Multi-Perspective Process Mining Method and System for Supporting Role-Activity Relationship Recognition

Abstract

The present invention discloses a multi - perspective process mining method and system for supporting role - activity relationship recognition, including: 1) obtaining the event log of the business process; 2) role recognition in the business process event log; 3) clustering the activities included in the business process event log, and performing role - activity relationship recognition on roles and activities; 4) mining the business process model, and annotating the process model according to the recognized role - activity relationship; 5) visualizing the annotated business process model containing role - activity relationships in the form of a swimlane diagram and annotating roles in the model to obtain a multi - perspective process model for supporting role - activity relationship recognition. The present invention breaks through the limitation of traditional business process mining in modeling and mining business process models from the perspective of activities, and adds role information to the mining of business process models, thereby more conveniently describing business processes and intuitively describing the logical relationships between various activities in business processes.

Description

Technical Field

[0001] The present invention relates to the technical field of business process management, and in particular to a multi-perspective process mining method, system, storage medium and computing device that support the identification of role-activity relationships. Background Art

[0002] Process mining can extract useful information from the event log records commonly generated by modern information systems. This technology provides new means for process discovery, monitoring and improvement in various application fields. Among the existing business process model mining methods, they can be divided into two categories. One is the business process model mining method based on activities included in the event log, such as directly analyzing the dependencies such as sequence, conflict, and concurrency between activities from the activity execution traces, and then constructing the Alpha Miner of the business process model; the other is the business process model mining method based on roles in the event log, using subprocesses to describe the responsibilities of process roles, so as to highlight the interaction between roles. In the existing methods, by analyzing the resources in the business process and analyzing the business relationships between resources for social network analysis. However, in the models obtained by the current business process mining methods, there is an urgent need for a business process model mining method that supports the identification of role-activity relationships, which can not only mine the role information in the business process, but also more intuitively describe the logical relationships between activities in the business process to improve the readability and understandability of the business process. Summary of the Invention

[0003] The first object of the present invention is to overcome the deficiencies of the prior art and provide a multi-perspective process mining method that supports the identification of role-activity relationships, breaking the limitation of the traditional business process mining in modeling and mining the business process model from the perspective of activities, and solving the problem that the traditional business process mining method ignores the relationship between roles and other elements in the business process.

[0004] The second object of the present invention is to provide a multi-perspective process mining system that supports the identification of role-activity relationships.

[0005] The third object of the present invention is to provide a storage medium.

[0006] The fourth object of the present invention is to provide a computing device.

[0007] The first object of the present invention is achieved by the following technical solutions: A multi-perspective process mining method that supports the identification of role-activity relationships, including the following steps:

[0008] 1) Obtain basic data, that is, the event log of the business process;

[0009] 2) Identify roles in the business process event log;

[0010] 3) Cluster the activities included in the business process event log, and identify the role-activity relationships for the roles and activities obtained in step 2).

[0011] 4) Mine the business process model, and annotate the process model based on the identified role-activity relationships.

[0012] 5) Visualize the annotated business process model containing role-activity relationships in the form of a swimlane diagram and by annotating roles in the model, to obtain a multi-perspective process model supporting role-activity relationship identification.

[0013] Furthermore, in step 1), the event log of the business process refers to the activity execution information recorded in the information management system of the business process. Each activity execution information is called an instance, and each instance contains multiple attributes, where the event ID, activity name, activity execution resource, and timestamp are basic attributes.

[0014] Furthermore, in step 2), identify the roles included in the business process event log, and use the distance between resources executing activities in the business process event log to determine whether two roles are the same role. Specifically, it includes the following steps:

[0015] 2.1) Taking the event log of the business process as input, traverse the activities of all event log traces to obtain the activity set A and the resource set R, where A = {a1, a2, a3,...., a n}, a j is each activity variant in the event log, j = 1, 2, 3,..., n, |A| = n is all activity variants included in this event log; R = {r1, r2, r3,...., r m}, r i is each resource executing activities in the event log, i = 1, 2, 3,..., m, |R| = m is all resource variants included in this event log;

[0016] 2.2) Traverse the event log, and count the number of times each resource executes each activity to obtain the resource-activity relationship count matrix RAD num = [(V r1 , V r2 , …, V rm ) T , where V ri is the vector of the number of times resource r i executes all activities;

[0017] 2.3) Calculate the frequency of resource r i executing activity a j according to formula (1) to obtain the resource-activity relationship frequency matrix RAD p = [(Pr1 , P r2 , …, P rm ) T , where P ri is the frequency vector for each resource r i to execute all activities;

[0018]

[0019] where P(r i , a j ) represents the frequency of resource r i to execute activity a j ; Num(r i , a j ) represents the number of times resource r i executes activity a j ; |V ri | represents the number of times resource r i executes all activities;

[0020] 2.4) Given a threshold TH, calculate the Euclidean distance for each resource in the resource-activity relationship frequency matrix to execute all activity vectors. If it is less than the threshold, classify the resources into the same role and obtain the final number of roles.

