CN115292258A - Cross-organization multi-source heterogeneous business process event log fusion method and system - Google Patents
Cross-organization multi-source heterogeneous business process event log fusion method and system Download PDFInfo
- Publication number
- CN115292258A CN115292258A CN202210897941.4A CN202210897941A CN115292258A CN 115292258 A CN115292258 A CN 115292258A CN 202210897941 A CN202210897941 A CN 202210897941A CN 115292258 A CN115292258 A CN 115292258A
- Authority
- CN
- China
- Prior art keywords
- event log
- business process
- population
- initial population
- individuals
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 168
- 230000008569 process Effects 0.000 title claims abstract description 158
- 238000007500 overflow downdraw method Methods 0.000 title claims description 12
- 230000004927 fusion Effects 0.000 claims abstract description 61
- 238000010367 cloning Methods 0.000 claims abstract description 36
- 230000035772 mutation Effects 0.000 claims abstract description 35
- 230000009133 cooperative interaction Effects 0.000 claims description 13
- 238000012163 sequencing technique Methods 0.000 claims description 12
- 230000008859 change Effects 0.000 claims description 10
- 230000008520 organization Effects 0.000 claims description 7
- 238000005065 mining Methods 0.000 abstract description 12
- 238000005516 engineering process Methods 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 6
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 101150095495 sctL1 gene Proteins 0.000 description 1
- 239000012085 test solution Substances 0.000 description 1
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
- G06F16/10—File systems; File servers
- G06F16/18—File system types
- G06F16/1805—Append-only file systems, e.g. using logs or journals to store data
- G06F16/1815—Journaling file systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention discloses a method and a system for fusing cross-organization multi-source heterogeneous business process event logs, which comprise the following steps: 1) Acquiring data and converting the data into an isomorphic service process event log; 2) Randomly matching tracks in event logs of any two isomorphic service processes to form an initial population; 3) Obtaining a random initial population and cloning to generate a clone population; 4) Selecting mutation and updating clone population; 5) Iteratively generating a fusion event log, repeating the steps 3) -5) until unselected individuals in the initial population are lower than a threshold value, obtaining a fusion population of isomorphic service process event logs, and arranging events in two tracks contained in each individual of the fusion population according to a timestamp to obtain the fusion event log; 6) And repeating the steps 2) -6) on the fusion event log and the isomorphic service process event log to be fused to obtain a cross-organization multi-source isomorphic service process event log after fusion. The invention breaks through the limitation that the existing flow mining technology and tools can not be utilized to analyze and improve the multi-organization business flow.
Description
Technical Field
The invention relates to the technical field of business process mining, in particular to a method and a system for fusing cross-organization multi-source heterogeneous business process event logs.
Background
With the rapid development of economic globalization and high and new technologies, cross-organization cooperative business processes in which a plurality of organizations cooperate to complete the same business target have become more and more common, a large number of event logs are generated in the cooperation process, and information systems in different organizations respectively record respective event logs, because the organization uses different information systems, the recorded event logs have case identifications which are difficult to unify, and the event logs have various complex matching relations, so that the difficulty is brought to the analysis of the whole cross-organization business process; in addition, the current business process mining tool takes a single event log as input for mining, modeling and analyzing, so that the event logs of the multi-source heterogeneous business process in the cross-organization business process are fused to obtain the single event log, and the current process mining technology and tool are applied to model mining and analyzing the cross-organization business process, so that the problems existing in the current cross-organization business process can be solved in a targeted manner.
Disclosure of Invention
The invention aims to overcome the defects and shortcomings of the prior art, provides a cross-organization multi-source heterogeneous business process event log fusion method, and breaks through the limitation that the existing process mining technology and tools cannot be used for analyzing and improving the multi-organization business process.
The invention also provides a cross-organization multi-source heterogeneous business process event log fusion system.
The first purpose of the invention is realized by the following technical scheme: a cross-organization multi-source heterogeneous business process event log fusion method comprises the following steps:
1) Acquiring basic data, namely cross-organization multi-source heterogeneous business process event logs, and converting the acquired cross-organization multi-source heterogeneous business process event logs into isomorphic business process event logs;
2) Obtaining an initial population: randomly selecting two event logs from the plurality of isomorphic service process event logs obtained in the step 1), and randomly matching tracks in the two event logs to form an initial population as an individual;
3) Obtaining a random initial population and cloning: randomly extracting a plurality of individuals from the initial population obtained in the step 2) to be used as a random initial population, calculating the affinity of each individual in the random initial population, sequencing, and selecting the top n% of individuals with high affinity values to clone so as to generate a new clone population;
4) Selection of mutations: selecting a specific number of individual clone populations to mutate so as to generate new individual, calculating the affinity of the new individual, comparing the affinity with the highest affinity value of the clone populations, accepting the mutant individual if the affinity of the mutant individual is higher, and rejecting the mutant individual if the affinity of the mutant individual is not higher;
5) And (3) iteratively generating a fusion event log: selecting some tracks from the remaining individuals of the initial population again to add into the random initial population, repeating the steps 3), 4) and 5) until the unselected individuals in the initial population are lower than the threshold value to obtain a fusion population of two event logs, and arranging the events in the two tracks contained in each individual of the fusion population according to the time stamps to obtain the fusion event logs;
6) Randomly extracting one event log from the rest isomorphic service process event logs in the step 1) to fuse with the fusion event log obtained in the step 5), and repeating the steps 2), 3), 4), 5) and 6) until the event log to be fused is empty, thereby finally obtaining the cross-organization multi-source isomorphic service process event log after fusion.
