A financial process mining method and related device based on an event graph

Abstract

The present application discloses a financial process mining method and related device based on an event graph. A business operation event graph is constructed according to the operation process of a financial system. According to the system operation logs generated by a user using the financial system, the number of occurrences of adjacent nodes in the business operation event graph is determined and used as the weight of the node connection corresponding to the adjacent nodes. Based on the weight of the connection between the target node and the corresponding node, the financial process is mined from the business operation event graph. Thus, it is no longer necessary to rely on a large amount of expert knowledge and complex predefined event-condition-action rule information. Only based on the relationship between operation steps in the operation process of the financial system, the financial process can be mined, reducing the development difficulty and cost.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular, to a financial process mining method and related device based on an event graph. Background Art

[0002] China is in the period of deep integration of digital technology and economic society, and the digital transformation and upgrading of traditional industries. Business requirements are constantly changing, and enterprises must continuously invest a large amount of manpower to design new processes to adapt to new business changes. Process mining technology is a new technology based on process modeling methods and data mining technology applied to business process management, and is one of the key elements of financial digital transformation.

[0003] Robotic Process Automation (RPA for short) is one of the key elements of financial digital transformation. However, from the historical experience of robotic process mining, it is not very clear which processes can be automated by RPA in the daily work of financial personnel at present. It is necessary to repeatedly communicate and investigate between demand personnel and financial personnel to complete the sorting and mining of business processes, and the efficiency is low. Summary of the Invention

[0004] In view of the above problems, the present application provides a financial process mining method and related device based on an event graph to improve the efficiency of financial process mining.

[0005] Based on this, the embodiments of the present application disclose the following technical solutions:

[0006] On the one hand, the embodiments of the present application provide a financial process mining method based on an event graph, and the method includes:

[0007] Construct a business operation event graph according to the operation process of the financial system; the business operation event graph includes nodes representing operation steps and node connections identifying the execution order between the operation steps;

[0008] Determine the number of occurrences of adjacent nodes in the business operation event graph according to the system operation logs generated by the user using the financial system, and use it as the weight of the node connection corresponding to the adjacent node;

[0009] Mine the financial process from the business operation event graph based on the target node and the weight of the corresponding node connection, where the target node is one of the multiple nodes included in the business operation event graph.

[0010] Optionally, before mining the financial process from the business operation event graph, the method further includes:

[0011] Determine a hypothetical event chain based on the first m nodes and the last n nodes connected to the target node, where the target node is one of the multiple nodes included in the business operation event graph;

[0012] Obtain the first weight sum of the node connections included in the hypothetical event chain, and the second weight of the target node connection connected to the target node in the hypothetical event chain;

[0013] If the first weight sum and the second weight do not meet the preset conditions, delete the target node connection to obtain an updated business operation event graph.

[0014] Optionally, the method further includes:

[0015] Match the financial process with the processes included in the system operation log;

[0016] If it is a complete match, the financial process mining is successful.

[0017] Optionally, constructing the business operation event graph according to the operation process of the financial system includes:

[0018] Construct an initial business operation event graph according to the system user manual of the financial system;

[0019] Extract knowledge chains from the system operation logs generated by users using the financial system. Each knowledge chain is used to identify an operation process, including multiple operation steps and the execution order between the operation steps;

[0020] Add the knowledge chain to the initial business operation event graph through knowledge alignment to generate the business operation event graph.

[0021] Optionally, mining the financial process from the business operation event graph based on the weights of the target node and the corresponding node connections includes:

[0022] Establish a starting point set including the target node;

[0023] Taking the nodes included in the starting point set as starting points, determine a path set having a connection relationship with the nodes included in the starting point set. The path set includes the node connections from each node in the starting point set to other nodes and the corresponding weights, and the other nodes are the nodes in the business operation event graph except the nodes included in the starting point set;

[0024] Add the nodes corresponding to the node connections with the largest weights in the path set to the starting point set, and execute the step of taking the nodes included in the starting point set as starting points to determine the path set having a connection relationship with the nodes included in the starting point set until no nodes are added to the starting point set;

[0025] Generate a financial process according to the set of starting points.

