CN114201460A - Block structure process mining method for incomplete event log caused by concurrency - Google Patents

Block structure process mining method for incomplete event log caused by concurrency Download PDF

Info

Publication number
CN114201460A
CN114201460A CN202111529372.XA CN202111529372A CN114201460A CN 114201460 A CN114201460 A CN 114201460A CN 202111529372 A CN202111529372 A CN 202111529372A CN 114201460 A CN114201460 A CN 114201460A
Authority
CN
China
Prior art keywords
event
logs
sub
log
cutting
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111529372.XA
Other languages
Chinese (zh)
Inventor
瞿鹏
杨帅豪
谭泽亚
肖芳雄
宋巍
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University of Science and Technology
Original Assignee
Nanjing University of Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanjing University of Science and Technology filed Critical Nanjing University of Science and Technology
Priority to CN202111529372.XA priority Critical patent/CN114201460A/en
Publication of CN114201460A publication Critical patent/CN114201460A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/10—File systems; File servers
    • G06F16/17—Details of further file system functions
    • G06F16/1734—Details of monitoring file system events, e.g. by the use of hooks, filter drivers, logs
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2465—Query processing support for facilitating data mining operations in structured databases
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/31—Indexing; Data structures therefor; Storage structures
    • G06F16/316—Indexing structures
    • G06F16/322—Trees

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Computational Linguistics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Mathematical Physics (AREA)
  • Fuzzy Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a block structure process mining method aiming at incomplete event logs caused by concurrency, which takes event logs described in an XES format as input and a process model file described in a PTML format as an output result; firstly, splitting an incomplete event log by using a method similar to an Induce Miner to determine the nesting relation between different block structures and the incomplete event log, and further processing the incomplete event log which cannot be correctly split and is caused by concurrence; secondly, deducing a possibly lost transmission previous relation by using collaborative filtering on the sub-logs, and performing segmentation operation by using segmentation operation applied to a transmission previous graph; and finally, combining all the block structures into a block structure process model represented by a process tree according to the nesting relation. The invention can process the incomplete event log caused by concurrency and excavate the process model close to the reality as much as possible.