Further, in step 1), obtaining a cross-organization multi-source heterogeneous business process event log and isomorphically structuring the heterogeneous business process event log, specifically comprising the following steps:
1.1 Obtaining cross-organization multi-source heterogeneous business process event logs; the cross-organization multi-source heterogeneous business process event log refers to a business process event log recorded in a plurality of organization internal information management systems which finish the same business process together, and heterogeneous business process event logs can be obtained due to different information management systems used by all organizations; the event log is a collection of finite event sequences, each of which is referred to as a trace;
1.2 Converting the format of the heterogeneous service process event log obtained in the step 1.1) into the same data format to form a cross-organization multi-source isomorphic service process event log;
the heterogeneous service process event log refers to an event log which is not acquired in the same system; the isomorphic business process event logs refer to event logs of the same data structure.
Further, in step 3), a random initial population is selected, and a specific number of individuals are selected for cloning to generate a new clone population, which specifically comprises the following steps:
3.1 Randomly extracting some tracks from the initial population obtained in the step 2) to be used as a random initial population;
3.2 Calculating the affinity f of each individual in the random initial population according to the formula (1) and sequencing;
f＝∑ST i +∑CI i +∑TD i (1)
in the formula, ST i Representing that two tracks matched in the ith individual have the same track identification; CI i Collaborative interaction information representing two tracks matched in the ith individual; TD i Representing the difference of the matched two track time stamps in the ith individual;
the collaborative interaction information is behavior constraint existing among events in tracks of different event logs;
3.3 From the individuals ranked from high to low according to affinity values obtained in step 3.2) the first n% of the individuals with high affinity values are selected for cloning, resulting in a new clonal population.
Further, in step 4), selecting a specific number of individual clones to perform mutation to generate new individuals, calculating the affinity of the new individuals, and comparing the affinity with the highest affinity value of the clone population, specifically comprising the following steps:
4.1 Num individuals are selected from the clone population obtained in the step 3), and the selection formula of Num is shown as a formula (2):
Num＝max(1,(1-e -(B-C) )*m) (2)
wherein B is the highest value of affinity in the current population; c is the affinity value of the current mutant individual; and m is the number of individuals of the current clone population.
4.2 Num individuals selected in the step 4.1) are mutated, and the mutation is the matching relation of changing the selected individuals to contain two tracks: add, delete, or change;
4.3 Using the affinity formula (1) to calculate the affinity value of the mutated individual and comparing with the highest affinity value of the clonal population, if the affinity value of the mutated individual is higher than the highest affinity value of the clonal population, accepting the mutated individual, otherwise rejecting the mutated individual.
Further, in step 5), adding new individuals to the random initial population, repeatedly calculating affinity values, cloning, selecting mutation, and stopping iteration until conditions are met to obtain a new fusion event log, specifically comprising the following steps:
5.1 Adding new individuals to the random initial population, namely randomly selecting some individuals from the rest individuals in the initial population again to add to the random initial population;
5.2 Step 3), step 4) and step 5) are repeated until all the individuals in the initial population are randomly extracted;
5.3 And) sequencing all events in the individual containing two tracks in the population obtained after the iteration is finished in the step 5.2) according to the time stamps to obtain a fusion event log of the two event logs.
The second purpose of the invention is realized by the following technical scheme: a cross-organization multi-source heterogeneous business process event log fusion system comprises an event log acquisition and isomorphism module, an initial population acquisition module, a random initial population acquisition and cloning module, a mutation selection module and a fusion event log iterative generation module;
the event log obtaining and isomorphic module is used for obtaining cross-organization multi-source heterogeneous business process event logs and isomorphizing the heterogeneous business process event logs to form cross-organization multi-source isomorphic business process event logs;
the initial population acquisition module is used for randomly selecting two event logs from a plurality of isomorphic service process event logs acquired by the event log acquisition and isomorphic module, and matching tracks in the two event logs randomly to form an initial population as an individual;
the random initial population acquisition and cloning module is used for randomly selecting individuals in the initial population, calculating the affinity, and cloning the first n% of individuals with high affinity values;
the mutation selection module is used for selecting Num individuals from the cloned population to perform mutation, namely adding, deleting and changing track matching;
the iterative fusion event log generation module is used for repeatedly carrying out the initial population acquisition module, the random initial population acquisition and cloning module and the selection mutation module to fuse the isomorphic business process event logs to be fused, and finally forming a fused cross-organization multi-source isomorphic business process event log.
