval path and the third node set, the approval failure node is determined. Starting from the approval failure node, the knowledge graph is traversed in a depth-first manner along the approval dependency edges to obtain the candidate approval sequence. Based on the path hop count between each approval authority node and the approval failure node in the candidate approval sequence, the approval authority nodes are sorted to generate the target approval sequence and sent to the corresponding approval department.

[0007] Optionally, the approval authority node includes the approval duration, and the method also includes: In the knowledge graph, traverse from the approval failure node to each approval authority node in the candidate approval sequence along the approval dependency edge to obtain the traversal path of each approval authority node. The approval time of each approval authority node is accumulated by summing the approval time of all approval authority nodes traversed along the path. Based on the path hop count between each approval authority node and the approval failure node in the candidate approval sequence, the approval authority nodes are sorted to generate a target approval sequence, which is then sent to the corresponding approval department. This includes: Based on the cumulative approval time and the path hop count between each approval authority node and the approval failure node in the candidate approval sequence, the approval authority nodes are sorted to generate the target approval sequence and sent to the corresponding approval department.

[0008] Optionally, based on the cumulative approval time and the path hop count between each approval authority node and the approval failure node in the candidate approval sequence, the approval authority nodes are sorted to generate a target approval sequence and sent to the corresponding approval department, including: Sort the approval authority nodes in the candidate approval sequence in ascending order of path hop count to obtain the initial sorted sequence; Extract approval authority nodes with the same path jump number from the initial sorted sequence to construct a fourth node set. Sort the approval authority nodes in each fourth node set in ascending order of cumulative approval time to obtain a fifth node set. The fifth node set is used to replace the approval authority nodes at the corresponding positions in the initial sorting sequence, generating the target approval sequence and sending it to the corresponding approval department.

[0009] Optionally, taking the approval rule node before the update as the first approval node and the updated approval rule node as the second approval node, the knowledge graph is traversed from the first and second approval nodes respectively, along the rule application relationship edges to the approval item nodes, and then along the permission association edges to the approval permission nodes, to obtain the first node set and the second node set, including: Starting from the first approval node in the knowledge graph, we traverse along the rule application relationship edge to all approval item nodes connected to the first approval node, and then traverse along the permission relationship edge from each approval item node to all approval permission nodes connected to the approval item node. Based on all approval permission nodes, we construct the first node set. Starting from the second approval node in the knowledge graph, we traverse along the rule application relationship edge to all approval item nodes connected to the second approval node. Then, from each approval item node, we traverse along the permission relationship edge to all approval permission nodes connected to the approval item node. Based on all approval permission nodes, we construct the second node set.

[0010] Optionally, a third node set is obtained by performing a difference operation on the first node set and the second node set, including: Extract the approval authority nodes that do not exist in the second node set from the first node set to obtain the first difference set; extract the approval authority nodes that do not exist in the first node set from the second node set to obtain the second difference set. The third node set is obtained by merging the first difference set and the second difference set.

[0011] Optionally, the approval rule node includes the effective time and expiration time of the approval rule version; Determine the rule version corresponding to each historical approval node at its approval completion time, and extract the approval path traversed by each historical approval node under that rule version from the knowledge graph, including: The time range of each approval rule node is constructed based on the effective time and the expiration time. When the approval completion time of a historical approval node falls into the time range of any approval rule node, the approval rule version of the corresponding approval rule node is determined as the rule version. In the knowledge graph, starting from the approval rule node corresponding to the rule version, we traverse along the rule application relationship edge to the approval item node, and then traverse along the permission relationship edge to the approval permission node that is the same as the approval permission node associated with the historical approval node, thus obtaining the approval path consisting of the approval rule node, the approval item node, and the approval permission node.

[0012] Optionally, approval failure nodes are determined by performing subgraph isomorphic matching on the approval path and the third node set. Starting from the approval failure node, a depth-first traversal of the knowledge graph is performed along the approval dependency edges to obtain the candidate approval sequence, including: Based on all approval authority nodes in the approval path and the permission association edges of the nodes in the knowledge graph, construct an approval path subgraph; based on all approval authority nodes in the third node set and the permission association edges of the nodes in the knowledge graph, construct a difference subgraph. When there is an approval authority node in the approval path subgraph that is identical to at least one node in the difference subgraph, the historical approval node corresponding to the approval path is determined to be an approval invalidation node. Starting from the approval failure node in the knowledge graph, a depth-first traversal is performed along the approval dependency edges, and all approval permission nodes traversed are arranged in the traversal order to construct a candidate approval sequence.

[0013] Secondly, this application provides a cross-departmental collaborative approval decision-making system based on knowledge graph reasoning, including: The acquisition module is used to acquire a knowledge graph constructed with approval permission nodes, approval rule nodes, approval item nodes, and historical approval nodes that have been completed as nodes, and permission association relationships, rule application relationships, and approval dependency relationships as edges. Historical approval nodes include the approval completion time. The generation module is used to take the approval rule node before the approval rule version update as the first approval node and the updated approval rule node as the second approval node. In the knowledge graph, starting from the first approval node and the second approval node respectively, it traverses along the rule application relationship edge to the approval item node and then along the permission relationship edge to each approval permission node to obtain the first node set and the second node set. The extraction module is used to perform a difference operation on the first node set and the second node set to obtain the third node set, determine the rule version corresponding to the historical approval node under the approval completion time, and extract the approval path traversed by the historical approval node under the rule version in the knowledge graph. The generation module is also used to determine the approval failure node by performing subgraph isomorphic matching on the approval path and the third node set, and to obtain the candidate approval sequence by performing a depth-first traversal of the knowledge graph along the approval dependency edges, starting from the approval failure node. The sending module is used to sort the approval authority nodes according to the path hop count between each approval authority node and the approval failure node in the candidate approval sequence, generate the target approval sequence, and send it to the corresponding approval department.

[0014] Thirdly, this application provides an electronic device, comprising: Memory, used to store computer programs; A processor, used to execute the computer program to implement the steps of the cross-departmental collaborative approval decision-making method based on knowledge graph reasoning as described in the first aspect above.