[0021] Furthermore, in step 3), cluster the activities included in the business process event log and identify the role-activity relationships for the roles and activities obtained in step 2). The specific steps are as follows:

[0022] 3.1) Denote the number of roles obtained in step 2) as Num, and cluster the activities in the event log according to the resources executing the activities, clustering them into Num activity classes, i.e., A cluster = {A1, A2, …, A Num}, where A cluster represents the set of all activity classes; A z represents each activity class, z = 1, 2, 3, …, Num;

[0023] 3.2) Denote the resources included in the roles obtained in step 2) as Role = {R1, R2, …, R Num |R z = {r1, r2,..., r p}}, where Role represents the set of all roles; R z represents each role, r x represents the resources classified into one role, 1 ≤ x ≤ p; Calculate, for each activity class A z ∈ A cluster in, the role R zThe ratio of the number of activities performed by all resources in ∈Role to the total number of activities in activity class A z If it is higher than a predefined threshold, then this role is associated with this activity class, that is, all activities in activity class A z belong to role R z .

[0024] Furthermore, in step 4), a process mining algorithm is used to mine the business process model, and the process model is annotated based on the identified role-activity relationships. Specifically, it includes the following steps:

[0025] 4.1) Use a process mining algorithm to mine the business process model of the event log to obtain the relationships between activities in the event log;

[0026] 4.2) Annotate the role-activity relationships identified in step 3) in the business process model mined in step 4.1) to obtain a business process model with role annotations.

[0027] Furthermore, in step 5), the annotated business process model containing role-activity relationships is visualized in the form of a swimlane diagram and role annotations in the model to obtain a multi-perspective process model supporting role-activity relationship identification. Specifically, it includes the following steps:

[0028] 5.1) Visualize and output the business process model with role annotations obtained in step 4) as a perspective process model;

[0029] 5.2) Rearrange the business process model with role annotations, store the activities belonging to the same role in the business process, and record the relationships between activities in the business process model;

[0030] 5.3) Use each role as the basis for dividing the swimlane diagram. The activities included in each role belong to the pools of the swimlane diagrams corresponding to each role. At the same time, visualize all activities and role information in the form of a swimlane diagram according to the relationships between activities recorded in step 5.2) to obtain a multi-perspective process model supporting role-activity relationship identification, that is, the swimlane diagram perspective model of the business process;

[0031] Among them, the swimlane diagram represents the activities of different roles in separate different areas. Each area is called a pool, which can conveniently describe the business process and intuitively describe the logical relationships between the activities in the business process.

[0032] The second object of the present invention is achieved through the following technical solutions: A multi-perspective process mining system supporting role-activity relationship identification is used to implement the above multi-perspective process mining method supporting role-activity relationship identification, and it includes:

[0033] A data acquisition module for acquiring the event logs of a business process; the event logs of the business process refer to the activity execution information recorded in the information management system in the business process, and each activity execution information is called an instance, and each instance contains multiple attributes. Among them, the activity ID, activity name, activity execution resource, and timestamp are basic attributes;

[0034] A role recognition module for identifying all roles from the event logs;

[0035] A role-activity relationship recognition module for identifying the roles and activities in the event logs;

[0036] A visualization module for visualizing the multi-perspective process model that supports the role-activity relationship recognition in combination with the identified role-activity relationship.

[0037] The third object of the present invention is achieved by the following technical solution: a storage medium stores a program, and when the program is executed by a processor, the multi-perspective process mining method for supporting the role-activity relationship recognition as described above is implemented.

[0038] The fourth object of the present invention is achieved by the following technical solution: a computing device includes a processor and a memory for storing a program executable by the processor. When the processor executes the program stored in the memory, the multi-perspective process mining method for supporting the role-activity relationship recognition as described above is implemented.

[0039] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0040] 1. For the first time, from the dimension of roles and in combination with the process mining method, the present invention breaks through the limitations of the traditional process mining method that mostly models and mines the business process based on activities.