Further, the event log obtaining and isomorphic module specifically executes the following operations:
acquiring a cross-organization multi-source heterogeneous business process event log; converting the format of the heterogeneous service process event logs into the same data format to form a cross-organization multi-source isomorphic service process event log; the cross-organization multi-source heterogeneous business process event log refers to an event log recorded in a plurality of organization internal information management systems which finish the same business process together, and the heterogeneous business process event log can be obtained due to different information management systems used by various organizations; the event log is a collection of finite event sequences, each of which is referred to as a trace;
the heterogeneous service process event log refers to an event log which is not acquired in the same system; the isomorphic business process event logs refer to event logs of the same data structure.
Further, the random initial population obtaining and cloning module specifically performs the following operations:
randomly extracting some tracks from the initial population obtained by the initial population obtaining module to serve as a random initial population; calculating the affinity f of each individual in the random initial population according to a formula (1), arranging the individuals from high to low according to the affinity, and selecting the first n percent of individuals with high affinity values for cloning to generate a new clone population;
f＝∑ST i +∑CI i +∑TD i (1)
in the formula, ST i Representing that two tracks matched in the ith individual have the same track identification; CI i Collaborative interaction information representing two tracks matched in the ith individual; TD i Representing the difference of the matched two track time stamps in the ith individual;
the collaborative interaction information is behavior constraints existing among events in the tracks of different event logs.
Further, the selection mutation module specifically performs the following operations:
num individuals are selected from the clone population obtained by the random initial population acquisition and clone module, and the selection formula of Num is shown as a formula (2):
Num＝max(1,(1-e -(B-C) )*m) (2)
wherein B is the highest value of affinity in the current population; c is the affinity value of the current mutant individual; m is the number of the current clone population individuals;
mutating the selected Num individuals, wherein the mutation is to change the matching relationship of the two tracks contained in the selected individuals: add, delete, or change; and then, calculating the affinity value of the mutated individual by using an affinity formula (1), comparing the affinity value with the highest affinity value of the clone population, accepting the mutated individual if the affinity value of the mutated individual is higher than the highest affinity value of the clone population, and rejecting the mutated individual if the affinity value of the mutated individual is not higher than the highest affinity value of the clone population.
Further, the iteration generation fusion event log module specifically executes the following operations:
adding new individuals into the random initial population, namely, randomly selecting some individuals from the rest individuals in the initial population again to add into the random initial population; repeating the random initial population obtaining and cloning module, the mutation selecting module and the fusion event log iterative generation module until all individuals in the initial population are randomly extracted; sequencing all events in the individual containing two tracks in the population obtained after the iteration is finished according to the time stamps to obtain a fusion event log with two event logs in the same structure; and then randomly extracting an event log from the event logs in the rest isomorphic service process event logs in the event log acquisition and isomorphic modules and fusing the event logs, repeating the initial population acquisition module, the random initial population acquisition and cloning module, the mutation selection module and the fusion event log generation module until the isomorphic service process event logs to be fused are empty, and finally obtaining a fused cross-organization multi-source isomorphic service process event log.
Compared with the prior art, the invention has the following advantages and beneficial effects:
1. the invention provides a multi-source heterogeneous business process event log fusion method, which fuses multi-source heterogeneous event logs into a single event log and is beneficial to modeling and mining of a cross-organization business process model.
2. The invention applies the multi-source heterogeneous business process event log fusion to the cross-organization business process in the process mining field for the first time, provides an event log fusion method for the event logs dispersed in the cross-organization business process, and can fully utilize the existing process mining tool to analyze the cross-organization business process by the method.
3. The invention provides the affinity function for the first time, takes the cooperative interaction information in the cross-organization business process as an index of the affinity function, and is beneficial to selecting the track with high affinity for matching, thereby obtaining the optimal matching track.
4. The invention provides a cross-organization multi-source heterogeneous business process event log fusion method based on the concept of affinity in artificial immune thinking and a genetic algorithm, fuses multi-source heterogeneous business process event logs, and can solve various complex matching relations existing in tracks in the cross-organization business process event logs.
5. The invention has wide use space in the cross-organization multi-source heterogeneous business process event log fusion and has wide prospect in the multi-source heterogeneous business process event log fusion.
Drawings
FIG. 1 is a logic flow diagram of the method of the present invention.
Fig. 2 is a schematic diagram of an xe format event log in an event log of a cross-organized multi-source heterogeneous business process in this embodiment.
Fig. 3 is a schematic diagram of a CSV format event log in a cross-organization multi-source heterogeneous service flow event log in this embodiment.