[0015] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps of the cross-departmental collaborative approval decision-making method based on knowledge graph reasoning as described in the first aspect above.

[0016] The cross-departmental collaborative approval decision-making method based on knowledge graph reasoning provided in this application first achieves temporal modeling of the evolution of approval rule versions by annotating the approval completion time attributes of historical approval nodes in the knowledge graph, enabling the knowledge graph to reconstruct the effective state of approval rules for each department at any historical moment. Then, by performing graph traversal and difference operations starting from the old and new versions of approval rule nodes respectively, the set of approval authority nodes that changed before and after the rule version update is accurately identified, thereby clarifying the effective boundaries of approval rules at different historical moments.

[0017] Then, for each historical approval node, the applicable rule version at that time is determined based on its approval completion time, and the corresponding approval path is extracted. By using subgraph isomorphic matching, it is determined whether the path includes difference nodes, thus achieving accurate determination of the compliance status of historical approvals under the new rules.

[0018] Finally, a depth-first traversal is performed based on the approval dependencies of the failed nodes to generate a minimal supplementary approval sequence that includes only the affected nodes. This avoids the crude approach of simply re-running the entire process or exempting all historical approvals, as is common in existing technologies. Therefore, this application solves the technical problem of not being able to accurately identify failed approval nodes after rule version updates, and improves the efficiency of cross-departmental collaborative approvals.

[0019] Furthermore, this application introduces the approval duration attribute as a dimension for optimizing the target approval sequence. By traversing along the approval dependency edges and calculating the cumulative approval duration from the failed node to each candidate approval node, the time cost of different supplementary approval paths is quantified. When sorting the candidate approval sequences, both the path hop count and the cumulative approval duration are considered. When multiple approval authority nodes have the same path hop count, nodes with shorter cumulative approval durations are prioritized, thus optimizing the execution order of supplementary approval tasks.

[0020] This dual-dimensional sorting strategy prioritizes approval paths with lower time costs while ensuring the correctness of approval dependencies, thus shortening the overall supplementary approval cycle. Therefore, this application further optimizes the time efficiency of the target approval sequence, achieving the dual goals of compliance assurance and improved approval efficiency. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 A flowchart illustrating a cross-departmental collaborative approval decision-making method based on knowledge graph reasoning, provided as an embodiment of this application; Figure 2 This is a schematic diagram of the structure of a knowledge graph provided in an embodiment of this application; Figure 3 A schematic diagram of a process for generating a target approval sequence is provided for an embodiment of this application; Figure 4 This is a schematic diagram of the structure of a cross-departmental collaborative approval decision-making system based on knowledge graph reasoning, provided as an embodiment of this application. Detailed Implementation

[0023] In the historical approval process after the cross-departmental collaborative approval rules were updated, the existing technology relied on a crude global processing strategy, which led to a key flaw: when all historical approvals were required to go through the entire process again, a large number of approval nodes that were not actually affected by the rule changes were forced to undergo repeated approvals, resulting in a serious waste of approval resources; while when a general exemption strategy was adopted, some nodes that should have been reviewed due to substantial changes in the rules were missed, creating compliance risk exposure.

[0024] This contradiction stems from the fact that existing technologies lack the ability to model the temporal evolution of approval rule versions, making it impossible to distinguish the effective boundaries of approval rules of various departments at different historical moments. Consequently, it is impossible to accurately locate which historical approval nodes have truly become invalid under the new rules, and there is an urgent need for an accurate identification method based on temporal knowledge graph reasoning.

[0025] To address the aforementioned issues, this application proposes a cross-departmental collaborative approval decision-making method based on knowledge graph reasoning. Its core lies in constructing a knowledge graph that includes time attributes, rule version difference identification, and subgraph isomorphic matching of historical approval paths to achieve precise location of failed approval nodes and minimize supplementary approvals.

[0026] Specifically, a knowledge graph is first constructed by including approval rule nodes, approval authority nodes, approval item nodes, and historical approval nodes, and the completion time of historical approval nodes is marked to enable the graph to perform temporal modeling. When the approval rule version is updated, the graph is traversed and the difference operation is performed starting from the old and new version rule nodes respectively to accurately extract the set of nodes with rule version differences.

[0027] Subsequently, for each historical approval node, its approval completion time is mapped to the applicable rule version at that time, and the approval path under that version is extracted. Failed approval nodes are accurately identified by determining whether the approval path includes differing nodes. Finally, a depth-first traversal is performed based on the approval dependencies of the failed nodes to generate a target approval sequence sorted by the number of path hops.

[0028] This method abandons the traditional global re-approval or global exemption mode. By identifying version differences in temporal knowledge graphs and matching subgraph isomorphisms, it ensures that supplementary approvals are triggered only for approval nodes that are truly affected by rule changes. This avoids the inefficiency caused by a large number of unnecessary duplicate approvals and eliminates the compliance risk exposure caused by blanket exemptions. It also solves the problem of inaccurate identification of failure nodes caused by the lack of temporal modeling capabilities in existing technologies, and significantly improves the accuracy and efficiency of cross-departmental collaborative approval processing after rule version updates.

[0029] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0030] To address the problems of existing technologies, embodiments of this application provide a cross-departmental collaborative approval decision-making method, apparatus, device, computer storage medium, and computer program product based on knowledge graph reasoning. The cross-departmental collaborative approval decision-making method based on knowledge graph reasoning provided in this application embodiment is described below.

[0031] Figure 1 This illustration shows a flowchart of a cross-departmental collaborative approval decision-making method based on knowledge graph reasoning, provided in one embodiment of this application. Figure 1 As shown, the method includes: S101. Obtain a knowledge graph constructed with approval authority nodes, approval rule nodes, approval item nodes, and historical approval nodes that have been completed as nodes, and permission association relationships, rule application relationships, and approval dependency relationships as edges. Historical approval nodes include the approval completion time.

[0032] Approval authority nodes refer to the approval authority that various departments within an enterprise possess in cross-departmental collaborative approval processes, including approval authority from the finance department, legal department, and business department, etc. Approval rule nodes refer to different versions of the enterprise's approval rules that are effective at different times, with each version corresponding to a complete set of approval standards and process requirements.