Description

Block structure process mining method for incomplete event log caused by concurrency
Technical Field
The invention belongs to the field of business processes, and particularly relates to a block structure process mining method for incomplete event logs caused by concurrency.
Background
In the era of big data and digitalization, many enterprises deploy information systems to better manage enterprise resources and improve operation efficiency. The information system of the enterprise collects and records a large amount of event data in daily operation, such as financial audit logs of the enterprise, supply chain purchase event data and the like. Because business requirements of an enterprise may change, business processes may also change, and a manager needs a current real business process of the enterprise to improve the current business process. And the event log of the enterprise in the recent period can reflect the current real business process of the enterprise. The process mining technology can mine a business process model from the event logs, and an enterprise manager can find a bottleneck and improve a business process by analyzing the process model, so that the purposes of reducing the operation cost and improving the operation efficiency are achieved.
The event log contains a number of sequences of events including information such as the name of the event, the time of occurrence, the operator, the action, etc. At present, the mainstream process mining algorithm depends on high-quality event logs, for example, the event logs are required to meet the defined completeness, and a process model with higher mining quality can be guaranteed. However, a real business process often includes concurrent activities, and event logs including the concurrent activities often hardly satisfy completeness required by a process mining algorithm, so that the quality of a mined process model is poor.
Disclosure of Invention
The invention aims to provide a mining method for a block structure process aiming at an incomplete event log caused by concurrency.
The technical solution for realizing the purpose of the invention is as follows: a block structure process mining method aiming at incomplete event logs caused by concurrency is characterized in that a block structure process model is mined according to event log information; taking an event log in an XES format as input, taking a process model in a PTML format as output, wherein the process model is a process tree which is a process model with a block structure, leaf nodes are events in the event log, non-leaf nodes are one of selection, sequence, circulation and concurrency, and each sub-tree of the process tree is a block structure; the method comprises the following specific steps:
step 1, for an event log, segmenting events contained in the event log by utilizing a segmentation operation based on a direct previous relation in an Inductive Miner, wherein the result of the segmentation operation is a plurality of event sets, and splitting corresponding sub-logs from an original event log according to projections according to the event sets; at this time, a preliminary process tree can be obtained, the type of the splitting operation corresponds to the root node, and the split child logs are child nodes of the split child logs; continuously splitting the sub-logs by repeating the above operations until the sub-logs only contain one event; for the sub-event logs which cannot be normally split, judging whether repeated and mutually exclusive events exist in each event sequence, if so, returning a block structure flower model capable of generating any event sequence to the event log, otherwise, entering the step 2;
step 2, traversing the event sequence of the child logs which are not correctly processed in the step 1 to acquire the transfer preceding relation among the events and generating a corresponding 0-1 matrix, wherein an element 1 in the matrix represents that the transfer preceding relation exists, a 0 represents that the transfer preceding relation does not exist, and the rest represents uncertainty; using collaborative filtering to deduce uncertain elements in the 0-1 matrix to determine whether the uncertain elements have a transfer preceding relationship; then generating a transfer previous graph according to the inference result, carrying out cutting operation by using selection cutting, sequential cutting or concurrent cutting applied to the transfer previous graph, and splitting the event log according to the splitting result; if the sub-logs can be normally split, the block structure corresponding to the sub-logs can be obtained, otherwise, a flower model corresponding to the event log is returned;
and 3, combining all the block structures into a process model of the block structure represented by a process tree according to the nesting relation.
An electronic device comprises a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the above-mentioned block structure process mining method for the incomplete event log caused by concurrency when executing the program.
A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, implements the above-described method for block-structured process mining for concurrency-induced incomplete event logs.
Compared with the Induced Miner and a derivative algorithm thereof and a mainstream process mining algorithm Alpha algorithm, the method can better process incomplete event logs caused by concurrency, mine the process model with higher accuracy, and the found block structure process model is easier to analyze and understand.
Drawings
FIG. 1 is a flow chart of a block structure process mining method for a concurrency-induced incomplete event log according to the invention.
FIG. 2 is a schematic diagram of a process model shown in process tree form.