[0041] 2. For the first time, the present invention visualizes the process model that supports the identification of the role-activity relationship in the form of a swimlane diagram, which is more conducive to the readability and understandability of the business process and helps with the subsequent optimization and improvement of the business process.

[0042] 3. For the first time, the present invention proposes a visualization method for the multi-perspective role-activity relationship process model. The multi-perspective process model provides two options for users, making it more convenient for users to use.

[0043] 4. The present invention has a wide range of application spaces in the mining of business process models and has broad prospects in the mining of business process models that support role-activity recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a schematic logical flow diagram of the method of the present invention.

[0045] Figure 2 is a business process model containing role annotations.

[0046] Figure 3 is a business process swimlane diagram perspective model that supports the identification of role-activity relationships.

[0047] Figure 4 is the system architecture diagram of the present invention. Detailed implementation manners

[0048] The present invention will be further described below in conjunction with specific embodiments.

[0049] As Figure 1 shown, this embodiment discloses a multi-perspective process mining method that supports the identification of role-activity relationships. First, the role-activity relationships are identified from the event log, then the interaction activities between roles are identified, the role-activity relationships are added to the process model, and finally, the multi-perspective process model that supports the identification of role-activity relationships is visualized. It includes the following steps:

[0050] 1) Obtain the basic data, which is the event log of the business process. The event log of the business process refers to the activity execution information recorded in the information management system in the business process. Each activity execution information is called an instance, and each instance contains multiple attributes. Among them, event ID, activity name (# act ), resource executing the activity (# Resource ), and timestamp (# time ) are basic attributes.

[0051] The traces contained in the event log of this business process are: <a 谢文 ,b 李冰 ,d 林帆 ,e 刘书 ,h 谢文 >,<a 林帆 ,d 林帆 ,c 谢文 ,e 刘书 ,g 张吉 >,<a 谢文 ,c 林帆 ,d 张吉 ,e 刘书 ,f 刘书 ,b 吕安 ,d 谢文 ,e 刘书 ,g 张吉 >,<a 谢文 ,d 林帆 ,b 吕安 ,e 刘书 ,h 张吉 >,<a 张吉 ,c 林帆 ,d 谢文, e 刘书 , f 刘书 , d 张吉 , c 林帆 , e 刘书 , f 刘书 , b 李冰 , d 谢文 , e 刘书 , h 林帆 >, <a 林帆 , c 张吉 , d 林帆 , e 刘书 , g 林帆 >, <a 谢文 , b 李冰 , d 林帆 , e 刘书 , g 谢文 >, among which the activity attributes of the business process event log of one case are shown in Table 1:

[0052] Table 1 Activity attributes of the business process event log of one case

[0053]

[0054]

[0055] The meanings of each activity in Table 1 are as follows: a: Register requirements; b: Comprehensive inspection; c: Random inspection; d: Ticket inspection; e: Make a decision; f: Reapply; g: Pay compensation; h: Reject request.

[0056] 2) Role identification in the business process event log, identify the roles included in the business process event log, and use the distance between resources executing activities in the business process event log to determine whether two roles are the same role, specifically including the following steps:

[0057] 2.1) Take the event log of the business process as input, traverse all the activities of the event log traces, and obtain the activity set A and the resource set R, where A = {a1, a2, a3,..., a n}, a j is each activity variant in the event log, j = 1, 2, 3,..., n, |A| = n is all the activity variants included in this event log; R = {r1, r2, r3,...., r m}, r i is each resource that executes activities in the event log, i = 1, 2, 3,..., m, |R| = m is all the resource variants included in this event log;

[0058] 2.2) Traverse the event log, count the number of times each resource executes each activity, and obtain the resource-activity relationship frequency matrix RAD num= [(V r1 , V r2 , …, V rm ) T , where V ri is the vector of the number of times each resource r i performs all activities;

[0059] 2.3) Calculate the frequency of resource r i performing activity a j according to formula (1) to obtain the resource-activity relationship frequency matrix RAD p = [(P r1 , P r2 , …, P rm ) T , where P ri is the frequency vector of each resource r i performing all activities;

[0060]

[0061] Among them, P(r i , a j ) represents the frequency of resource r i performing activity a j ; Num(r i , a j ) represents the number of times resource r i performs activity a j ; |V ri | represents the number of times resource r i performs all activities;

[0062] 2.4) Given a threshold TH, calculate the Euclidean distance of each resource performing all activity vectors in the resource-activity relationship frequency matrix. If it is less than the threshold, classify the resources into the same role and obtain the final number of roles.