[0033] Approval item nodes refer to specific types of items requiring cross-departmental collaborative approval, including procurement applications, contract signings, budget applications, and project initiation. Historical approval nodes refer to historical records of completed approvals in enterprise project management, with each historical approval node recording the status of a specific approval item completed by a specific department at a specific point in time.

[0034] Permission association refers to the association between approval item nodes and approval authority nodes, indicating the departmental approval authority required for an approval item. Rule applicability refers to the applicability between approval rule nodes and approval item nodes, indicating the approval items to which the rule version applies. Approval dependency refers to the dependency between different approval authority nodes, indicating that a department's approval depends on the approval result of another department.

[0035] Specifically, the approval authority data of each department is first obtained from the organizational structure management module of the enterprise project management system. The approval authority identifier of each department is extracted, and the average time taken by the department to complete historical approval tasks is calculated from the historical approval record database as the approval duration attribute to construct a set of approval authority nodes. Then, retrieve the historical version change records of the approval rules from the approval rule management module, extract the identifier, effective time, and expiration time of each rule version to construct an approval rule node set. .

[0036] Next, the enterprise-defined approval item type data is obtained from the approval item configuration module, and a set of approval item nodes is constructed. Simultaneously, complete historical approval records are retrieved from the approval record database, and information such as the approval item, approval department, and approval completion time for each record is extracted to construct a set of historical approval nodes. Each historical approval node Carry the approval completion time attribute .

[0037] Then, based on the approval rule configuration and item definition, a set of permission association edges is established between approval items and approval permissions. Establish a set of rule application relationships between approval rules and approval items. Finally, based on the inter-departmental approval sequence requirements defined in the enterprise approval process, a set of approval dependency edges is established between approval authority nodes. .

[0038] Historical approval nodes Through implicit attribute mapping and the set of approval item nodes and approval permission node set Relatedly, during graph retrieval, the corresponding approval item node and approval authority node in the graph can be directly located by reading the record attributes of historical approval nodes. This involves combining the aforementioned node sets. Sum of edges Organization as a graph structure Complete the construction of the knowledge graph.

[0039] Figure 2 This is a schematic diagram of the structure of the knowledge graph provided in an embodiment of this application. For example... Figure 2 As shown, the topology node examples in the diagram include nodes belonging to the approval rule node set. Internal discrete approval rule nodes and It belongs to the set of approval item nodes. Internal discrete approval item nodes and It belongs to the set of historical approval nodes. Internal discrete historical approval nodes and and the set of nodes belonging to the approval authority Internal discrete approval authority nodes , and .

[0040] Discrete approval rule nodes , Along the rule application relationship edges respectively Directly pointing to the corresponding discrete approval item node , Discrete approval item nodes , Along the permission association edges respectively Many-to-many branching points to the corresponding discrete approval authority nodes. , and Each discrete approval authority node , , Along the approval dependency edge Sequential sequential guidance. Discrete historical approval nodes. and Each node precisely points to a discrete approval item node that matches its completion history along the implicit mapping edges in the cross-level logical dimension. , and discrete approval authority nodes , .

[0041] S102. Taking the approval rule node before the update of the approval rule version as the first approval node and the updated approval rule node as the second approval node, start from the first approval node and the second approval node in the knowledge graph respectively, traverse along the rule application relationship edge to the approval item node, and then traverse along the permission relationship edge to each approval permission node to obtain the first node set and the second node set.

[0042] Optionally, step S102, which takes the approval rule node before the update as the first approval node and the updated approval rule node as the second approval node, and traverses along the rule application relationship edge to the approval item node and then along the permission association edge to each approval permission node in the knowledge graph to obtain the first node set and the second node set, can specifically include: S1021. Starting from the first approval node in the knowledge graph, traverse along the rule application relationship edge to all approval item nodes connected to the first approval node. Then, from each approval item node, traverse along the permission relationship edge to all approval permission nodes connected to the approval item node. Construct the first node set based on all approval permission nodes.

[0043] The first approval node refers to the approval rule node that took effect before the update of the approval rule version and corresponds to the old version of the approval rule recorded in the knowledge graph. The first node set refers to the set of all approval permission nodes reached after starting from the first approval node and traversing through two hops: the rule application relationship edge and the permission association relationship edge.

[0044] Specifically, firstly in knowledge graphs The first approval node is identified as the approval rule node before the version update. Then, starting from the first approval node, the rules are applied along the set of rule applicability edges. The first hop traversal is performed on the edges in the process, reaching all approval item nodes connected to the approval rule node through the rule application relationship edge, thus obtaining the set of intermediate item nodes.

[0045] Next, starting from each approval item node in the set of intermediate item nodes, follow the set of permission association edges. The edges in the array are traversed a second time to reach all approval authority nodes connected to these approval item nodes through permission association edges. Finally, all approval authority nodes obtained from the second traversal are collected and summarized to construct the first node set. The traversal process employs a breadth-first search strategy to ensure that all approval authority nodes reachable from the first approval node are visited.

[0046] S1022. Starting from the second approval node in the knowledge graph, traverse along the rule application relationship edge to all approval item nodes connected to the second approval node, and then traverse along the permission relationship edge from each approval item node to all approval permission nodes connected to the approval item node. Construct the second node set based on all approval permission nodes.

[0047] The second approval node refers to the approval rule node that takes effect after the approval rule version is updated, and corresponds to the new version of the approval rule recorded in the knowledge graph. The second node set refers to the set of all approval permission nodes reached after starting from the second approval node and traversing through two hops via the rule application relationship edge and the permission association relationship edge.

[0048] Specifically, firstly in knowledge graphs The node representing the updated approval rule version is identified as the second approval node. Then, starting from the second approval node, the edges along the rule application relationship set are used. The first hop traversal is performed on the edges in the process, reaching all approval item nodes connected to the approval rule node through the rule application relationship edge, thus obtaining the set of intermediate item nodes.