[0026] On the other hand, an embodiment of the present application provides a financial process mining device based on an event graph, and the device includes: a construction unit, a determination unit, and a mining unit;

[0027] The construction unit is configured to construct a business operation event graph according to the operation process of the financial system; the business operation event graph includes nodes representing operation steps and node connections identifying the execution order between the operation steps;

[0028] The determination unit is configured to determine the number of occurrences of adjacent nodes in the business operation event graph according to the system operation logs generated by the user using the financial system, and use it as the weight of the node connection corresponding to the adjacent nodes;

[0029] The mining unit is configured to mine a financial process from the business operation event graph based on the target node and the weight of the corresponding node connection, and the target node is one of the multiple nodes included in the business operation event graph.

[0030] Optionally, the device further includes an update unit, configured to:

[0031] Determine a hypothetical event chain according to the first m nodes and the last n nodes connected to the target node, where the target node is one of the multiple nodes included in the business operation event graph;

[0032] Obtain the sum of the first weights of the node connections included in the hypothetical event chain, and the second weight of the target node connection connected to the target node in the hypothetical event chain;

[0033] If the sum of the first weights and the second weight do not meet the preset conditions, delete the target node connection to obtain an updated business operation event graph.

[0034] On the other hand, the present application provides a computer device, and the device includes a processor and a memory:

[0035] The memory is used to store program code and transmit the program code to the processor;

[0036] The processor is configured to execute the method described in the above aspect according to the instructions in the program code.

[0037] On the other hand, the present application provides a computer-readable storage medium, and the computer-readable storage medium is used to store a computer program, and the computer program is used to execute the method described in the above aspect.

[0038] On the other hand, an embodiment of the present application provides a computer program product or a computer program. The computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the method described in the above aspect.

[0039] Compared with the prior art, the advantages of the above technical solution of the present application are as follows:

[0040] Construct a business operation event graph according to the operation process of the financial system. Determine the number of occurrences of adjacent nodes in the business operation event graph based on the system operation logs generated by users using the financial system, and use it as the weight of the node connection corresponding to the adjacent node. Based on the weight of the connection between the target node and the corresponding node, mine the financial process from the business operation event graph. Thus, it is no longer necessary to rely on a large amount of expert knowledge and complex predefined event-condition-action rule information. Only according to the relationship between operation steps in the operation process of the financial system, the financial process can be mined, reducing the development difficulty and cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments recorded in the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to these drawings.

[0042] Figure 1 It is a flowchart of a financial process mining method based on an event graph provided by an embodiment of the present application;

[0043] Figure 2 It is a schematic diagram of a business operation event graph provided by an embodiment of the present application;

[0044] Figure 3 It is a schematic diagram of the generation of a business operation event graph provided by an embodiment of the present application;

[0045] Figure 4 It is a schematic diagram of a business operation event graph provided by an embodiment of the present application;

[0046] Figure 5 It is a schematic diagram of generating a financial process provided by an embodiment of the present application;

[0047] Figure 6 It is a schematic diagram of a financial process mining device based on an event graph provided by an embodiment of the present application;

[0048] Figure 7 The structural diagram of a computer device provided by an embodiment of the present application. Detailed implementation manners

[0049] In order to enable those skilled in the art to better understand the solution of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0050] In view of the low efficiency of financial process mining, in the related art, a configurable process mining method based on event logs is mainly used to complete the process mining of event logs by artificially defining a large amount of event-condition-action rule information. However, the implementation of the configurable process mining method based on event logs depends on a large amount of expert knowledge and complex predefined event-condition-action rule information, resulting in a high development cost.

[0051] Based on this, the embodiments of the present application provide a financial process mining method based on an event graph. By utilizing the good ability of the event graph to organize and utilize information, the non-structured data and structured data knowledge are integrated to construct a complete business operation process event graph, and based on this, the financial process mining is realized. It does not need to rely on a large amount of expert knowledge and complex predefined event-condition-action rule information, and the financial process can be mined only according to the relationship between operation steps in the operation process of the financial system, reducing the development difficulty and cost.

[0052] In order to make the purpose, technical solutions and advantages of the invention clearer, the following further details the invention with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the invention, rather than to limit the invention. On the premise of conforming to the technical concept of the invention, the features, structures, characteristics or expression details described in a specific embodiment may not be excluded from being combined in a suitable manner or in more other embodiments. The flowcharts shown in the accompanying drawings are only for illustrative purposes, not necessarily including the content and operation steps, nor necessarily in the order described. For example, some operation steps can be decomposed, while some operation steps can be combined or partially combined, so the actual execution order may be changed according to the actual situation.