[0063] Using the above steps, the activity set A = {a, b, c, d, e, f, g, h} and resource set R = {Xie Wen, Lin Fan, Zhang Ji, Liu Shu, Lv An, Li Bing} of this event log are mined; the obtained resource-activity relationship times table is shown in Table 2, and the resource-activity relationship times matrix is as shown in RAD num . Then, according to formula (1), the resource-activity relationship frequency matrix is as shown in RAD p , and we get:

[0064] Table 1 Resource-Activity Relationship Times Table

[0065] a b c d e f g h Xie Wen 4 0 1 3 0 0 1 1 Lin Fan 2 0 3 5 0 0 1 1 Zhang Ji 1 0 1 2 0 0 2 1 Li Bing 0 3 0 0 0 0 0 0 Lv An 0 2 0 0 0 0 0 0 Liu Shu 0 0 0 0 10 3 0 0

[0066]

[0067] Calculate the Euclidean distance between roles as follows:

[0068]

[0069]

[0070]

[0071]

[0072]

[0073]

[0074]

[0075]

[0076]

[0077]

[0078]

[0079]

[0080]

[0081]

[0082]

[0083] Determine the threshold TH = 0.4. Through the calculated Euclidean distance between roles, Xie Wen, Lin Fan, and Zhang Ji can be determined as one role, called worker; Li Bing and Lu An can be determined as one role, called expert; Liu Shu can be determined as one role, called manager, that is, Role = {worker, expert, manager}.

[0084] 3) Cluster the activities included in the business process event log, and identify the role-activity relationships between the roles and activities obtained in step 2). The specific steps are as follows:

[0085] 3.1) Call the number of roles obtained in step 2) as Num, and cluster the activities in the event log according to the resources executing the activities into Num activity classes, that is, A cluster = {A1, A2,..., A Num}, where A cluster represents the set of all activity classes; A zRepresenting each activity class, z = 1, 2, 3, …, Num;

[0086] 3.2) Represent the resources included in the role obtained in step 2) as Role = {R1, R2, …, R Num |R z = {r1, r2,..., r p}}, where Role represents the set of all roles; R z represents each role, and r x represents the resources classified into one role, 1 ≤ x ≤ p; Calculate, for each activity class A z ∈ A cluster , the ratio of the number of activities executed by the resources included in the role R z ∈ Role to the total number of activities in the activity class A z . If it is higher than the established threshold, this role can be corresponding to this activity class, that is, all activities in the activity class A z belong to the role R z .

[0087] Adopt the above steps: Num = 3, cluster the activities in the event log into 3 classes, and obtain 3 activity sets A cluster1 = {a, c, d, g, h}, A cluster2 = {b}, A cluster3 = {e, f}.

[0088] Determine the threshold as 0.8, and calculate the proportion of activities executed by each role: P(A cluster1 , worker) = 1, P(A cluster1 , expert) = P(A cluster1 , manager) = 0; P(A cluster2 , worker) = P(A cluster2 , manager) = 0, P(A cluster2 , expert) = 1; P(A cluster3 , worker) = P(A cluster3 , Exper) = 0, P(A cluster3 , manager) = 0; Therefore, the role-activity relationships can be identified: <worker, A cluster1 >, <Exper, A cluster2 >, <manager, A cluster3 >.

[0089] 4) Use the process mining algorithm to mine the business process model and annotate the process model according to the identified role-activity relationships, which specifically includes the following steps:

[0090] 4.1) Use the process mining algorithm to mine the business process model of the event log to obtain the relationships between activities in the event log;

[0091] 4.2) Annotate the role-activity relationships identified in step 3) in the business process model mined in step 4.1) to obtain a business process model with role annotations.

[0092] 5) Visualize the annotated business process model containing role-activity relationships in the form of a swimlane diagram and role annotations in the model to obtain a multi-perspective process model supporting role-activity relationship identification, which specifically includes the following steps:

[0093] 5.1) Visualize and output the business process model with role annotations obtained in step 4) as a perspective process model.

[0094] 5.2) Rearrange the business process model with role annotations, store the activities belonging to the same role in the business process, and record the relationships between activities in the business process model.

[0095] 5.3) Use each role as the basis for dividing the swimlane diagram. The activities included in each role belong to the pools of the swimlane diagrams corresponding to each role. At the same time, visualize all activities and role information in the form of a swimlane diagram according to the relationships between activities recorded in step 5.2) to obtain a multi-perspective process model supporting role-activity relationship identification, that is, the business process swimlane diagram perspective model.