Fig. 4 is a schematic diagram of the event log after the event log in the CSV format is converted into the XES format in this embodiment.
Fig. 5 is a schematic diagram of an event log of a cross-organization multi-source isomorphic service process obtained by final fusion in this embodiment.
Fig. 6 is a system architecture diagram of the present invention.
Detailed Description
The present invention will be further described with reference to the following specific examples.
Example 1
As shown in fig. 1, the embodiment discloses a method for fusing event logs of a cross-organization multi-source heterogeneous business process, which includes the following steps:
1) The method comprises the following steps of obtaining basic data, namely cross-organization multi-source heterogeneous business process event logs, and converting the obtained cross-organization multi-source heterogeneous business process event logs into isomorphic business process event logs, and specifically comprises the following steps:
1.1 Obtaining cross-organization multi-source heterogeneous business process event logs; the cross-organization multi-source heterogeneous business process event log refers to a business process event log recorded in a plurality of organization internal information management systems which finish the same business process together, and heterogeneous business process event logs can be obtained due to different information management systems used by all organizations; the event log is a collection of finite event sequences, each of which is referred to as a trace;
1.2 Converting the format of the heterogeneous service process event log obtained in the step 1.1) into the same data format to form a cross-organization multi-source isomorphic service process event log;
the heterogeneous service process event log refers to an event log which is not acquired in the same system; the isomorphic business process event logs refer to event logs of the same data structure;
by adopting the steps, two event logs of the back-end IT support process are obtained, including active AT (start), AP (analysis problem), PC (program change), AD (adjustment document), TS (test solution problem), SI (search problem), SM (mail sending) and CT (close). Wherein, the active AT, AP, SI, SM, CT are recorded in the XES event log, which is OrgA, and part of the event log is shown in fig. 2; the activity PC, AD, TS are recorded in the event log of CSV format, orgB, and part of the event log is shown in FIG. 3; and converts the event log in the CSV format into an event log in the XES format, so as to obtain the event log in the XES format as shown in fig. 4.
2) Obtaining an initial population: randomly selecting two event logs from the event logs of the isomorphic service process obtained in the step 1), and randomly matching tracks in the two event logs to form an initial population as an individual;
3) Obtaining a random initial population and cloning: randomly extracting 20% of individuals from the initial population obtained in the step 2) to be used as a random initial population, calculating the affinity of each individual in the random initial population, sequencing, selecting the first 20% of individuals with high affinity values for cloning, and generating a new clone population, wherein the method specifically comprises the following steps:
3.1 Randomly drawing 20% of tracks from the initial population obtained in the step 2) as a random initial population;
3.2 Calculating the affinity f of each individual in the random initial population according to the formula (1) and sequencing;
f＝∑ST i +∑CI i +∑TD i (1)
in the formula, ST i Representing that two tracks matched in the ith individual have the same track identification; CI i Collaborative interaction information representing two tracks matched in the ith individual; TD i Representing the difference of the matched two track time stamps in the ith individual;
the collaborative interaction information is behavior constraint existing among events in tracks of different event logs;
3.3 From the high-to-low arranged individuals according to affinity values obtained in step 3.2) the first 20% of the high-affinity value individuals are selected for cloning, resulting in a new clonal population.
4) Selection of mutations: selecting a certain number of individual clone populations to mutate so as to generate new individual, calculating the affinity of the new individual, comparing the affinity with the highest affinity value of the clone populations, accepting the mutant individual if the affinity of the mutant individual is higher, and rejecting the mutant individual if the affinity of the mutant individual is higher, specifically comprising the following steps:
4.1 Num individuals are selected from the clone population obtained in the step 3), and the selection formula of Num is shown as a formula (2):
Num＝max(1,(1-e -(B-C) )*m) (2)
wherein B is the highest value of affinity in the current population; c is the affinity value of the current mutant individual; m is the number of the current clone population individuals;
4.2 Num individuals selected in the step 4.1) are mutated, and the mutation is the matching relation of changing the selected individuals to contain two tracks: add, delete, or change;
4.3 Using the affinity formula (1) to calculate the affinity value of the mutated individual and comparing with the highest affinity value of the clonal population, if the affinity value of the mutated individual is higher than the highest affinity value of the clonal population, accepting the mutated individual, otherwise rejecting the mutated individual.
5) And (3) iteratively generating a fusion event log: selecting 20% of tracks from the remaining individuals of the initial population again to add to the random initial population, repeating the steps 3), 4) and 5) until the unselected individuals in the initial population are lower than the threshold value 1 (the threshold value is set as 1 in the experiment), obtaining a fusion population of the two event logs, and arranging the events in the two tracks contained in each individual of the fusion population according to the time stamps to obtain a fusion event log, wherein the method specifically comprises the following steps:
5.1 Adding new individuals to the random initial population, namely randomly selecting 20% of individuals from the rest individuals in the initial population again to add to the random initial population;
5.2 Step 3), step 4) and step 5) are repeated until all the individuals in the initial population are randomly extracted;
5.3 And) sequencing all events in the individual containing two tracks in the population obtained after the iteration is finished in the step 5.2) according to the time stamps to obtain a fusion event log of the two event logs.