[0053] The following combines Figure 1 , to introduce a financial process mining method based on an event graph provided by an embodiment of the present application. Refer to Figure 1, This figure is a flowchart of a financial process mining method based on an event graph provided by an embodiment of the present application. This method may include S101 - S103.

[0054] S101: Construct a business operation event graph according to the operation process of the financial system.

[0055] Currently, there is a large amount of valid pattern knowledge hidden in the financial system log data. By using process mining technology, these data can be analyzed and mined from multiple views such as control flow, organization, and resources, which can better realize the link between event data and process models, automatically mine the processes with high daily repetition and clear regularity of financial personnel, and give suggestions on the automation of relevant processes for financial personnel.

[0056] There are many operations in the financial system, and different operations constitute different operation processes. For example, operation step A - operation step B - operation step C constitutes an operation process, and operation step B - operation step C - operation step A constitutes another operation process.

[0057] Based on different operation processes, a business operation event graph for the financial system can be constructed. Among them, the business operation event graph includes multiple nodes, each node represents an operation step, and there is an order between operation steps. Based on the operation order (node relationship), two nodes can be connected by a directed line. As Figure 2 shown, this figure is a business operation event graph composed of the above two operation processes.

[0058] As a possible implementation manner, an embodiment of constructing a business operation event graph according to the operation process of the financial system is introduced below.

[0059] S1011: Construct an initial business operation event graph according to the system user manual of the financial system.

[0060] It should be noted that the system user manual of the financial system is the content constructed to guide users on how to use the financial system when the financial system is built. It can be understood that the system user manual of the financial system belongs to unstructured data.

[0061] The system user manual records multiple operation steps and the execution order between operation steps. Based on the system user manual, an initial business operation event graph can be constructed. Among them, the initial business operation event graph includes multiple nodes, each node represents an operation step, and node connections indicating the execution order between operation steps. Compared with the business operation event graph, the initial business operation event graph includes less content. In order to improve the initial business operation event graph, a knowledge chain will be added on the basis of the initial business operation event graph. The following is a specific description.

[0062] S1012: Extract a knowledge chain based on the system operation logs generated by the user using the financial system.

[0063] In practical applications, when users, or financial personnel, use the financial system, system operation logs will be generated. The system operation logs record when and how the users use the financial system. Through the system operation logs, it can be clearly determined which operation processes in the daily work of users can be automated through RPA, and then the knowledge chain can be extracted through the system operation logs.

[0064] Among them, a knowledge chain represents an operation process commonly used by a user, including multiple operation steps and the execution order between multiple operation steps.

[0065] S1013: Add the knowledge chain to the initial business operation event graph through knowledge alignment to generate a business operation event graph.

[0066] Among them, knowledge alignment is to merge the nodes with the same shape or even different shapes but the same meaning in the nodes of the initial business operation event graph and the nodes of the knowledge chain according to the relationship between the corresponding nodes.

[0067] See Figure 3 , which is a schematic diagram of the generation of a business operation event graph provided by an embodiment of this application. Figure 3 The left figure shows an initial business operation event graph. The knowledge chain generated through S1012 is operation step A - operation step B - operation step D. When the two are fused through knowledge alignment, the business operation event graph shown in the Figure 3 right figure is generated.

[0068] S102: Determine the number of times adjacent nodes appear in the business operation event graph according to the system operation logs generated by the user using the financial system, and use it as the weight of the node connection corresponding to the adjacent nodes.

[0069] Analyze the number of occurrences of the relationship between nodes in the business operation event graph from the system operation logs and record it on the edge corresponding to the nodes.

[0070] Continuing with Figure 2 as an example, if operation step A - operation step B appears 10 times in the system operation logs, the weight of the node connection corresponding to nodes A and B is 10. If operation step B - operation step C appears 5 times in the system operation logs, the weight of the node connection corresponding to nodes B and C is 5. If operation step C - operation step A appears 1 time in the system operation logs, the weight of the node connection corresponding to nodes B and C is 1. Thus, a weighted business operation event graph as shown in Figure 4 is generated.

[0071] S103: Mine the financial process from the business operation event graph based on the weights of the connections between the target node and the corresponding nodes.

[0072] Taking one node (the target node) among the multiple nodes included in the business operation event graph as an example, multiple paths including the target node can be obtained in the business operation event graph. Each path corresponds to a financial process. The weight corresponding to each path can be determined through the weights of the node connections. The greater the weight, the higher the credibility, thus realizing the financial process from the business operation event graph.