[0049] Next, starting from each approval item node in the set of intermediate item nodes, follow the set of permission association edges. The edges in the node are traversed in a second hop, reaching all approval authority nodes connected to these approval item nodes through permission association edges. Finally, all approval authority nodes obtained from the second hop traversal are collected and summarized to construct a second node set. The traversal strategy remains consistent with the process of constructing the first node set, employing breadth-first search to ensure completeness.

[0050] This embodiment accurately extracts the first node set and the second node set, and fully restores the approval authority scope of each department before and after the rule version update.

[0051] S103. Perform a difference operation on the first node set and the second node set to obtain the third node set. Determine the rule version corresponding to the historical approval node under the approval completion time. Extract the approval path traversed by the historical approval node under the rule version from the knowledge graph.

[0052] Optionally, the process of performing a difference operation on the first node set and the second node set to obtain the third node set in step S103 may specifically include: S1031. Extract the approval authority nodes that do not exist in the second node set from the first node set to obtain the first difference set. Extract the approval authority nodes that do not exist in the first node set from the second node set to obtain the second difference set.

[0053] The first difference set refers to the set of approval authority nodes that exist in the first node set but not in the second node set, representing departmental approval authority that was required in the old version of the approval rules but has been cancelled in the new version. The second difference set refers to the set of approval authority nodes that exist in the second node set but not in the first node set, representing departmental approval authority that has been added in the new version of the approval rules but did not exist in the old version.

[0054] Specifically, firstly, for the first node set Iterate through each approval permission node in the second node set and determine whether the node exists in the second node set. In the middle. If a certain approval authority node and If the node is found to be a match, then add that node to the first difference set. After completing the traversal, the first difference set is obtained. .

[0055] Then for the second node set Iterate through each approval permission node in the set and determine whether the node exists in the first node set. In the middle. If a certain approval authority node and If the node is found to be a duplicate node, then add that node to the second difference set. After completing the traversal, the second difference set is obtained. The difference operation uses set traversal and comparison to determine whether a node exists in the target set by checking the consistency of its node identifier.

[0056] S1032. By merging the first difference set and the second difference set, the third node set is obtained.

[0057] The third node set refers to the set of approval authority nodes after merging the first difference set and the second difference set, and includes all approval authority nodes that have changed before and after the rule version update, that is, both canceled department approval authority and newly added department approval authority.

[0058] Specifically, first obtain the first difference set. Second difference set Then, all approval authority nodes in the two difference sets are merged. If a certain approval authority node... Existence or If a node is found in any set in the set, it is added to the third node set. The merge operation uses the set union operation to obtain the third node set. This set represents the complete range of all differing nodes before and after the approval rule version update.

[0059] Optionally, the approval rule node includes the effective time and expiration time of the approval rule version. Step S103, which determines the rule version corresponding to the historical approval node at the approval completion time, and extracts the approval path traversed by the historical approval node under the rule version in the knowledge graph, may specifically include: S1033. Construct the time interval for each approval rule node based on the effective time and the expiration time. When the approval completion time of a historical approval node falls within the time interval of any approval rule node, determine the approval rule version of the corresponding approval rule node as the rule version.

[0060] Approval rule nodes include the effective date and expiration date of an approval rule version. The effective date refers to the point in time when a particular approval rule version begins to take effect, and the expiration date refers to the point in time when a particular approval rule version ceases to be effective. A time interval refers to the range of time constituted by the effective date and expiration date of a particular approval rule node, representing the validity period of that rule version within the enterprise's approval process. A rule version refers to the approval rule version actually applicable at the time of approval completion for historical approval nodes, determined by comparing the approval completion time with the time intervals of each approval rule node.

[0061] Specifically, starting with the set of approval rule nodes Get the effective time carried by each approval rule node. and expiration time Construct time intervals Then from the historical approval node set Obtain each historical approval node Approval completion time .

[0062] Next, for each historical approval node, its approval completion time is recorded. Time intervals of all approval rule nodes Compare them one by one to determine whether they meet the requirements. When the approval is completed Falling into a certain approval rule node time interval At that time, determine the node of the approval rule. The corresponding approval rule version is the rule version for that historical approval node. .

[0063] If the approval completion time If it does not fall within any defined time interval, then the source is determined according to the temporal logic: if Earlier than all known effective dates If so, the approval rule version corresponding to the approval rule node with the earliest effective time will be directly determined as the rule version of that historical approval node; if The gap between two adjacent time intervals is compared. The absolute difference between the expiration time of the previous version and the effective time of the next version at the endpoints of adjacent time intervals is used to determine the rule version corresponding to the nearest neighboring approval rule node with the smaller absolute difference. .

[0064] S1034. Starting from the approval rule node corresponding to the rule version in the knowledge graph, traverse along the rule application relationship edge to the approval item node, and then traverse along the permission relationship edge to the approval permission node that is the same as the approval permission node associated with the historical approval node, to obtain the approval path consisting of the approval rule node, approval item node, and approval permission node.

[0065] An approval path refers to a graph path that starts from the approval rule node corresponding to the rule version, passes through approval item nodes, and finally reaches the approval authority node associated with the historical approval node. This path includes the approval rule nodes, approval item nodes, approval authority nodes, and the rule application relationship edges and authority relationship edges connecting these nodes. The approval authority node associated with the historical approval node refers to the approval authority node corresponding to the department that actually completed the approval, as recorded by the historical approval node upon completion.

[0066] Specifically, firstly, according to the rule version determined in step S1033 Location knowledge graph The corresponding approval rule node. Then, starting from this approval rule node, along the set of rule application relationship edges. The first hop traversal is performed on the edges in the algorithm to reach the approval item node that is connected to the approval rule node through the rule application relationship edge.

[0067] Next, starting from each approval item node, we set up the collection along the permission association edges. The edges in the algorithm are traversed a second time to reach the approval authority node connected to the approval item node through the permission association edge. Historical approval nodes are also retrieved. The recorded approval authority node identifiers are used to match the reached approval authority nodes with the approval authority nodes recorded in the historical approval node records. Finally, the successfully matched approval authority nodes, along with all nodes and edges traversed during the traversal from the approval rule node to the approval authority node, are organized into an approval path. .