6) Randomly extracting one event log from the rest isomorphic service process event logs in the step 1) to fuse with the fusion event log obtained in the step 5), and repeating the steps 2), 3), 4), 5) and 6) until the event log to be fused is empty, thereby finally obtaining the cross-organization multi-source isomorphic service process event log after fusion.
By adopting the steps, the event log after the cross-organization multi-source heterogeneous business process event log fusion is shown in fig. 5.
Example 2
The embodiment discloses a cross-organization multi-source heterogeneous business process event log fusion system, as shown in fig. 6, the system includes the following functional modules: the system comprises an event log acquisition and isomorphism module, an initial population acquisition module, a random initial population acquisition and cloning module, a mutation selection module and an iterative generation and fusion event log module.
The event log obtaining and isomorphic module is used for obtaining cross-organization multi-source heterogeneous business process event logs and isomorphizing the heterogeneous business process event logs to form cross-organization multi-source isomorphic business process event logs;
the initial population acquisition module is used for randomly selecting two event logs from a plurality of isomorphic service process event logs acquired by the event log acquisition and isomorphic module, and matching tracks in the two event logs randomly to form an initial population as an individual;
the random initial population acquisition and cloning module is used for randomly selecting individuals in an initial population, calculating the affinity, and cloning the first 20% of individuals with high affinity values;
the mutation selection module is used for selecting Num individuals from the cloned population to perform mutation, namely adding, deleting and changing track matching;
the iterative fusion event log generation module is used for repeatedly carrying out the initial population acquisition module, the random initial population acquisition and cloning module and the selection mutation module to fuse the isomorphic business process event logs to be fused, and finally forming a fused cross-organization multi-source isomorphic business process event log.
Further, the event log obtaining and isomorphic module specifically performs the following operations:
acquiring a cross-organization multi-source heterogeneous business process event log; and converting the format of the heterogeneous business process event log into the same data format to form a cross-organization multi-source isomorphic business process event log. The cross-organization multi-source heterogeneous business process event log refers to event logs recorded in a plurality of organization internal information management systems which finish the same business process together, and heterogeneous business process event logs can be obtained due to different information management systems used by all organizations; the event log is a collection of finite event sequences, each of which is referred to as a trace;
the heterogeneous event log refers to an event log which is not acquired in the same system; the homogeneous event log refers to an event log with the same data structure.
Further, the random initial population obtaining and cloning module specifically performs the following operations:
randomly extracting 20% of tracks from the initial population obtained by the initial population obtaining module to serve as a random initial population; and (3) calculating the affinity f of each individual in the random initial population according to the formula (1), arranging the individuals from high to low according to the affinity, and selecting the first 20 percent of individuals with high affinity values for cloning to generate a new clone population.
f＝∑ST i +∑CI i +∑TD i (1)
In the formula, ST i Representing that two tracks matched in the ith individual have the same track identification; CI i Collaborative interaction information representing two tracks matched in the ith individual; TD i Representing the difference of the matched two track time stamps in the ith individual;
the collaborative interaction information is behavior constraints existing among events in the tracks of different event logs.
Further, the selection mutation module specifically performs the following operations:
num individuals are selected from the clone population obtained by the random initial population acquisition and clone module, and the selection formula of Num is shown as a formula (2):
Num＝max(1,(1-e -(B-C) )*m) (2)
wherein B is the highest value of affinity in the current population; c is the affinity value of the current mutant individual; m is the number of the current clone population individuals;
carrying out mutation on the Num individuals selected, wherein the mutation is the change of the matching relation of the two tracks contained in the selected individuals: add, delete, or change; and then, calculating the affinity value of the mutated individual by using an affinity formula (1), comparing the affinity value with the highest affinity value of the clone population, accepting the mutated individual if the affinity value of the mutated individual is higher than the highest affinity value of the clone population, and rejecting the mutated individual if the affinity value of the mutated individual is not higher than the highest affinity value of the clone population.
Further, the iteration generation fusion event log module specifically executes the following operations:
adding new individuals into the random initial population, namely, randomly selecting 20% of individuals from the rest individuals in the initial population again to add into the random initial population; repeating the random initial population obtaining and cloning module, the mutation selecting module and the fusion event log iterative generation module until all individuals in the initial population are randomly extracted; sequencing all events in the individual containing two tracks in the population obtained after the iteration is finished according to the time stamps to obtain a fusion event log with two event logs in the same structure; and then randomly extracting an event log from the event logs of the rest isomorphic business process in the event log acquisition and isomorphic modules and fusing the event logs, repeating the initial population acquisition module, the random initial population acquisition and cloning module, the mutation selection module and the fusion event log generation module until the event logs of the isomorphic business process to be fused are empty, and finally obtaining a fused cross-organization multi-source isomorphic business process event log.