[0073] It can be obtained from the above solution that the business operation event graph is constructed according to the operation process of the financial system. According to the system operation logs generated by the user using the financial system, the number of occurrences of adjacent nodes in the business operation event graph is determined and used as the weight of the node connection corresponding to the adjacent nodes. Based on the weights of the connections between the target node and the corresponding nodes, the financial process is mined from the business operation event graph. Thus, there is no longer a need to rely on a large amount of expert knowledge and complex predefined event-condition-action rule information. Only based on the relationship between the operation steps in the operation process of the financial system, the financial process can be mined, reducing the development difficulty and cost.

[0074] As a possible implementation, establish a starting point set including the target node; taking the nodes included in the starting point set as starting points, determine a path set having a connection relationship with the nodes included in the starting point set. The path set includes the node connections from each node in the starting point set to other nodes and the corresponding weights, and the other nodes are the nodes in the business operation event graph except the nodes included in the starting point set; add the nodes corresponding to the node connections with the largest weights in the path set to the starting point set, and execute the step of determining the path set having a connection relationship with the nodes included in the starting point set taking the nodes included in the starting point set as starting points until no nodes are added to the starting point set; generate the financial process according to the starting point set.

[0075] The following combines Figure 5 An embodiment is used to illustrate a possible implementation of S103.

[0076] Using the business operation event graph, assume that the vertices in the event graph are divided into 2 sets, namely the starting point set U and the path set V - U. It can be understood that the starting point set U includes the target node. Based on the target starting point, the path set V - U includes the nodes having a connection relationship with the target node and the connection relationships. Each time, select the one with the largest weight corresponding to the node connection in the path set V - U, and add the node connected by this node connection to the set U until all the connectable nodes are added to the starting point set U, then a potential financial process can be obtained.

[0077] If the established event graph is as shown in Figure 5 (a) below, where the Chinese and English letters V (V1 - V6) in the business operation event graph represent nodes, and the edges between the nodes represent the sequential relationship between two nodes (the sequential relationship is ignored here), and the numbers on the edges represent the frequency of the sequential relationship appearing in the system operation log.

[0078] First, select any node V in the graph as the starting node of the financial process. After that, if node W is added to this financial process, there must be an edge between node V and node W, and the weight of this edge is the largest among the weights of the edges connecting node V to other nodes.

[0079] Taking point V2 as the starting node (target node), at this time the starting point set U = {V2}, and the paths included in the path set V - U and their corresponding weights are V2 - V1 = 1, V2 - V4 = 2, V2 - V3 = 7. Take the largest V2 - V3, so add point V3 to the starting point set U, that is, U = {V2, V3}, as shown in Figure 3 (b) below.

[0080] When U = {V2, V3}, the paths included in the path set V - U and their corresponding weights are V2 - V1 = 1, V2 - V4 = 2, V3 - V1 = 4, V3 - V4 = 4, V3 - V6 = 5. Take the largest V3 - V6, and add point V6 to the starting point set U, that is, U = {V2, V3, V6}, as shown in Figure 3 (c) below.

[0081] When U = {V2, V3, V6}, the paths included in the path set V - U and their corresponding weights are V2 - V1 = 1, V2 - V4 = 2, V3 - V1 = 4, V3 - V4 = 4, V6 - V4 = 5, V6 - V5 = 1. Take the largest V6 - V4, and add point V4 to the starting point set U, that is, U = {V2, V3, V6, V4}, as shown in Figure 3 (d) below.

[0082] When U = {V2, V3, V6, V4}, the paths included in the path set V - U and their corresponding weights are V2 - V1 = 1, V2 - V4 = 2, V3 - V1 = 4, V3 - V4 = 4, V6 - V5 = 1. Take the largest V3 - V1. It can be understood that if there are the same sizes, any one can be chosen, and add point V1 to the starting point set U, that is, U = {V2, V3, V6, V4, V1}, as shown in Figure 3 (e) below.

[0083] Select continuously in turn until all nodes are added, or in other words, no nodes can be added to the starting point set, then the potentially possible financial process is obtained, as shown inFigure 3 as shown in (f) of the middle

[0084] As a possible implementation, before S103, in order to avoid the influence of abnormal log data, abnormal nodes can be removed to make the subsequent mined financial process more accurate. The following is a specific description.