[0068] This embodiment clarifies the specific scope of rule changes, restores the true state of historical approvals under the rule system at that time, and realizes accurate correlation analysis between historical approvals and rule version differences.

[0069] S104. By performing subgraph isomorphic matching on the approval path and the third node set, the approval failure node is determined. Starting from the approval failure node, the knowledge graph is traversed in a depth-first manner along the approval dependency edges to obtain the candidate approval sequence.

[0070] Optionally, step S104, which determines the approval failure node by performing subgraph isomorphic matching on the approval path and the third node set, and obtains the candidate approval sequence by performing a depth-first traversal of the knowledge graph along the approval dependency edges starting from the approval failure node, may specifically include: S1041. Based on all approval authority nodes of the approval path and the permission association edges of the nodes in the knowledge graph, construct an approval path subgraph. Based on all approval authority nodes of the third node set and the permission association edges of the nodes in the knowledge graph, construct a difference subgraph.

[0071] Approval path subgraph refers to the subgraph of the approval path The graph includes all the approval authority nodes and the edges connecting these approval authority nodes to each other in the knowledge graph through permission associations. It represents the scope of approval authority coverage and connection relationships that historical approval nodes actually experienced under the rule version at that time.

[0072] A difference subgraph is a subgraph consisting of a set of third nodes. The local graph structure includes all the approval authority nodes and the edges connecting these approval authority nodes to each other in the knowledge graph through permission association relationships. It represents all the approval authority nodes that have changed before and after the rule version update and their connection relationships.

[0073] Specifically, the approval path extracted in step S1034 is first obtained. From the approval path Extract all approval authority nodes and their associated approval item nodes from the knowledge graph, and then... Find the set of edges that connect these nodes based on their permissions. The interconnected edges, together with the extracted approval authority nodes, approval item nodes, and corresponding permission relationship edges, form an approval path subgraph. .

[0074] Next, obtain the third node set constructed in step S1032. ,from Extract all approval authority nodes and backtrack to obtain the approval item nodes associated with these approval authority nodes in the knowledge graph. Find the set of edges that link these approval item nodes and approval authority nodes through permission relationships. The interconnected edges that link the aforementioned nodes and their corresponding permission association edges together form a difference subgraph. .

[0075] S1042. When there is an approval authority node in the approval path subgraph that is the same as at least one node in the difference subgraph, the historical approval node corresponding to the approval path is determined to be the approval invalidation node.

[0076] An approval failure node refers to a historical approval node in which the approval path subgraph and the difference subgraph have the same approval authority node in the subgraph isomorphic matching. It means that the scope of approval authority experienced by the historical approval node under the rule version applicable when the approval was completed overlaps with the scope of approval authority affected by the rule version update. In other words, the historical approval node is affected by the rule version update and has compliance issues under the new version of the approval rule.

[0077] Specifically, first, the approval path subgraph constructed in step S1041 is obtained. and difference subgraph Then iterate through the difference subgraph. For each approval authority node in the graph, determine whether the node exists in the approval path subgraph. middle.

[0078] The judgment method is to first analyze the difference subgraphs. The node identifier and approval path subgraph for each approval authority node. The node identifiers of all approval authority nodes are mapped and compared to find any intersection nodes with identical node identifiers. If approval authority nodes with the same node identity exist, the isomorphism constraint must be strictly enforced, that is, the intersection node must be further verified in the difference subgraph. Approval path subgraph The topological neighbors in the model include the associated approval item nodes and the direction and type of the edges of their connected permission relationships, and whether they are completely equivalent.

[0079] If and only if the approval path subgraph Existence and difference subgraphs The approval path is determined when at least one node is identical and the local edge topology around that node is completely isomorphic. Corresponding historical approval nodes Approval failure node .

[0080] S1043. Starting from the approval failure node in the knowledge graph, perform a depth-first traversal along the approval dependency edges, and arrange all the approval permission nodes traversed in the traversal order to construct a candidate approval sequence.

[0081] A candidate approval sequence is a sequence of approval authority nodes obtained by performing a depth-first traversal of the knowledge graph starting from the approval failure node and following the approval dependency edges, and then arranging them according to the traversal access order. It represents an ordered set of all approval authority nodes that have an approval dependency relationship with the approval failure node.

[0082] Specifically, firstly, the approval failure node determined in step S1042 is obtained. Based on the implicit mapping relationships in the graph, the node corresponding to the approval failure was located. The corresponding expired approval permission node. Then, in the knowledge graph... Starting from this invalidated approval permission node, the set of approval dependency edges is formed. Edge-based depth-first traversal is initiated.

[0083] The specific process of depth-first traversal is as follows: Starting from the invalid approval permission node in the mapping, prioritize visiting the downstream approval permission nodes directly connected to it along the approval dependency edges. After visiting the node, continue to extend to deeper approval permission nodes along the approval dependency edges of that node until it is impossible to continue traversing downwards. At that point, backtrack to the previous level node and then visit other unvisited approval permission nodes at the same level. This process is repeated until all reachable approval permission nodes have been visited. Finally, all approval permission nodes recorded in the order of visit during the traversal are arranged into a candidate approval sequence. ,in The first node to be accessed for approval permissions.

[0084] This embodiment accurately identifies approval failure nodes affected by rule version updates, avoiding the inefficient process of re-examining all historical approval nodes, and fully collects all approval permission nodes that have approval dependencies on the approval failure nodes.

[0085] S105. Based on the path hop count between each approval authority node and the approval failure node in the candidate approval sequence, sort the approval authority nodes to generate the target approval sequence and send it to the corresponding approval department.

[0086] Optionally, the approval authority node includes the approval duration, and the method also includes: In the knowledge graph, traverse along the approval dependency edges from the approval failure node to each approval authority node in the candidate approval sequence to obtain the traversal path of each approval authority node. Accumulate the approval times of all approval authority nodes traversed along the traversal path to obtain the cumulative approval time of each approval authority node.