In conclusion, after the scheme is adopted, the invention provides a new method and a new system for the cross-organization multi-source heterogeneous business process event log fusion, which can fuse the heterogeneous business process event logs dispersed in the actual scene, break through the limitation of dispersion and heterogeneity of the event logs during the mining of the original cross-organization business process model, facilitate the subsequent analysis of the business process, effectively promote the development of the cross-organization business process mining technology, have practical application value and are worthy of popularization.
The above-mentioned embodiments are merely preferred embodiments of the present invention, and the scope of the present invention is not limited thereto, so that the changes in the shape and principle of the present invention should be covered within the protection scope of the present invention.
Claims (10)
1. A cross-organization multi-source heterogeneous business process event log fusion method is characterized by comprising the following steps:
1) Acquiring basic data, namely cross-organization multi-source heterogeneous business process event logs, and converting the acquired cross-organization multi-source heterogeneous business process event logs into isomorphic business process event logs;
2) Obtaining an initial population: randomly selecting two event logs from the plurality of isomorphic service process event logs obtained in the step 1), and randomly matching tracks in the two event logs to form an initial population as an individual;
3) Obtaining a random initial population and cloning: randomly extracting a plurality of individuals from the initial population obtained in the step 2) to be used as a random initial population, calculating the affinity of each individual in the random initial population, sequencing, and selecting the top n% of individuals with high affinity values to clone so as to generate a new clone population;
4) Selection of mutations: selecting a specific number of individual clone populations to mutate so as to generate new individual, calculating the affinity of the new individual, comparing the affinity with the highest affinity value of the clone populations, accepting the mutant individual if the affinity of the mutant individual is higher, and rejecting the mutant individual if the affinity of the mutant individual is not higher;
5) And (3) iteratively generating a fusion event log: selecting some tracks from the remaining individuals of the initial population again to add into the random initial population, repeating the steps 3), 4) and 5) until the unselected individuals in the initial population are lower than the threshold value to obtain a fusion population of two event logs, and arranging the events in the two tracks contained in each individual of the fusion population according to the time stamps to obtain the fusion event logs;
6) Randomly extracting one event log from the rest isomorphic service process event logs in the step 1) to fuse with the fusion event log obtained in the step 5), and repeating the steps 2), 3), 4), 5) and 6) until the event log to be fused is empty, thereby finally obtaining the cross-organization multi-source isomorphic service process event log after fusion.
2. The cross-organization multi-source heterogeneous business process event log fusion method according to claim 1, characterized in that: in step 1), acquiring a cross-organization multi-source heterogeneous business process event log and isomorphism the heterogeneous business process event log, specifically comprising the following steps:
1.1 Obtaining cross-organization multi-source heterogeneous business process event logs; the cross-organization multi-source heterogeneous business process event log refers to a business process event log recorded in a plurality of organization internal information management systems which finish the same business process together, and heterogeneous business process event logs can be obtained due to different information management systems used by all organizations; the event log is a collection of finite event sequences, each of which is referred to as a trace;
1.2 Converting the format of the heterogeneous service process event log obtained in the step 1.1) into the same data format to form a cross-organization multi-source isomorphic service process event log;
the heterogeneous service process event log refers to an event log which is not acquired in the same system; the isomorphic business process event logs refer to event logs of the same data structure.
3. The cross-organization multi-source heterogeneous business process event log fusion method according to claim 1, characterized in that: in step 3), selecting a random initial population, and selecting a specific number of individuals for cloning to generate a new clone population, specifically comprising the following steps:
3.1 Randomly extracting some tracks from the initial population obtained in the step 2) to be used as a random initial population;
3.2 Calculating the affinity f of each individual in the random initial population according to the formula (1) and sequencing;
f＝∑ST i +∑CI i +∑TD i (1)
in the formula, ST i Representing that two tracks matched in the ith individual have the same track identification; CI i Collaborative interaction information representing two tracks matched in the ith individual; TD (time division) i Representing the difference of the matched two track time stamps in the ith individual;
the collaborative interaction information is behavior constraint existing among events in tracks of different event logs;
3.3 From the individuals ranked from high to low according to affinity values obtained in step 3.2) the first n% of the individuals with high affinity values are selected for cloning, resulting in a new clonal population.
4. The cross-organization multi-source heterogeneous business process event log fusion method according to claim 3, characterized in that: in step 4), selecting a specific number of individual clonal populations to mutate to generate new individuals, calculating the affinity of the new individuals, and comparing the affinity with the highest affinity value of the clonal populations, specifically comprising the following steps:
4.1 Num) individuals are selected from the clone population obtained in step 3), and the selection formula of Num is shown as formula (2):
Num＝max(1,(1-e -(B-C) )*m) (2)
wherein B is the highest value of affinity in the current population; c is the affinity value of the current mutant individual; and m is the number of individuals of the current clone population.