Fig. 3 is a schematic diagram of an event log file described in an XES format.
Fig. 4 is a schematic diagram of an event sequence set S obtained by parsing an event log file in an XES format.
FIG. 5 is a schematic diagram of a delivery preceding matrix extracted from an event log.
Fig. 6 is a schematic diagram of the process of applying concurrent cuts on the previous figures.
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention discloses a block structure process mining method aiming at incomplete event logs caused by concurrency, wherein the event logs are a set of a plurality of event sequences, events comprise information such as activity names, time stamps and resources, and if only the event names and the time stamps are considered, the event sequences can be simplified into a series of event name sequences with a sequence occurrence order. The invention takes an event log described by an XES format as input and a process model described by a PTML format as output, the specific flow is shown as the attached figure 1, and the specific steps are as follows:
step 1, for an event log, segmenting events contained in the event log by utilizing segmentation operations (selection segmentation, sequential segmentation, circular segmentation and concurrent segmentation) based on direct previous relations in an Inductive Miner, and splitting the event log according to a segmentation result until only one event is contained in a sub-log, wherein the specific steps are as follows:
step 1-1, splitting the incomplete event log by using a splitting operation based on a direct previous relation defined in an Inductive Miner to determine a nesting relation between different block structures and the incomplete event log until the sub-log only contains one event (namely the splitting cannot be continued), and further processing the sub-log which cannot be normally split;
step 1-2, for the sub-logs which cannot be processed normally in step 1-1, judging whether each event sequence contained in the sub-logs has repeated events or mutually exclusive events (namely whether the event sequence corresponds to a cycle or a selection structure), if so, returning a flower model with a block structure to the event log, otherwise, further processing is needed;
step 2, traversing the event sequence of the child logs which are not correctly processed in the step 1 to acquire the transfer preceding relation among the events and generating a corresponding 0-1 matrix, wherein an element 1 in the matrix represents that the transfer preceding relation exists, a 0 represents that the transfer preceding relation does not exist, and the rest represents uncertainty; predicting uncertain elements in the 0-1 matrix by using collaborative filtering to determine whether a transfer preceding relation exists; then generating a previous graph transmitted according to the prediction result, performing segmentation operation by using selection segmentation, sequential segmentation or concurrent segmentation applied to the previous graph transmitted, and splitting the event log according to the segmentation result; if the split can be normally carried out, the block structure corresponding to the sub-log can be obtained, otherwise, a flower model corresponding to the event log is returned, and the specific steps are as follows:
and 2-1, traversing all event sequences for the logs which cannot be processed in the step 1, and acquiring all event pairs with a transmission previous relation. The method specifically comprises the following steps: any one event sequence a1,a2,a3,……,an-1,anContains n events, wherein aiRepresenting an event i, arranging the N events according to the occurrence sequence, and obtaining N x (N-1)/2 groups of delivery preceding relation pairs: a is1＞a2,a1＞a3,…,a1＞an,a2＞a3,a2＞a4,…,a2＞an,…,an-1＞an. Wherein, ai＞ajRepresenting event aiThe transfer occurs at event ajBefore. Generating a pre-transmission 0-1 matrix according to the pre-transmission relation among the events, wherein the first row and the first column in the matrix are event names, if the rest elements are 1, the pre-transmission relation exists between the event pairs corresponding to the elements, if the rest elements are 0, the pre-transmission relation does not exist, and the rest elements do not determine whether the pre-transmission relation exists or not;
step 2-2, for the event pairs corresponding to the uncertain elements in the 0-1 matrix<ai,aj>Record aiIs located in the row vector [ a1,a2,……,ai-1,ai+1,……,aj-1,aj+1,……,an-1,an]Is v isi，ajIs located in the column vector [ a1,a2,……,ai-1,ai+1,……,aj-1,aj+1,……,an-1,an]TIs v isjWherein n is the number of events in the matrix, and uncertain elements are regarded as 0 in the calculation process; according to the formula
Figure BDA0003411164830000041
Where A and B represent vectors, calculate viCosine similarity with other line vectors, except for ajFront of highest similarity outside the line vector
Figure BDA0003411164830000042
The event of each is recorded as a set Su(ii) a Calculating vjCosine similarity with other column vectors, except for aiTop of highest similarity outside the column vector
Figure BDA0003411164830000046
The event of each is recorded as a set Si；
Step 2-3, calculating S of all event pairs which are uncertain whether the transfer previous relation exists or not according to the step 2-2uAnd SiSet, for each undetermined element M in the matrixx,yThe inference result is Ratingx,yAccording to the formula
Figure BDA0003411164830000043
Is calculated, where M isr,cRepresenting an event a in a matrixrAnd event acValue of the element, Mr,yAnd Mx,cIn the same way, the method for preparing the composite material,
Figure BDA0003411164830000044
r denotes a symbol from the set SuEvent a ofrC denotes a symbol from the set SiEvent a ofcλ and δ are linear interpolation parameters, fixed as 0.5 by default; arranging all the inference results in ascending order, and taking the total number of the inference results as m