[0085] In the related art, the method for mining low-frequency behaviors of business processes based on Petri nets is inevitably interfered by noise logs or abnormal behavior logs. The mined process is often a local optimal solution rather than a global optimal solution. Based on this, the method for mining low-frequency behaviors of business processes based on Petri nets uses behavioral semantics to distinguish low-frequency behaviors and noise. However, this method only uses the semantics between two nodes and still inevitably has a low recognition rate. Based on this, the embodiments of the present application provide a method for removing abnormal nodes based on the analysis of multiple nodes, as specifically shown in S201-S203.

[0086] S201: Determine a hypothetical event chain according to the first m nodes and the last n nodes connected to the target node.

[0087] Among them, m and n are positive integers, which can be equal or unequal. The present application does not make specific limitations on this. Here, taking m = 3 and n = 3 as an example, it will be described in combination with the target node.

[0088] S202: Obtain the first weight sum of the node connections included in the hypothetical event chain, and the second weight of the target node connection connected to the target node in the hypothetical event chain.

[0089] The hypothetical event chain includes m + n + 1 nodes and m + n node connections. Each node connection has a weight. Therefore, the first weight sum can be obtained based on the weights corresponding to the m + n node connections.

[0090] It should be noted that in the hypothetical event chain, there are 3 node connections connected to the target node, namely, the node connection between the target node and the previous node, the node connection between the target node and the next node, and the connection formed by the aforementioned two node connections. The above 3 node connections can all be used as the target node connection. The present application does not make specific limitations on this. It can be understood that corresponding to different target node connections, the corresponding second weight sizes are different, and the subsequent corresponding preset conditions will also be different.

[0091] S203: If the first weight sum and the second weight do not meet the preset conditions, delete the target node connection to obtain an updated business operation event graph.

[0092] If the target node connection appears a relatively large number of times and the difference between the corresponding second weight and the sum of the first weights is not too much, it indicates that it is more likely that the target node connection is not generated by abnormal log data. If the target node connection appears a relatively small number of times and the difference between the corresponding second weight and the sum of the first weights is relatively large, it indicates that it is more likely that the target node connection is generated by abnormal log data.

[0093] Among them, the preset condition is set to determine that the target node connection appears a relatively large number of times. For example, if the second weight exceeds half of the sum of the first weights.

[0094] As a possible implementation, each node connection in the business operation event graph can be used as the target node connection for judgment to obtain an updated business operation event graph, and the accuracy of the obtained business operation event graph is higher, and the subsequent mined financial process is more accurate.

[0095] The following is illustrated by an example. In the business operation event graph, based on the execution relationship between operation steps, 3 nodes before the target node and 3 nodes after it are obtained. Together with the current node, a total of 7 nodes form a hypothetical event chain. Taking the node connection between the target node and the previous node as the target node connection, if the second weight of the target node connection does not exceed half of the sum of the first weights corresponding to the hypothetical event chain, it is considered that the sum of the first weights and the second weight do not meet the preset condition, and it is more likely that the target node comes from abnormal logs. Then, delete the target node connection to obtain an updated business operation event graph.

[0096] Therefore, compared with the method for mining low-frequency behaviors in business processes based on Petri nets that uses behavioral semantics to distinguish low-frequency behaviors and noises, the method provided in the embodiments of the present application can start from the overall hypothetical event chain, rather than relying solely on the semantic relationship between 2 event nodes to distinguish low-frequency and noise log data, which can improve the noise log recognition rate.

[0097] Furthermore, the deleted target node can delete the system operation log corresponding to the hypothetical event chain, match the financial process with the process included in the system operation log. If it is completely matched, the financial process mining is successful. If it is not completely matched, delete this financial process to further improve the accuracy of the financial process.

[0098] In addition to the financial process mining method based on event graphs provided in the embodiments of the present application, a financial process mining device based on event graphs is also provided, as Figure 6 shown. The device includes: a construction unit 601, a determination unit 602, and a mining unit 603;

[0099] The building unit 601 is used to construct a business operation event graph according to the operation process of the financial system; the business operation event graph includes nodes representing operation steps and node connections identifying the execution order between the operation steps.

[0100] The determining unit 602 is used to determine the number of occurrences of adjacent nodes in the business operation event graph according to the system operation logs generated by the user using the financial system, and use it as the weight of the node connection corresponding to the adjacent node.

[0101] The mining unit 603 is used to mine financial processes from the business operation event graph based on the target node and the weight of the corresponding node connection, where the target node is one of the multiple nodes included in the business operation event graph.