4.2 Num individuals selected in the step 4.1) are mutated, and the mutation is the matching relation of changing the selected individuals to contain two tracks: add, delete, or change;
4.3 Using the affinity formula (1) to calculate the affinity value of the mutated individual and comparing with the highest affinity value of the clonal population, if the affinity value of the mutated individual is higher than the highest affinity value of the clonal population, accepting the mutated individual, otherwise rejecting the mutated individual.
5. The cross-organization multi-source heterogeneous business process event log fusion method according to claim 1, characterized in that: in step 5), adding new individuals to the random initial population, repeatedly calculating affinity values, cloning and selecting mutation until conditions are met, and stopping iteration to obtain a new fusion event log, wherein the method specifically comprises the following steps:
5.1 Adding new individuals to the random initial population, namely randomly selecting some individuals from the rest individuals in the initial population again to add to the random initial population;
5.2 Step 3), step 4) and step 5) are repeated until all the individuals in the initial population are randomly extracted;
5.3 And) sequencing all events in the individual containing two tracks in the population obtained after the iteration is finished in the step 5.2) according to the time stamps to obtain a fusion event log of the two event logs.
6. A cross-organization multi-source heterogeneous business process event log fusion system is characterized by comprising an event log acquisition and isomorphism module, an initial population acquisition module, a random initial population acquisition and cloning module, a mutation selection module and an iterative generation fusion event log module;
the event log obtaining and isomorphic module is used for obtaining cross-organization multi-source heterogeneous business process event logs and isomorphizing the heterogeneous business process event logs to form cross-organization multi-source isomorphic business process event logs;
the initial population acquisition module is used for randomly selecting two event logs from a plurality of isomorphic service process event logs obtained by the event log acquisition and isomorphic module, and randomly matching tracks in the two event logs to form an initial population as an individual;
the random initial population acquisition and cloning module is used for randomly selecting individuals in the initial population, calculating the affinity, and cloning the first n% of individuals with high affinity values;
the mutation selection module is used for selecting Num individuals from the cloned population to perform mutation, namely adding, deleting and changing track matching;
the iterative fusion event log generation module is used for repeatedly performing the initial population acquisition module, the random initial population acquisition and cloning module and the selection mutation module to fuse the isomorphic business process event logs to be fused, and finally forming a fused cross-organization multi-source isomorphic business process event log.
7. The cross-organization multi-source heterogeneous business process event log fusion system of claim 6, wherein: the event log obtaining and isomorphic module specifically executes the following operations:
acquiring a cross-organization multi-source heterogeneous business process event log; converting the format of the heterogeneous business process event logs into the same data format to form a cross-organization multi-source isomorphic business process event log; the cross-organization multi-source heterogeneous business process event log refers to event logs recorded in a plurality of organization internal information management systems which finish the same business process together, and heterogeneous business process event logs can be obtained due to different information management systems used by all organizations; the event log is a collection of finite event sequences, each of which is referred to as a trace;
the heterogeneous service process event log refers to an event log which is not acquired in the same system; the isomorphic business process event logs refer to event logs of the same data structure.
8. The cross-organization multi-source heterogeneous business process event log fusion system of claim 6, wherein: the random initial population obtaining and cloning module specifically executes the following operations:
randomly extracting some tracks from the initial population obtained by the initial population obtaining module to serve as a random initial population; calculating the affinity f of each individual in the random initial population according to a formula (1), arranging the individuals from high to low according to the affinity, and selecting the first n percent of individuals with high affinity values for cloning to generate a new clone population;
f＝∑ST i +∑CI i +∑TD i (1)
in the formula, ST i Representing that two tracks matched in the ith individual have the same track identification; CI i Collaborative interaction information representing two tracks matched in the ith individual; TD i Representing the difference of the matched two track time stamps in the ith individual;
the collaborative interaction information is behavior constraints existing among events in the tracks of different event logs.
9. The cross-organization multi-source heterogeneous business process event log fusion system of claim 8, wherein: the selection mutation module specifically performs the following operations:
num individuals are selected from the clone population obtained by the random initial population acquisition and clone module, and the selection formula of Num is shown as a formula (2):
Num＝max(1,(1-e -(B-C) )*m) (2)
wherein, B is the highest value of the affinity in the current population; c is the affinity value of the current mutant individual; m is the number of the current clone population individuals;
mutating the selected Num individuals, wherein the mutation is the change of the matching relation of the selected individuals comprising two tracks: add, delete, or change; and then, calculating the affinity value of the mutated individual by using an affinity formula (1), comparing the affinity value with the highest affinity value of the clone population, accepting the mutated individual if the affinity value of the mutated individual is higher than the highest affinity value of the clone population, and rejecting the mutated individual if the affinity value of the mutated individual is not higher than the highest affinity value of the clone population.