[0087] Approval duration refers to the time required for the department corresponding to the approval authority node to complete an approval, and it reflects the actual approval cycle occupied by that department in the cross-departmental collaborative approval process. Traversal path refers to the set of nodes and edges traversed in the knowledge graph from the approval failure node along the approval dependency relationship edges to reach a certain approval authority node in the candidate approval sequence, and it represents the complete approval dependency link required to reach that approval authority node.

[0088] The cumulative approval time refers to the sum of the approval times of all approval authority nodes traversed along the traversal path, and represents the total time cost required to reach the target approval authority node from the approval failure node through the traversal path.

[0089] Specifically, the candidate approval sequence constructed in step S1043 is first obtained. and the approval failure node determined in step S1042. Then, regarding the candidate approval sequence... Each approval permission node in In knowledge graphs Starting from the invalidation approval permission node in the aforementioned mapping, the set of approval dependency edges is traversed. Traversing the edges in When there are multiple arrivals When traversing connected paths, the shortest hop path with the fewest nodes is uniformly extracted as the baseline path; if there are still multiple short hop paths, the one with the largest sum of approval times is selected as the unique record, recording all nodes and edges traversed during the baseline traversal, thus obtaining the traversal path. .

[0090] Next, obtain the traversal path. The approval duration attribute carried by each of the approval permission nodes traversed will be used to iterate through the path. The approval time of all approval authority nodes is summed up to obtain the approval authority node. Cumulative approval time .

[0091] Step S105, based on the path hop count between each approval authority node and the approval failure node in the candidate approval sequence, sorts the approval authority nodes to generate a target approval sequence and sends it to the corresponding approval department, including: S1051. Based on the cumulative approval time and the path hop count between each approval authority node and the approval failure node in the candidate approval sequence, sort the approval authority nodes to generate the target approval sequence and send it to the corresponding approval department.

[0092] The path hop count refers to the number of approval dependency edges traversed in the knowledge graph from the failed approval node to a certain approval authority node in the candidate approval sequence, and represents the approval dependency hierarchy distance between the approval authority node and the failed approval node. The target approval sequence is the final supplementary approval execution sequence generated after sorting all approval authority nodes in the candidate approval sequence based on both the path hop count and the cumulative approval time.

[0093] Specifically, the candidate approval sequence constructed in step S1043 is first obtained. Approval failure point and each approval authority node Corresponding path hop count and cumulative approval time Then by path hop count Sort by main keywords and by cumulative approval time Using the secondary sorting key, the candidate approval sequence All approval authority nodes are sorted in a two-level manner: first, by the number of path hops in ascending order; then, when two approval authority nodes have the same number of path hops, they are sorted by the cumulative approval time in ascending order. Finally, the sorted sequence of approval authority nodes is used as the target approval sequence. Send it to the corresponding approval department.

[0094] Optionally, step S1051, which sorts the approval authority nodes according to the cumulative approval time and the path hop count between each approval authority node and the approval failure node in the candidate approval sequence, generates a target approval sequence and sends it to the corresponding approval department, may specifically include: S10511. Sort the approval authority nodes in the candidate approval sequence in ascending order of path hop count to obtain the initial sorting sequence.

[0095] The initial sorting sequence refers to the ordered sequence obtained by sorting all approval authority nodes in the candidate approval sequence from smallest to largest according to the number of path hops, and it represents the preliminary supplementary approval execution order based on the distance of approval dependency levels as the main sorting criterion.

[0096] Specifically, the candidate approval sequence constructed in step S1043 is first obtained. and approval failure points Then, regarding the candidate approval sequence... Each approval permission node in In knowledge graphs Based on the traversal path The statistics start from the mapped invalid approval permission node and proceed along the approval dependency edge to reach... The number of approval dependency edges traversed determines the path hop count for that approval authority node. Then, follow the path jump count. Candidate approval sequence in ascending order Sort all approval permission nodes to obtain the initial sorting sequence. .

[0097] S10512. Extract approval authority nodes with the same path jump number from the initial sorting sequence, construct the fourth node set, and sort the approval authority nodes in each fourth node set in ascending order of cumulative approval time to obtain the fifth node set.

[0098] The fourth node set refers to the set of approval authority nodes with the same number of path hops extracted from the initial sorting sequence. It represents the group of approval authority nodes that need to be further sorted by cumulative approval time when the distance between approval dependency levels is the same. The fifth node set refers to the ordered set of nodes obtained by sorting the approval authority nodes in the fourth node set in ascending order of cumulative approval time. It represents the refined sorting result based on time cost under the premise of the same number of path hops.

[0099] Specifically, firstly, the initial sorting sequence constructed in step S10511 is obtained. Traverse the initial sorted sequence All approval permission nodes in the set are grouped together, with nodes having the same number of path hops forming a fourth node set. Then for each fourth node set Get the cumulative approval time for each approval authority node in the set, and sort the fourth node set in ascending order of cumulative approval time. Sort the approval authority nodes within the node set to obtain the corresponding fifth node set. .

[0100] S10513. Use the fifth node set to replace the approval authority nodes at the corresponding positions in the initial sorting sequence, generate the target approval sequence and send it to the corresponding approval department.

[0101] Specifically, firstly, the initial sorting sequence constructed in step S10511 is obtained. and the fifth node set constructed in step S10512 Then locate the initial sorted sequence. The location interval corresponding to the fourth node set is used to determine the approval authority nodes within that location interval according to the fifth node set. The order of the nodes in the sequence is replaced, while the positions of approval authority nodes with different path hop counts remain unchanged. Finally, the sequence after replacement is used as the target approval sequence. According to the target approval sequence The node sequence will push the supplementary approval tasks to the corresponding approval departments of each node in turn.

[0102] Figure 3 This is a schematic diagram of the process for generating a target approval sequence provided in an embodiment of this application. Figure 3 As shown, in the specific operation and logical flow, the candidate sequence obtained by depth-first traversal is first sliced ​​hierarchically, and the minimum number of path jumps along the approval dependency relationship edge is counted when starting from the invalid approval authority node to reach each candidate.