[0102] As a possible implementation, the device further includes an updating unit, which is used to:

[0103] Determine a hypothetical event chain according to the first m nodes and the last n nodes connected to the target node, where the target node is one of the multiple nodes included in the business operation event graph;

[0104] Obtain the sum of the first weights of the node connections included in the hypothetical event chain, and the second weight of the target node connection connected to the target node in the hypothetical event chain;

[0105] If the sum of the first weights does not meet the preset condition with the second weight, delete the target node connection to obtain an updated business operation event graph.

[0106] As a possible implementation, the device further includes a matching unit, which is used to:

[0107] Match the financial process with the processes included in the system operation logs;

[0108] If it is a complete match, the financial process mining is successful.

[0109] As a possible implementation, the building unit 601 is used to:

[0110] Construct an initial business operation event graph according to the system user manual of the financial system;

[0111] Extract knowledge chains according to the system operation logs generated by the user using the financial system. Each knowledge chain is used to identify an operation process, including multiple operation steps and the execution order between the operation steps;

[0112] Add the knowledge chain to the initial business operation event graph through knowledge alignment to generate the business operation event graph.

[0113] As a possible implementation, the mining unit 603 is configured to:

[0114] Establish a starting point set including target nodes;

[0115] Taking the nodes included in the starting point set as starting points, determine a path set having a connection relationship with the nodes included in the starting point set, where the path set includes node connections from each node in the starting point set to other nodes and corresponding weights, and the other nodes are nodes in the business operation event graph except for the nodes included in the starting point set;

[0116] Add the nodes corresponding to the node connections with the largest weights in the path set to the starting point set, and execute the step of taking the nodes included in the starting point set as starting points to determine a path set having a connection relationship with the nodes included in the starting point set until no nodes are added to the starting point set;

[0117] Generate a financial process according to the starting point set.

[0118] It can be seen from the above technical solutions that a business operation event graph is constructed according to the operation process of the financial system, the number of times adjacent nodes appear in the business operation event graph is determined according to the system operation logs generated by the user using the financial system and used as the weights of the node connections corresponding to the adjacent nodes, and based on the target nodes and the weights of the corresponding node connections, a financial process is mined from the business operation event graph. Thus, it is no longer necessary to rely on a large amount of expert knowledge and complex pre-defined event-condition-action rule information, and the financial process can be mined only according to the relationship between operation steps in the operation process of the financial system, reducing the development difficulty and cost.

[0119] The embodiment of the present application also provides a computer device. Refer to Figure 7 , which shows the structure diagram of a computer device provided by the embodiment of the present application. As Figure 7 shown, the device includes a processor 710 and a memory 720:

[0120] The memory 710 is used to store program codes and transmit the program codes to the processor;

[0121] The processor 720 is used to execute any one of the financial process mining methods based on the event graph provided in the above embodiment according to the instructions in the program codes.

[0122] The embodiment of the present application provides a computer-readable storage medium, and the computer-readable storage medium is used to store a computer program, and the computer program is used to execute any one of the financial process mining methods based on the event graph provided in the above embodiment.

[0123] The embodiments of the present application also provide a computer program product or a computer program. The computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the method for event-graph-based financial process mining provided in various alternative implementations of the above aspects.

[0124] As described above, it is only a preferred embodiment of the present invention and does not impose any form of limitation on the present invention. Although the present invention is disclosed by the preferred examples as above, it is not intended to limit the present invention. Any person skilled in the art can make many possible changes and modifications to the technical solution of the present invention by using the methods and technical contents disclosed above without departing from the scope of the technical solution of the present invention, or modify it into equivalent change and equivalent embodiments. Therefore, any simple modification, equivalent change and modification made to the above examples based on the essence of the technical solution of the present invention without departing from the content of the technical solution of the present invention still fall within the protection scope of the technical solution of the present invention.

[0125] It should be noted that the embodiments in this specification are described in a progressive manner, and the key points of each embodiment are the differences from other embodiments. The same or similar parts among the embodiments can be referred to each other. For the systems or devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple, and the relevant parts can be referred to the descriptions in the method part.

[0126] It should be understood that in the present application, "at least one (item)" means one or more, and "a plurality" means two or more. "And / or" is used to describe the association relationship of associated objects and indicates that three relationships may exist. For example, "A and / or B" may mean: only A exists, only B exists, and both A and B exist at the same time. Among them, A and B can be singular or plural. The character " / " generally means that the associated objects before and after are in an "or" relationship. "At least one (one) of the following" or its similar expression means any combination of these items, including any combination of single item (one) or plural items (ones). For example, at least one (one) of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a, b and c", where a, b, c can be single or multiple.