10. The cross-organization multi-source heterogeneous business process event log fusion system of claim 6, wherein: the iteration generation fusion event log module specifically executes the following operations:
adding new individuals into the random initial population, namely, randomly selecting some individuals from the rest individuals in the initial population again to add into the random initial population; repeating the random initial population obtaining and cloning module, the mutation selecting module and the fusion event log iterative generation module until all individuals in the initial population are randomly extracted; sequencing all events in the individual containing two tracks in the population obtained after the iteration is finished according to the time stamps to obtain a fusion event log with two event logs in the same structure; and then randomly extracting an event log from the event logs of the rest isomorphic business process in the event log acquisition and isomorphic modules and fusing the event logs, repeating the initial population acquisition module, the random initial population acquisition and cloning module, the mutation selection module and the fusion event log generation module until the event logs of the isomorphic business process to be fused are empty, and finally obtaining a fused cross-organization multi-source isomorphic business process event log.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210897941.4A CN115292258A (en) | 2022-07-28 | 2022-07-28 | Cross-organization multi-source heterogeneous business process event log fusion method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210897941.4A CN115292258A (en) | 2022-07-28 | 2022-07-28 | Cross-organization multi-source heterogeneous business process event log fusion method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115292258A true CN115292258A (en) | 2022-11-04 |
Family
ID=83823907
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210897941.4A Pending CN115292258A (en) | 2022-07-28 | 2022-07-28 | Cross-organization multi-source heterogeneous business process event log fusion method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115292258A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117035380A (en) * | 2023-07-11 | 2023-11-10 | 山东理工大学 | Cross-organization business process consistency detection and abnormal behavior diagnosis method and system |
-
2022
- 2022-07-28 CN CN202210897941.4A patent/CN115292258A/en active Pending
Non-Patent Citations (2)
| Title |
|---|
| CLAES, JAN 等: "Integrating Computer Log Files for Process Mining: A Genetic Algorithm Inspired Technique", 《CAISE 2011: ADVANCED INFORMATION SYSTEMS ENGINEERING WORKSHOPS 》, vol. 83, pages 282 - 293 * |
| 徐杨;袁峰;林琪;汤德佑;李东;: "基于混合人工免疫算法的流程挖掘事件日志融合方法", 软件学报, no. 02, pages 396 - 416 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117035380A (en) * | 2023-07-11 | 2023-11-10 | 山东理工大学 | Cross-organization business process consistency detection and abnormal behavior diagnosis method and system |
| CN117035380B (en) * | 2023-07-11 | 2024-04-16 | 山东理工大学 | Cross-organization business process consistency detection and abnormal behavior diagnosis method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7356479B2 (en) | Device and method for accommodating business process | |
| CN103812872B (en) | A kind of network navy behavioral value method and system based on mixing Di Li Cray process | |
| CN105023196A (en) | Analysis method and device for charging transaction data of charging stations | |
| CN104484582B (en) | The biological information project automatic analysis method and system realized by modularization selection | |
| CN109409561B (en) | Construction method of multi-time scale time sequence collaborative prediction model | |
| CN115292258A (en) | Cross-organization multi-source heterogeneous business process event log fusion method and system | |
| CN112147967A (en) | Automatic generation method and device for scene test case of automobile controller | |
| CN113488106A (en) | Method for rapidly acquiring comparison result data of target genome region | |
| CN108108883B (en) | Clustering algorithm-based vehicle scheduling network elastic simplification method | |
| CN113688490A (en) | Network co-construction sharing processing method, device, equipment and storage medium | |
| CN111931022A (en) | AI hot spot content intelligent editing system | |
| EP2256677A1 (en) | Workflow processing program, method, and device | |
| CN106056299A (en) | Decision support method and device based on process mining | |
| CN112052232B (en) | Business process context extraction method based on replay technology | |
| CN113344708A (en) | Large-scale system production and exercise application evaluation method and system | |
| CN112835960A (en) | Data analysis method and system for digital exhibition | |
| CN110597796B (en) | Big data real-time modeling method and system based on full life cycle | |
| CN109669777B (en) | Industrial internet big data element demand service providing method and system | |
| CN111444106A (en) | Analysis method and system for software testable requirements | |
| Fu et al. | Charting the landscape of enterprise architecture complexity cybernetics: A systematic literature analysis | |
| CN112860899B (en) | Label generation method and device, computer equipment and computer readable storage medium | |
| CN111428756B (en) | Planning data fusion real-time state method and device based on time sequence information entropy | |
| CN114926027A (en) | Distributed business process compliance checking method and system based on model decomposition | |
| CN112181952A (en) | Method, system, device and storage medium for constructing data model | |
| CN110955754A (en) | Model construction method for repeated incoming call analysis and recognition |
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 |