Figure BDA0003411164830000045
The number is used as a threshold value, the inferred value is larger than the threshold value or reaches 1, the corresponding event pair is judged to have the transmission prior relationship, otherwise, the corresponding event pair does not exist, the transmission prior graph TPG can be generated by combining the previously determined event pair with the transmission prior relationship, the vertex is an event, and the directed arc represents that the transmission prior relationship exists from the starting time to the target event;
step 2-4, defining three cutting operations applied to the TPG according to the TPG obtained in the step 2-3, and selecting cutting, sequential cutting and concurrent cutting, wherein the specific definition is as follows:
selecting and cutting: one selection cut may cut the TPG into multiple event sets Σ1,…,∑mThe event set must satisfy:
·
Figure BDA0003411164830000051
cutting in sequence: a sequential cut may cut the TPG into multiple ordered sets of events Σ1,…,∑mThe event set must satisfy:
·
Figure BDA0003411164830000052
·
Figure BDA0003411164830000053
concurrent cutting: one concurrent cut may cut the TPG into multiple event sets Σ1,…,∑mThe event set must satisfy:
·
Figure BDA0003411164830000054
wherein
Figure BDA0003411164830000055
Represents an event aiTo event ajThere is a transitive predecessor relationship between them, i.e. there is a directly reachable path in the TPG; and according to the sequence of the selective cutting, the sequential cutting and the concurrent cutting, selecting one cutting which can cut the TPG into a plurality of sub TPGs for carrying out the cutting operation, obtaining an event set corresponding to a plurality of sub TPGs, and recording the type of the cutting operation. Splitting corresponding sub-logs from incomplete sub-logs according to the event set, and repeating the splitting operation until all the sub-logs only contain one event; if the sub-log which cannot be normally split is encountered in the process, a flower model corresponding to the sub-log is returned.
And 3, combining all the block structures into a process model of the block structure represented by a process tree according to the nesting relation.
The present invention will be further described with reference to the following specific examples.
Examples
The invention relates to a mining method for a block structure process of an incomplete event log caused by concurrency. And outputting a block structure process model represented by a process tree by taking the event log described in the XES format as input.
With reference to the example, the method comprises the following specific operation steps:
step 1, fig. 2 shows a business process P modeled by a process tree, fig. 3 shows a partial display of an event log L generated by P, the file is described by an XES format, wherein a trace label represents an event sequence, and an event label represents an event; the event log comprises 6 event sequences (trace) in total, only the activity name of the event is considered, and the obtained event sequence set S is shown in FIG. 4; the preliminary result obtained by performing the segmentation through the segmentation operation based on the direct preceding relationship defined by the Inductive Miner is as follows: sequence of<{a1},{a2}, circulation of<Sequence of<{a3},{a4,a5,a6,a7,a8,a9},{a10}>,{a11}>,{a12}>(ii) a The event log is subjected to projection splitting according to the splitting result to obtain a process tree (block structure), and the expression of the process tree is as follows: sequence (a)1,a2Cycling (sequence)
Figure BDA0003411164830000061
Set a4,a5,a6,a7,a8,a9Corresponding sub-logs
Figure BDA0003411164830000062
Due to concurrent activities, the direct prior completeness of the index Miner requirement cannot be met, so that the segmentation operation cannot be normally carried out; and because the flower model does not contain repeated events or mutually exclusive occurrences, the flower model does not need to be generated.
Step 2, extracting the matters which cannot be processed in the step 1Part log
Figure BDA0003411164830000063
The previous relationship in (a) generates a 0-1 matrix as shown in fig. 5. In the delivery preceding matrix, there are 5 event pairs that cannot determine whether a delivery preceding relationship exists, respectively<a5,a4>，<a7,a4>，<a7,a6>，<a9,a4>，<a9,a8>The prediction score is calculated from small to large as: r<a7,a6>＝0.8917，R<a9,a8>＝0.8917，R<a5,a4>＝0.9442，R<a7,a4>＝0.9452，R<a9,a4>When the 2/3 digit 0.9442 is taken as the threshold value 0.9452, the event pair is determined<a7,a4>And<a9,a4>there is a transfer-ahead relationship; the result of the transfer before graph is obtained according to the prediction result, and the result is obtained by using the concurrent cutting applied to the transfer before graph as shown in FIG. 6: concurrence of<{a4,a5},{a6,a7},{a8,a9}>(ii) a According to the segmentation result pair
Figure BDA0003411164830000064
The projection splitting is performed to obtain a subprocess tree (block structure), which has the expression: concurrence of
Figure BDA0003411164830000065
Continue to sub-log
Figure BDA0003411164830000066
And
Figure BDA0003411164830000067
the resolution was carried out with the results: sequence (a)4,a5) Sequence (a)6,a7) Sequence (a)8,a9)。
Step 3, combining all the block structures in the step 2 to successfully obtain a process model shown in a process tree form as shown in fig. 1; the process model is consistent with the business process corresponding to the input event log, which shows that the invention can correctly process the incomplete event log caused by concurrency.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims (7)