[0127] It should also be noted that, in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising said element.

[0128] The steps of the methods or algorithms described in connection with the embodiments disclosed herein may be implemented directly in hardware, in software modules executed by a processor, or in a combination thereof. The software modules may be placed in a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium well known in the art.

[0129] The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A financial process mining method based on an event graph, characterized in that, the method includes: Constructing a business operation event graph according to the operation process of the financial system; the business operation event graph includes nodes representing operation steps and node connections identifying the execution order between the operation steps; Determining the number of occurrences of adjacent nodes in the business operation event graph according to the system operation logs generated by the user using the financial system, and using it as the weight of the node connection corresponding to the adjacent nodes; Mining financial processes from the business operation event graph based on the target node and the weight of the corresponding node connection, where the target node is one of the multiple nodes included in the business operation event graph; Before mining the financial process from the business operation event graph, the method further includes: Determining a hypothetical event chain according to the first m nodes and the last n nodes connected to the target node, where the target node is one of the multiple nodes included in the business operation event graph; Obtaining the sum of the first weights of the node connections included in the hypothetical event chain, and the second weight of the target node connection connected to the target node in the hypothetical event chain; If the sum of the first weights does not meet the preset condition with the second weight, delete the target node connection to obtain an updated business operation event graph.

2. The method according to claim 1, characterized in that, the method further includes: Matching the financial process with the processes included in the system operation logs; If it is a complete match, the financial process mining is successful.

3. The method according to claim 1, characterized in that, Constructing the business operation event graph according to the operation process of the financial system includes: Constructing an initial business operation event graph according to the system user manual of the financial system; Extracting knowledge chains according to the system operation logs generated by the user using the financial system, and each knowledge chain is used to identify an operation process, including multiple operation steps and the execution order between the operation steps; Adding the knowledge chain to the initial business operation event graph through knowledge alignment to generate the business operation event graph.

4. The method according to any one of claims 1-3, characterized in that, Mining the financial process from the business operation event graph based on the target node and the weight of the corresponding node connection includes: Establishing a starting point set including the target node; Taking the nodes included in the starting point set as starting points, determining a path set having a connection relationship with the nodes included in the starting point set, the path set including the node connections from each node in the starting point set to other nodes and the corresponding weights, and the other nodes are the nodes in the business operation event graph except the nodes included in the starting point set; Adding the nodes corresponding to the node connections with the largest weights in the path set to the starting point set, and executing the step of taking the nodes included in the starting point set as starting points and determining the path set having a connection relationship with the nodes included in the starting point set until no nodes are added to the starting point set; Generating a financial process according to the starting point set.

5. A financial process mining device based on an event graph, characterized in that, the device includes: a construction unit, a determination unit, a mining unit, and an update unit; the construction unit is configured to construct a business operation event graph according to the operation process of the financial system; the business operation event graph includes nodes representing operation steps and node connections identifying the execution order between the operation steps; the determination unit is configured to determine the number of occurrences of adjacent nodes in the business operation event graph according to the system operation logs generated by the user using the financial system, and use it as the weight of the node connection corresponding to the adjacent nodes; the mining unit is configured to mine financial processes from the business operation event graph based on the target node and the weight of the corresponding node connection, where the target node is one of the multiple nodes included in the business operation event graph; the update unit is configured to determine a hypothetical event chain according to the first m nodes and the last n nodes connected to the target node before mining the financial process from the business operation event graph, where the target node is one of the multiple nodes included in the business operation event graph; obtain the sum of the first weights of the node connections included in the hypothetical event chain, and the second weight of the target node connection connected to the target node in the hypothetical event chain; if the sum of the first weights does not meet the preset condition with the second weight, delete the target node connection to obtain an updated business operation event graph.

6. A computer device, characterized in that, the device includes a processor and a memory: the memory is used to store program code and transmit the program code to the processor; the processor is configured to execute the method according to any one of claims 1-4 according to the instructions in the program code.

7. A computer-readable storage medium, characterized in that, the computer-readable storage medium is used to store a computer program, and the computer program is used to execute the method according to any one of claims 1-4.

8. A computer program product, characterized in that, it includes a computer program or instruction; when the computer program or instruction is executed by a processor, the method according to any one of claims 1-4 is executed.

Images