[0103] Next, it is determined whether there are approval authority nodes with the same number of path jumps. If the result is yes, these nodes with the same number of jumps are classified and constructed into a fourth node set. The cumulative approval time of each node in the fourth node set is further calculated. Then, the nodes in the fourth node set are rearranged locally according to the ascending order of cumulative approval time from smallest to largest to generate a fifth node set. Finally, node replacement is implemented at the corresponding interval position of the initial sorting sequence.

[0104] If the judgment result is negative, an initial sorted sequence is generated and output directly according to the path hop count from smallest to largest. Through the above two-layer judgment and rearrangement process, the optimal generation of the target approval sequence is finally achieved, and the orderly distribution of cross-departmental collaborative supplementary approval tasks is completed in the most efficient manner while strictly and compliantly following the approval dependencies between departments.

[0105] This embodiment ensures that supplementary approvals strictly adhere to inter-departmental approval dependencies, guaranteeing process compliance. Prioritizing approval tasks with lower time costs optimizes the execution order of supplementary approvals while ensuring compliance, effectively shortening the overall supplementary approval cycle and achieving a dual improvement in approval efficiency and process accuracy.

[0106] Figure 4 This application provides a schematic diagram of the structure of a specific implementation of a cross-departmental collaborative approval decision-making system based on knowledge graph reasoning, as shown in the following embodiments. Figure 4 The system may include: The acquisition module 410 is used to acquire a knowledge graph constructed with approval permission nodes, approval rule nodes, approval item nodes and historical approval nodes that have been completed as nodes, and permission association relationships, rule application relationships and approval dependency relationships as edges. The historical approval nodes include the approval completion time. The generation module 420 is used to take the approval rule node before the approval rule version update as the first approval node and the updated approval rule node as the second approval node. In the knowledge graph, starting from the first approval node and the second approval node respectively, it traverses along the rule application relationship edge to the approval item node and then along the permission association relationship edge to each approval permission node to obtain the first node set and the second node set. The extraction module 430 is used to perform a difference operation on the first node set and the second node set to obtain the third node set, determine the rule version corresponding to the historical approval node under the approval completion time, and extract the approval path traversed by the historical approval node under the rule version in the knowledge graph. The generation module 420 is also used to determine the approval failure node by performing subgraph isomorphic matching on the approval path and the third node set, and to obtain the candidate approval sequence by performing a depth-first traversal of the knowledge graph along the approval dependency edge, starting from the approval failure node. The sending module 440 is used to sort the approval authority nodes according to the path hop count between each approval authority node and the approval failure node in the candidate approval sequence, generate the target approval sequence, and send it to the corresponding approval department.

[0107] The cross-departmental collaborative approval decision system based on knowledge graph reasoning in this application is used to implement the aforementioned cross-departmental collaborative approval decision method based on knowledge graph reasoning. Therefore, the specific implementation of the cross-departmental collaborative approval decision system based on knowledge graph reasoning can be found in the embodiment section of the cross-departmental collaborative approval decision method based on knowledge graph reasoning above. The specific implementation can be referred to the description of the corresponding embodiments, which will not be repeated here.

[0108] Electronic devices may include processors and memory storing computer program instructions.

[0109] Specifically, the processor may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0110] The memory may include a large-capacity storage device for data or instructions. For example, and not limitingly, the memory may include a hard disk drive (HDD), a floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory may include removable or non-removable (or fixed) media. Where appropriate, the memory may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, the memory is non-volatile solid-state memory.

[0111] Memory may include read-only memory (ROM), random access memory (RAM), disk storage media devices, optical storage media devices, flash memory devices, and electrical, optical, or other physical / tangible memory storage devices. Therefore, typically, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to the first aspect of this disclosure.

[0112] The processor reads and executes computer program instructions stored in memory to implement any of the cross-departmental collaborative approval decision-making methods based on knowledge graph reasoning in the above embodiments.

[0113] In one example, the electronic device may also include a communication interface and a bus. The processor, memory, and communication interface are connected via the bus and communicate with each other.

[0114] The communication interface is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0115] A bus, including hardware, software, or both, couples components of an online data traffic metering device together. For example, and not limitingly, a bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, a bus may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, this application contemplates any suitable bus or interconnect.

[0116] The electronic device can execute the cross-departmental collaborative approval decision-making method based on knowledge graph reasoning in the embodiments of this application, thereby realizing the cross-departmental collaborative approval decision-making method based on knowledge graph reasoning described in conjunction with the accompanying drawings.

[0117] Furthermore, in conjunction with the cross-departmental collaborative approval decision-making method based on knowledge graph reasoning in the above embodiments, this application embodiment can provide a computer-readable storage medium for implementation. This computer-readable storage medium stores computer program instructions; when executed by a processor, these computer program instructions implement any of the cross-departmental collaborative approval decision-making methods based on knowledge graph reasoning in the above embodiments.

[0118] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0119] The functional blocks shown in the above-described structural diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0120] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0121] The aspects of this application have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations 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, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0122] The foregoing has provided a detailed description of a cross-departmental collaborative approval decision-making method and system based on knowledge graph reasoning, as provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A cross-departmental collaborative approval decision-making method based on knowledge graph reasoning, characterized in that, include: Obtain a knowledge graph constructed with approval authority nodes, approval rule nodes, approval item nodes, and historical approval nodes that have been completed as nodes, and permission association relationships, rule application relationships, and approval dependency relationships as edges. The historical approval nodes include the approval completion time. Taking the approval rule node before the update of the approval rule version as the first approval node and the updated approval rule node as the second approval node, in the knowledge graph, starting from the first approval node and the second approval node respectively, traverse along the rule application relationship edge to the approval item node and then traverse along the permission association relationship edge to each approval permission node to obtain the first node set and the second node set. A third node set is obtained by performing a difference operation on the first node set and the second node set. The rule version corresponding to the historical approval node under the approval completion time is determined. The approval path traversed by the historical approval node under the rule version is extracted from the knowledge graph. By performing subgraph isomorphic matching on the approval path and the third node set, the approval failure node is determined. Starting from the approval failure node, the knowledge graph is traversed in a depth-first manner along the approval dependency edges to obtain the candidate approval sequence. Based on the path hop count between each approval authority node in the candidate approval sequence and the approval failure node, the approval authority nodes are sorted to generate a target approval sequence, which is then sent to the corresponding approval department.