1. A block structure process mining method aiming at incomplete event logs caused by concurrency is characterized in that a block structure process model is mined according to event log information; taking an event log in an XES format as input, taking a process model in a PTML format as output, wherein the process model is a process tree which is a process model with a block structure, leaf nodes are events in the event log, non-leaf nodes are one of selection, sequence, circulation and concurrency, and each sub-tree of the process tree is a block structure; the method comprises the following specific steps:
step 1, for an event log, segmenting events contained in the event log by utilizing a segmentation operation based on a direct previous relation in an Inductive Miner, wherein the result of the segmentation operation is a plurality of event sets, and splitting corresponding sub-logs from an original event log according to projections according to the event sets; at this time, a preliminary process tree can be obtained, the type of the splitting operation corresponds to the root node, and the split child logs are child nodes of the split child logs; continuously splitting the sub-logs by repeating the above operations until the sub-logs only contain one event; for the sub-event logs which cannot be normally split, judging whether repeated and mutually exclusive events exist in each event sequence, if so, returning a block structure flower model capable of generating any event sequence to the event log, otherwise, entering the step 2;
step 2, traversing the event sequence of the child logs which are not correctly processed in the step 1 to acquire the transfer preceding relation among the events and generating a corresponding 0-1 matrix, wherein an element 1 in the matrix represents that the transfer preceding relation exists, a 0 represents that the transfer preceding relation does not exist, and the rest represents uncertainty; using collaborative filtering to deduce uncertain elements in the 0-1 matrix to determine whether the uncertain elements have a transfer preceding relationship; then generating a transfer previous graph according to the inference result, carrying out cutting operation by using selection cutting, sequential cutting or concurrent cutting applied to the transfer previous graph, and splitting the event log according to the splitting result; if the sub-logs can be normally split, the block structure corresponding to the sub-logs can be obtained, otherwise, a flower model corresponding to the event log is returned;
and 3, combining all the block structures into a process model of the block structure represented by a process tree according to the nesting relation.
2. The method for mining the block structure process aiming at the incomplete event log caused by the concurrency according to claim 1, wherein the step 1 specifically comprises:
step 1-1, splitting the incomplete event log by using a splitting operation based on a direct previous relation defined in an Inductive Miner to determine the nesting relation between different block structures and the incomplete event log until the sub-log only contains one event, and further processing the sub-log which cannot be normally split;
and step 1-2, judging whether repeated events or mutually exclusive events exist in each event sequence contained in the sub-logs which cannot be normally processed in the step 1-1, if so, returning a flower model with a block structure to the event log, and otherwise, further processing the sub-logs.
3. The method for mining the block structure process aiming at the incomplete event log caused by the concurrency as claimed in claim 1, wherein the slicing operation in the step 1 comprises selection slicing, sequential slicing, loop slicing and concurrent slicing.
4. The method for mining the block structure process aiming at the incomplete event log caused by the concurrency according to claim 1, wherein the step 2 specifically comprises:
step 2-1, traversing all event sequences of the logs which cannot be processed in the step 1, and acquiring all event pairs with a transmission prior relationship; the method specifically comprises the following steps: any one event sequence a1,a2,a3,……,an-1,anContains n events, wherein aiRepresenting an event i, arranging the N events according to the occurrence sequence, and obtaining N x (N-1)/2 groups of delivery preceding relation pairs: a is1＞a2,a1＞a3,…,a1＞an,a2＞a3,a2＞a4,…,a2＞an,…,an-1＞an(ii) a Wherein, ai＞ajRepresenting event aiThe transfer occurs at event ajBefore; generating a pre-transmission 0-1 matrix according to the pre-transmission relation among the events, wherein the first row and the first column in the matrix are event names, if the rest elements are 1, the pre-transmission relation exists between the event pairs corresponding to the elements, if the rest elements are 0, the pre-transmission relation does not exist, and the rest elements do not determine whether the pre-transmission relation exists or not;
step 2-2, for the event pairs corresponding to the uncertain elements in the 0-1 matrix<ai,aj>Record aiIs located in the row vector [ a1,a2,……,ai-1,ai+1,……,aj-1,aj+1,……,an-1,an]Is v isi，ajIs located in the column vector [ a1,a2,……,ai-1,ai+1,……,aj-1,aj+1,……,an-1,an]TIs v isjWherein n is the number of events in the matrix, and uncertain elements are regarded as 0 in the calculation process; according to the formula
Figure FDA0003411164820000021
Wherein A and B represent a radicalQuantity, calculate viCosine similarity with other line vectors, except for ajFront of highest similarity outside the line vector
Figure FDA0003411164820000022
The event of each is recorded as a set Su(ii) a Calculating vjCosine similarity with other column vectors, except for aiTop of highest similarity outside the column vector
Figure FDA0003411164820000023
The event of each is recorded as a set Si；
Step 2-3, calculating S of all event pairs which are uncertain whether the transfer previous relation exists or not according to the step 2-2uAnd SiSet, for each undetermined element M in the matrixx,yThe inference result is Ratingx,yAccording to the formula
Figure FDA0003411164820000024
Is calculated, where M isr,cRepresenting an event a in a matrixrAnd event acValue of the element, Mr,yAnd Mx,cIn the same way, the method for preparing the composite material,
Figure FDA0003411164820000031
r denotes a symbol from the set SuEvent a ofrC denotes a symbol from the set SiEvent a ofcλ and δ are linear interpolation parameters; arranging all the inference results in ascending order, and taking the total number of the inference results as m
Figure FDA0003411164820000032
The number is used as a threshold value, the inference result is larger than the threshold value or reaches 1, the corresponding event pair is judged to have the transmission prior relationship, otherwise, the corresponding event pair does not exist, the transmission prior graph TPG is generated by combining the previously determined event pair with the transmission prior relationship, and the vertex is the eventA directed arc representing a transitive preceding relationship from a start time to a target event;
step 2-4, defining three cutting operations applied to the TPG according to the TPG obtained in the step 2-3, and selecting cutting, sequential cutting and concurrent cutting, wherein the specific definition is as follows:
selecting and cutting: one selection cut may cut the TPG into multiple event sets Σ1,…,∑mThe event set must satisfy:
Figure FDA0003411164820000033
cutting in sequence: a sequential cut may cut the TPG into multiple ordered sets of events Σ1,…,∑mThe event set must satisfy:
Figure FDA0003411164820000034
Figure FDA0003411164820000035
concurrent cutting: one concurrent cut may cut the TPG into multiple event sets Σ1,…,∑mThe event set must satisfy:
Figure FDA0003411164820000036
wherein
Figure FDA0003411164820000037
Represents an event aiTo event ajThere is a transitive predecessor relationship between them, i.e. there is a directly reachable path in the TPG; according to the sequence of selective cutting, sequential cutting and concurrent cutting, selecting a cutting capable of cutting TPG into a plurality of sub TPGs for cutting operation, and obtaining events corresponding to a plurality of sub TPGsCollecting and recording the type of the segmentation operation; splitting corresponding sub-logs from incomplete sub-logs according to the event set, and repeating the splitting operation until all the sub-logs only contain one event; if the sub-log which cannot be normally split is encountered in the process, a flower model corresponding to the sub-log is returned.
5. The method of block structured process mining for concurrency driven incomplete event logs according to claim 4, wherein λ and δ have values of 0.5.
6. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the method for block structured process mining for concurrency-induced incomplete event logs according to any one of claims 1 to 5 when executing the program.
7. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements a method of block-structured process mining for concurrency-induced incomplete event logs according to any one of claims 1 to 5.
CN202111529372.XA 2021-12-14 2021-12-14 Block structure process mining method for incomplete event log caused by concurrency Pending CN114201460A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111529372.XA CN114201460A (en) 2021-12-14 2021-12-14 Block structure process mining method for incomplete event log caused by concurrency