[0096] Among them, the swimlane diagram represents the activities of different roles in separate different areas, and each area is called a pool, which can conveniently describe the business process and intuitively describe the logical relationships between the activities in the business process.

[0097] Using the above steps, taking the business process event log where Table 1 is located as an example, a multi-perspective process model supporting role-activity relationship identification is finally obtained, that is, Figure 2 the business process model with role annotations shown as follows and Figure 3 the business process swimlane diagram perspective model supporting role-activity relationship identification shown as follows.

[0098] Embodiment 2

[0099] This embodiment discloses a multi-perspective process mining system supporting role-activity relationship identification, which is used to implement the multi-perspective process mining method supporting role-activity relationship identification described in Embodiment 1. As Figure 4 shown in the figure, this system includes the following functional modules:

[0100] A data acquisition module for acquiring the event logs of a business process; the event logs of the business process refer to the activity execution information recorded in the information management system in the business process, and each activity execution information is called an instance, and each instance contains multiple attributes. Among them, the activity ID, activity name, activity execution resource, and timestamp are basic attributes;

[0101] A role recognition module for identifying all roles from the event logs;

[0102] The role-activity relationship recognition module for identifying roles and activities from the event logs;

[0103] The visualization module for visualizing a multi-perspective process model that supports role-activity relationship recognition in combination with the identified role-activity relationship.

[0104] Embodiment 3

[0105] This embodiment discloses a storage medium storing a program, which when executed by a processor, implements the multi-perspective process mining method for supporting role-activity relationship recognition described in Embodiment 1.

[0106] The storage medium in this embodiment may be a magnetic disk, an optical disc, a computer memory, a read-only memory (ROM), a random access memory (RAM), a USB flash drive, a mobile hard disk, or other media.

[0107] Embodiment 4

[0108] This embodiment discloses a computing device including a processor and a memory for storing a program executable by the processor. When the processor executes the program stored in the memory, it implements the multi-perspective process mining method for supporting role-activity relationship recognition described in Embodiment 1.

[0109] The computing device described in this embodiment may be a desktop computer, a laptop computer, a smart phone, a PDA handheld terminal, a tablet computer, a programmable logic controller (PLC), or other terminal devices with processor functions.

[0110] In summary, after adopting the above solutions, the present invention provides a new method and system for model mining of business process mining. By identifying the role-activity relationship of business process event logs, a multi-perspective process model that supports the role-activity relationship is obtained, which can clearly describe the relationship of the business process, effectively improve the readability and understandability of the business process, has practical promotion value, and is worthy of promotion.

[0111] The above-described embodiments are only preferred embodiments of the present invention, and do not limit the scope of implementation of the present invention. Therefore, any changes made according to the shape and principle of the present invention should be covered within the protection scope of the present invention.

Claims

1. A multi-perspective process mining method for supporting role-activity relationship recognition, characterized in that Including the following steps: 1) Obtain the basic data, which is the event log of the business process; 2) Identify the roles included in the business process event log, and use the distance between the resources executing activities in the business process event log to determine whether two roles are the same role. Specifically, it includes the following steps: 2.1) Taking the event log of the business process as input, traverse the activities of all event log traces to obtain the activity set A and the resource set R, where A = {a1, a2, a3, …, a n}, a j is each activity variant in the event log, j = 1, 2, 3, …, n, |A| = n is all the activity variants included in this event log; R = {r1, r2, r3, …, r m}, r i is each resource that executes the activity in the event log, i = 1, 2, 3, …, m, |R| = m is all the resource variants included in this event log; 2.2) Traverse the event log, count the number of times each resource executes each activity, and obtain the resource-activity relationship count matrix RAD num = [(V r1 , V r2 , …, V rm ) T , where V ri is the count vector of the number of times resource r i executes all activities; 2.3) Calculate the resource r according to formula (1) i Execute the activity a j frequency to obtain the resource-activity relationship frequency matrix RAD p = [(P r1 , P r2 , …, P rm ) T , where P ri is the frequency vector of each resource r i executing all activities; where, P(r i , a j ) represents the frequency of resource r i executing activity a j ; Num(r i , a j ) represents the number of times resource r i executes activity a j ; |V ri | represents the number of times resource r i executes all activities; 2.4) Given a threshold TH, calculate the Euclidean distance of each resource executing all activity vectors in the resource-activity relationship frequency matrix. If it is less than the threshold, classify the resources as the same role and obtain the final number of roles; 3) Cluster the activities included in the business process event log, and identify the role-activity relationship between the roles and activities obtained in step 2); 4) Mine the business process model and annotate the process model according to the identified role-activity relationship; 5) Visualize the annotated business process model including the role-activity relationship in the form of a swimlane diagram and annotating roles in the model to obtain a multi-perspective process model supporting role-activity relationship identification.

2. The multi-perspective process mining method for supporting role-activity relationship recognition according to claim 1, wherein: In step 1), the event log of the business process refers to the activity execution information recorded in the information management system in the business process. Each activity execution information is called an instance, and each instance contains multiple attributes, where the event ID, activity name, resource executing the activity, and timestamp are basic attributes.

3. The multi - perspective process mining method for supporting role - activity relationship recognition according to claim 2, characterized in that: In step 3), cluster the activities included in the business process event log, and identify the role-activity relationship between the roles and activities obtained in step 2). The specific steps are as follows: 3.1) Denote the number of roles obtained in step 2) as Num, and cluster the activities in the event log according to the resources executing the activities into Num activity classes, namely A cluster = {A1, A2, …, A Num}, where A cluster represents the set of all activity classes; A z represents each activity class, z = 1, 2, 3, …, Num; 3.2) Represent the resources included in the role obtained in step 2) as Role = {R1, R2, …, R Num |R z = {r1, r2,..., r p}}, where Role represents the set of all roles; R z represents each role, and r x represents the resources classified into one role, 1 ≤ x ≤ p; Calculate the ratio of the number of activities executed by all the resources included in the role R z ∈ A cluster in each activity class A z to the total number of activities in the activity class A z . If it is higher than the established threshold, then correspond this role to this activity class, that is, all the activities in the activity class A z belong to the role R z .

4. The multi-perspective process mining method for supporting role-activity relationship recognition according to claim 3, wherein: In step 4), use the process mining algorithm to mine the business process model and annotate the process model according to the identified role-activity relationship. Specifically, it includes the following steps: 4.1) Use the process mining algorithm to mine the business process model of the event log to obtain the relationship between activities in the event log; 4.2) Annotate the role-activity relationship identified in step 3) in the business process model mined in step 4.1) to obtain a business process model with role annotations.

5. The multi-perspective process mining method for supporting role-activity relationship recognition according to claim 4, wherein: In step 5), visualize the annotated business process model including the role-activity relationship in the form of a swimlane diagram and annotating roles in the model to obtain a multi-perspective process model supporting role-activity relationship identification. Specifically, it includes the following steps: 5.1) Visualize and output the business process model with role annotations obtained in step 4) as a perspective process model; 5.2) Rearrange the business process model with role annotations, store the activities belonging to the same role in the business process, and record the relationship between activities in the business process model at the same time; 5.3) Use each role as the basis for dividing the swimlane diagram. The activities included in each role belong to the pools of the swimlane diagrams corresponding to each role. At the same time, visualize all activities and role information in the form of a swimlane diagram according to the relationship between activities recorded in step 5.2) to obtain a multi-perspective process model supporting role-activity relationship identification, that is, the business process swimlane diagram perspective model. Among them, the swimlane diagram represents the activities of different roles in separate different areas, and each area is called a pool, which can conveniently describe the business process and intuitively describe the logical relationship between the activities in the business process.

6. A multi-perspective process mining system for supporting role-activity relationship recognition, characterized in that, A multi-perspective process mining method for supporting role-activity relationship recognition according to any one of claims 1 to 5, comprising: A data acquisition module for acquiring the event log of the business process; the event log of the business process refers to the activity execution information recorded in the information management system in the business process, and each activity execution information is called an instance, and each instance contains multiple attributes. Among them, activity ID, activity name, activity execution resource, and timestamp are basic attributes; A role recognition module for identifying all roles from the event log; A role-activity relationship recognition module for identifying the roles and activities in the event log; A visualization module for visualizing the multi-perspective process model that supports role-activity relationship recognition in combination with the identified role-activity relationship.

7. A storage medium stores a program, characterized in that, When the program is executed by the processor, the multi-perspective process mining method for supporting role-activity relationship recognition according to any one of claims 1 to 5 is implemented.

8. A computing device, comprising a processor and a memory for storing processor-executable programs, characterized in that, When the processor executes the program stored in the memory, the multi-perspective process mining method for supporting role-activity relationship recognition according to any one of claims 1 to 5 is implemented.

Images