2. The method according to claim 1, characterized in that, The approval authority node includes the approval duration, and the method further includes: In the knowledge graph, traverse along the approval dependency edges from the approval failure node to each approval authority node in the candidate approval sequence to obtain the traversal path of each approval authority node; The cumulative approval time of each approval authority node is obtained by summing the approval times of all the approval authority nodes traversed on the traversal path. The step of sorting the approval authority nodes according to the path hop count between each approval authority node in the candidate approval sequence and the approval failure node, generating a target approval sequence, and sending it to the corresponding approval department includes: Based on the cumulative approval time and the path hop count between each approval authority node in the candidate approval sequence and the approval failure node, the approval authority nodes are sorted to generate a target approval sequence and sent to the corresponding approval department.

3. The method according to claim 2, characterized in that, The step of sorting the approval authority nodes according to the cumulative approval time and the path hop count between each approval authority node in the candidate approval sequence and the approval failure node to generate a target approval sequence and sending it to the corresponding approval department includes: Sort the approval authority nodes in the candidate approval sequence in ascending order of the path hop count to obtain the initial sorting sequence; Extract the approval authority nodes with the same path jump number from the initial sorting sequence, construct a fourth node set, and sort the approval authority nodes in each fourth node set in ascending order of cumulative approval time to obtain a fifth node set; The fifth node set is used to replace the approval authority nodes at the corresponding positions in the initial sorting sequence, and the target approval sequence is generated and sent to the corresponding approval department.

4. The method according to claim 1, characterized in that, The first approval node is defined as the approval rule node before the update, and the second approval node is defined as the updated approval rule node. In the knowledge graph, starting from the first and second approval nodes respectively, the process traverses along the rule application relationship edges to the approval item nodes, and then along the permission association edges to each approval permission node, resulting in a first node set and a second node set, including: Starting from the first approval node in the knowledge graph, the system traverses along the rule application relationship edge to all approval item nodes connected to the first approval node, and then traverses along the permission association relationship edge from each approval item node to all approval permission nodes connected to the approval item node, and constructs a first node set based on all the approval permission nodes. Starting from the second approval node in the knowledge graph, the process traverses along the rule application relationship edge to all approval item nodes connected to the second approval node, and then traverses along the permission association edge from each approval item node to all approval permission nodes connected to the approval item node, thus constructing a second node set based on all the approval permission nodes.

5. The method according to claim 1, characterized in that, The step of performing a difference operation on the first node set and the second node set to obtain the third node set includes: Extract the approval authority nodes that do not exist in the second node set from the first node set to obtain the first difference set; extract the approval authority nodes that do not exist in the first node set from the second node set to obtain the second difference set. The third node set is obtained by merging the first difference set and the second difference set.

6. The method according to claim 1, characterized in that, The approval rule node includes the effective time and expiration time of the approval rule version; The step of determining the rule version corresponding to the historical approval node at the approval completion time, and extracting the approval path traversed by the historical approval node under the rule version from the knowledge graph, includes: Based on the effective time and the expiration time, a time interval is constructed for each approval rule node. When the approval completion time of the historical approval node falls into the time interval of any of the approval rule nodes, the approval rule version of the corresponding approval rule node is determined as the rule version. In the knowledge graph, starting from the approval rule node corresponding to the rule version, the process traverses along the rule application relationship edge to the approval item node, and then along the permission association relationship edge to the approval permission node that is the same as the approval permission node associated with the historical approval node, thus obtaining the approval path composed of the approval rule node, approval item node, and approval permission node.

7. The method according to claim 1, characterized in that, The process involves determining approval failure nodes by performing subgraph isomorphic matching on the approval path and the third node set. Starting from the approval failure node, a depth-first traversal of the knowledge graph is performed along the approval dependency edges to obtain a candidate approval sequence, including: Based on all the approval authority nodes of the approval path and the permission association edges of the nodes in the knowledge graph, an approval path subgraph is constructed, and a difference subgraph is constructed based on all the approval authority nodes of the third node set and the permission association edges of the nodes in the knowledge graph. When there is an approval authority node in the approval path subgraph that is the same as at least one node in the difference subgraph, the historical approval node corresponding to the approval path is determined to be an approval failure node. Starting from the approval failure node in the knowledge graph, a depth-first traversal is performed along the approval dependency edges, and all the approval authority nodes traversed are arranged in the traversal order to construct the candidate approval sequence.

8. A cross-departmental collaborative approval and decision-making system based on knowledge graph reasoning, characterized in that, include: The acquisition module is used to acquire a knowledge graph constructed with approval permission nodes, approval rule nodes, approval item nodes, and historical approval nodes that have been completed as nodes, and permission association relationships, rule application relationships, and approval dependency relationships as edges. The historical approval nodes include the approval completion time. The generation module is used to take the approval rule node before the approval rule version update as the first approval node and the updated approval rule node as the second approval node. In the knowledge graph, starting from the first approval node and the second approval node respectively, it traverses along the rule application relationship edge to the approval item node and then along the permission association relationship edge to each approval permission node to obtain the first node set and the second node set. The extraction module is used to perform a difference operation on the first node set and the second node set to obtain a third node set, determine the rule version corresponding to the historical approval node under the approval completion time, and extract the approval path traversed by the historical approval node under the rule version in the knowledge graph. The generation module is also used to determine the approval failure node by performing subgraph isomorphic matching on the approval path and the third node set, and to obtain the candidate approval sequence by performing a depth-first traversal of the knowledge graph along the approval dependency edges, starting from the approval failure node. The sending module is used to sort the approval authority nodes according to the path hop count between each approval authority node in the candidate approval sequence and the approval failure node, generate a target approval sequence, and send it to the corresponding approval department.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the cross-departmental collaborative approval decision-making method based on knowledge graph reasoning as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, enables the cross-departmental collaborative approval decision-making method based on knowledge graph reasoning as described in any one of claims 1 to 7.

Images