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111529372.XA CN114201460A (en) 2021-12-14 2021-12-14 Block structure process mining method for incomplete event log caused by concurrency

Publications (1)

Publication Number Publication Date
CN114201460A true CN114201460A (en) 2022-03-18

Family

ID=80653687

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111529372.XA Pending CN114201460A (en) 2021-12-14 2021-12-14 Block structure process mining method for incomplete event log caused by concurrency

Country Status (1)

Country Link
CN (1) CN114201460A (en)

Similar Documents

Publication Publication Date Title
US10733149B2 (en) Template based data reduction for security related information flow data
CN111726248A (en) Alarm root cause positioning method and device
US11841839B1 (en) Preprocessing and imputing method for structural data
CN109118155B (en) Method and device for generating operation model
JP2007328712A (en) Time series pattern generation system and time series pattern generating method
US8326982B2 (en) Method and apparatus for extracting and visualizing execution patterns from web services
CN104573124A (en) Education cloud application statistics method based on parallelized association rule algorithm
CN114430365B (en) Fault root cause analysis method, device, electronic equipment and storage medium
CN110990403A (en) Business data storage method, system, computer equipment and storage medium
CN110275889B (en) Feature processing method and device suitable for machine learning
CN110941554A (en) Method and device for reproducing fault
CN115455429A (en) Vulnerability analysis method and system based on big data
CN112052232B (en) Business process context extraction method based on replay technology
CN109063040B (en) Client program data acquisition method and system
CN112949778A (en) Intelligent contract classification method and system based on locality sensitive hashing and electronic equipment
CN114201460A (en) Block structure process mining method for incomplete event log caused by concurrency
CN116015939A (en) Advanced persistent threat interpretation method based on atomic technology template
CN115409541A (en) Cigarette brand data processing method based on data blood relationship
CN115688853A (en) Process mining method and system
CN115049060A (en) Business process task execution knowledge recommendation method based on deep learning
CN114756602A (en) Real-time streaming process mining method and system and computer readable storage medium
US8631391B2 (en) Method and a system for process discovery
CN114465875A (en) Fault processing method and device
KR20220115859A (en) Edge table representation of the process
CN112750047A (en) Behavior relation information extraction method and device, storage medium and electronic equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination