WO2025058122A1 - System and method for generating multi-perspective event log - Google Patents

Abstract

The present invention relates to a system and a method for generating a multi-perspective event log, and may comprise: receiving a data source; extracting metadata from the data source; generating an event map and a sample event log by using the metadata; calculating the quality of the sample event log; and, if the quality value of the sample event log is greater than or equal to a reference value, generating, on the basis of the event map, a multi-perspective event log by collecting data from the data source.

Description

System and method for generating multi-view event logs

The following embodiments relate to systems and methods for generating logs with multi-view events.

Process mining analysis uses as input an event log, which consists of a series of events that can be grouped in a specific way to represent different aspects of a business process.

In practice, event logs are not always readily available in most real-world situations. In most cases, event logs must be extracted from a data source (e.g., an information system or database) and converted into a specific structure and format that process mining can interpret.

The process of extracting and identifying (e.g., cases, activities, timestamps) events requires manual effort and domain knowledge about the process, making it a major challenge in generating event logs, which is time-consuming and costly.

Another challenge in generating event logs is dealing with process perspectives. Event logs are usually bound to specific process analysis objectives, i.e., business perspectives and perspectives themselves can be correlated with entities in the data.

Multiple views can be extracted from the database, which means that multiple event logs and process models can be generated for the same process data.

In particular, several approaches have been proposed to automatically extract event logs from databases, such as rdb2log or OpenSLEX. However, these approaches require partial or full scans of the database, which results in long computation times and limits the sources to relational databases.

The present invention aims to provide a system and method for generating a multi-view event log.

A method for generating a multi-view event log according to one embodiment of the present invention may include the steps of: receiving a data source; extracting metadata from the data source; generating an event map and a sample event log using the metadata; calculating a quality of the sample event log; and, if a quality value of the sample event log is equal to or higher than a reference value, collecting data from the data source based on the event map to generate a multi-view event log.

At this time, the step of generating the event map and the sample event log using the metadata may include the steps of: converting the extracted metadata into a standardized entity relationship data model; generating a data catalog by connecting the entity relationship data models; inferring an activity concept and a time stamp concept from the data catalog; inferring a case concept from the data catalog; generating the event map by associating the activity concept, the time stamp concept, and the case concept; and extracting sample data from the data source and extracting data from the sample data based on the event map to generate the sample event log.

At this time, the step of creating a data catalog by connecting the entity relationship data models may include the step of confirming the similarity of field names included in the entity relationship data models; the step of connecting the entity relationship data models so that the data catalog becomes acyclic by using the confirmed similarity; and the step of removing optional fields that are not essential from each of the entity relationship data models included in the data catalog.

At this time, the step of inferring the activity concept and the time stamp concept from the data catalog comprises: a step of analyzing the data type of each field of the entity relationship data models included in the data catalog; a step of analyzing each field name of the entity relationship data models using dictionary-based pattern matching; a step of inferring the activity concept and the time stamp concept through date or time analysis among the fields included in the entity relationship data models; a step of removing the remaining time stamp fields except for the representative time stamp field from each of the entity relationship data models; a step of removing duplicate column names in the entity relationship data models included in the data catalog; a step of calculating the centrality of each of the entity relationship data models based on the number of entity relationship data models that depend on each of the entity relationship data models included in the data catalog; a step of removing an entity relationship data model that has no time stamp concept candidate and no entity relationship data model that depends on the entity relationship data model among the entity relationship data models included in the data catalog; The method may further include: removing fields that are not primary keys or foreign keys and fields that do not correspond to timestamps so that each of the entity-relationship data models included in the data catalog has at most one category field and one numeric field; inferring a timestamp concept and an activity life cycle through text similarity analysis of a timestamp field for each of the entity-relationship data models included in the data catalog; and removing a date and time pattern from the timestamp concept and inferring an activity concept by connecting the fields that are not primary keys or foreign keys and the fields that do not correspond to timestamps.

At this time, the step of inferring the case concept from the data catalog can generate the case concept by inferring the number of times each field of the entity relationship data models included in the data catalog is cited and the ranking of the number of times cited.

At this time, the method for generating a multi-view event log may further include a step of generating an event map based on data quality if at least one of the activity concept, the timestamp concept and the case concept is not found; and a step of generating a multi-view event log from the data source based on the event map generated based on the data quality.

At this time, the step of calculating the quality of the sample event log can be calculated by considering at least one of the case identifier ratio, the trace mutation ratio, the average ratio of unique activities per case, the endpoint activity ratio, and the start-end activity ratio of the sample event log.

At this time, the method for generating a multi-viewpoint event log may further include a step of generating an event map based on data quality if the quality value of the sample event log is less than a reference value; and a step of generating a multi-viewpoint event log from the data source based on the event map generated based on the data quality.

At this time, the step of generating an event map based on the data quality may include: a step of calculating a data quality index from the data source; a step of calculating a data quality dimension and an average value from the data source; a step of inferring an event concept candidate using a classification algorithm; a step of generating a plurality of candidate event maps based on the event concept candidates; a step of extracting sample data from the data source and generating a sample candidate event log based on each of the plurality of candidate event maps; a step of calculating a quality for each of the sample candidate event logs; and a step of determining the event map from among the plurality of candidate event maps based on a quality of the sample candidate event log.

At this time, the step of inferring the event concept candidate using the classification algorithm may include the step of predicting the event concept for each field of the entity relationship data models extracted from the data source as the event concept candidate using the classification model generated by using the previously learned event log as learning data as input to the classification algorithm; the step of removing fields that are not primary keys or foreign keys and are not event concept fields from the entity relationship data models; the step of setting the average value of the data quality indicators for all the event concept fields included in the entity relationship data models as a threshold value and removing the event concept field whose average value of the data quality indicator of the event concept field is lower than the threshold value; the step of modifying fields with the same field name included in the entity relationship data models but whose event concept candidates of the event concept field are different to have the same event concept candidate; and the step of removing fields with the same field name included in the entity relationship data models except for the field with the highest average value of the data quality indicator.

At this time, the step of determining the event map from among the plurality of candidate event maps based on the quality of the sample candidate event log may determine the event map as the candidate event map corresponding to the sample candidate event log having the highest quality.

At this time, the step of determining the event map from among the plurality of candidate event maps based on the quality of the sample candidate event logs may include providing the user with quality information of the sample candidate event logs corresponding to the plurality of candidate event maps, and determining the event map based on the candidate event map selected by the user.

A system for generating a multi-view event log according to one embodiment of the present invention comprises: a memory for storing a data source; and a processor, wherein the processor extracts metadata from the data source, generates an event map and a sample event log using the metadata, calculates a quality of the sample event log, and if the quality value of the sample event log is equal to or higher than a reference value, collects data from the data source based on the event map to generate a multi-view event log.

At this time, the processor may convert the extracted metadata into a standardized entity relationship data model when generating the event map and the sample event log using the metadata, generate a data catalog by connecting the entity relationship data models, infer an activity concept and a time stamp concept from the data catalog, infer a case concept from the data catalog, generate the event map by associating the activity concept, the time stamp concept, and the case concept, extract sample data from the data source, and extract data from the sample data based on the event map to generate the sample event log.

At this time, the processor can generate an event map based on data quality if at least one of the activity concept, the timestamp concept and the case concept is not found, and can generate a multi-view event log from the data source based on the event map generated based on the data quality.

At this time, the processor may calculate the quality of the sample event log by considering at least one of the case identifier ratio, the trace mutation ratio, the average ratio of unique activities per case, the endpoint activity ratio, and the start-end activity ratio of the sample event log.

At this time, the processor can generate an event map based on data quality if the quality value of the sample event log is less than a reference value, and generate a multi-viewpoint event log from the data source based on the event map generated based on the data quality.

The present invention relates to a system and method for generating a multi-perspective event log, which receives a data source, extracts metadata from the data source, generates an event map and a sample event log using the metadata, calculates the quality of the sample event log, and generates a multi-perspective event log by collecting data from the data source based on the event map if the quality value of the sample event log is higher than a reference value. Since various case concepts are automatically inferred, the multi-perspective event log can be extracted without forcing the user to specify or infer case concepts.

FIG. 1 is a diagram schematically illustrating a configuration of a system for generating a multi-view event log according to an embodiment of the present invention.

FIG. 2 is a flowchart schematically illustrating a process for generating a multi-view event log in a system according to one embodiment of the present invention.

FIG. 3 is a flowchart illustrating a process of generating an event map and a sample event log using metadata in a system according to an embodiment of the present invention.

FIG. 4 is a flowchart illustrating a process of creating a data catalog in a system according to one embodiment of the present invention.

FIG. 5 is a flowchart illustrating a process of inferring an activity concept and a time stamp concept from a data catalog in a system according to one embodiment of the present invention.

FIG. 6 is a flowchart illustrating a process of generating an event map based on data quality in a system according to an embodiment of the present invention.

FIG. 7 is a flowchart illustrating a process of inferring event opening candidates using a classification algorithm in a system according to one embodiment of the present invention.

Hereinafter, embodiments will be described in detail with reference to the attached drawings. However, since various modifications may be made to the embodiments, the scope of rights of the patent application is not limited or restricted by these embodiments. It should be understood that all modifications, equivalents, or substitutes to the embodiments are included in the scope of rights.

The terms used in the examples are for the purpose of description only and should not be construed as limiting. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this specification, the terms "comprises" or "has" and the like are intended to specify the presence of a feature, number, step, operation, component, part or combination thereof described in the specification, but should be understood to not exclude in advance the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts or combinations thereof.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the embodiments belong. Terms defined in commonly used dictionaries, such as those defined in common dictionaries, should be interpreted as having a meaning consistent with the meaning they have in the context of the relevant art, and shall not be interpreted in an idealized or overly formal sense, unless expressly defined in this application.

In addition, when describing with reference to the attached drawings, the same components will be given the same reference numerals regardless of the drawing numbers, and redundant descriptions thereof will be omitted. When describing an embodiment, if it is determined that a specific description of a related known technology may unnecessarily obscure the gist of the embodiment, the detailed description thereof will be omitted.

Also, in describing components of the embodiments, terms such as first, second, A, B, (a), (b), etc. may be used. These terms are only intended to distinguish the components from other components, and the nature, order, or sequence of the components are not limited by the terms. When it is described that a component is "connected," "coupled," or "connected" to another component, it should be understood that the component may be directly connected or connected to the other component, but another component may also be "connected," "coupled," or "connected" between each component.

Components included in one embodiment and components that have common functions will be described using the same names in other embodiments. Unless otherwise stated, descriptions made in one embodiment may be applied to other embodiments, and specific descriptions will be omitted to the extent of overlap.

Hereinafter, a system and method for generating a multi-view event log according to an embodiment of the present invention will be described in detail with reference to the attached FIGS. 1 to 7.

FIG. 1 is a diagram schematically illustrating a configuration of a system for generating a multi-view event log according to an embodiment of the present invention.

Referring to FIG. 1, a system (100) for generating a multi-view event log may be configured to include a processor (110), a communication unit (120), and a memory (130).

The communication unit (120) is a communication interface device including a receiver and a transmitter, and can transmit and receive data wired or wirelessly. The communication unit (120) can receive a data source by connecting to an external database server, etc.

The memory (130) stores an operating system, application programs and storage data for controlling the overall operation of the system (100), and can also store data sources, metadata, entity relationship data models, data catalogs, event maps, sample event logs and multi-view event logs according to the present invention.

The processor (110) may be configured to include a metadata extraction unit (111), a first event map generation unit (112), a quality calculation unit (113), a second event map generation unit (114), and a multi-viewpoint event log generation unit (115).

The metadata extraction unit (111) can receive a data source from the memory (130) and extract metadata from the data source.

The first event map generation unit (112) can generate an event map and a sample event log using metadata.

More specifically, the first event map generation unit (112) can 1) convert extracted metadata into a standardized entity relationship data model, 2) generate a data catalog by connecting entity relationship data models, 3) infer an activity notion and a timestamp notion from the data catalog, infer a case notion from the data catalog, 4) generate an event map by associating the activity concept, the timestamp concept, and the case concept, and 5) extract sample data from a data source and extract data from the sample data based on the event map to generate a sample event log.

At this time, if at least one of the activity concept, the time stamp concept, and the case concept is not found, the first event map generation unit (112) can request the second event map generation unit (114) to generate an event map based on data quality.

The first event map generation unit (112) can generate a data catalog by connecting entity relationship data models through the following process. The first event map generation unit (112) can generate a data catalog by 1) checking the similarity of field names included in entity relationship data models, 2) using the checked similarity to connect entity relationship data models so that the data catalog is acyclic, and 3) removing optional fields that are not essential from each of the entity relationship data models included in the data catalog.

The first event map generation unit (112) can infer the activity concept and the timestamp concept from the data catalog through the following process. The first event map generation unit (112) 1) analyzes the data type of each field for the entity relationship data models included in the data catalog, 2) analyzes each field name of the entity relationship data models using dictionary-based pattern matching, 3) infers the activity concept and the timestamp concept through date or time analysis among the fields included in the entity relationship data models, 4) removes the remaining timestamp fields except for the representative timestamp field from each of the entity relationship data models, 5) removes duplicate column names in the entity relationship data models included in the data catalog, 6) calculates the centrality of each entity relationship data model based on the number of entity relationship data models that depend on each entity relationship data model included in the data catalog, 7) removes entity relationship data models that do not have a timestamp concept candidate and do not have an entity relationship data model that depends on the entity relationship data model included in the data catalog, and 8) generates a centrality map for each entity relationship data model included in the data catalog. 9) For each entity-relationship data model included in the data catalog, the timestamp concept and activity life cycle can be inferred through text similarity analysis of the timestamp field, and 10) the date and time pattern can be removed from the timestamp concept and the activity concept can be inferred by linking the fields that are not primary or foreign keys and timestamps so that there is at most one category field and one numeric field.

When inferring a case concept from a data catalog, the first event map generation unit (112) can generate a case concept by inferring the number of times each field of entity relationship data models included in the data catalog is cited and the ranking of the number of times cited.

The quality calculation unit (113) can calculate the quality of a sample event log.

More specifically, the quality calculation unit (113) can calculate the quality of the sample event log by considering at least one of the case identifier ratio, the trace mutation ratio, the average ratio of unique activities per case, the endpoint activity ratio, and the start-end activity ratio of the sample event log.

The second event map generation unit (114) can generate an event map based on data quality if the quality value of the sample event log is lower than a reference value. More specifically, the second event map generation unit (114) can 1) calculate a data quality index from a data source, 2) calculate a data quality dimension and an average value from the data source, 3) infer event concept candidates using a classification algorithm, 4) generate a plurality of candidate event maps based on the event concept candidates, 5) extract sample data from the data source and generate sample candidate event logs based on each of the plurality of candidate event maps, 6) calculate the quality for each of the sample candidate event logs, and 7) determine an event map from among the plurality of candidate event maps based on the quality of the sample candidate event logs.

When the second event map generation unit (114) determines an event map from among a plurality of candidate event maps based on the quality of the sample candidate event logs, if set to fully automatic, it can determine the candidate event map corresponding to the sample candidate event log with the highest quality as the event map.

The second event map generation unit (114) determines an event map from among a plurality of candidate event maps based on the quality of the sample candidate event logs. If manually set, the second event map generation unit (114) provides the user with quality information of the sample candidate event logs corresponding to the plurality of candidate event maps, and determines the event map based on the candidate event map selected by the user.

The second event map generation unit (114) can infer event concept candidates by the following process when inferring them using a classification algorithm. The second event map generation unit (114) 1) predicts event concepts for each field of entity relationship data models extracted from a data source as event concept candidates by using a classification model generated by using pre-learned event logs as learning data as input to a classification algorithm, 2) removes fields that are not primary keys or foreign keys and are not event concept fields from the entity relationship data models, 3) sets the average value of data quality indicators for all event concept fields included in the entity relationship data models as a threshold value, and removes event concept fields whose average value of data quality indicators of event concept fields is lower than the threshold value, 4) modifies fields with the same field names included in the entity relationship data models but whose event concept candidates are different to have the same event concept candidates, and 5) removes fields with the same field names included in the entity relationship data models except for the field with the highest average value of data quality indicators, thereby inferring event concept candidates.

In addition, if the second event map generation unit (114) is requested to generate an event map because at least one of the activity concept, the time stamp concept, and the case concept is not found by the first event map generation unit (112), 1) an event map can be generated based on data quality, and 2) a multi-viewpoint event log can be generated from a data source based on the event map generated based on data quality.

The multi-viewpoint event log generation unit (115) can generate a multi-viewpoint event log by collecting data from a data source based on an event map if the quality value of the sample event log calculated by the quality calculation unit (113) is higher than a reference value.

The multi-viewpoint event log generation unit (115) can generate a multi-viewpoint event log from a data source based on the event map generated based on data quality by generating an event map generated based on data quality in the second event map generation unit (114).

In Fig. 1, the metadata extraction unit (111), the first event map generation unit (112), the quality calculation unit (113), the second event map generation unit (114), and the multi-viewpoint event log generation unit (115) are configured to be included in the processor (110), but are not limited thereto and may be configured as separate devices. In addition, they may be implemented in the form of program commands that can be executed through various computer means.

Hereinafter, the method according to the present invention configured as described above will be described with reference to the drawings below.

FIG. 2 is a flowchart schematically illustrating a process for generating a multi-view event log in a system according to one embodiment of the present invention.

Referring to FIG. 2, the system (100) can receive a data source (210).

And, the system (100) can extract metadata from a data source (220).

At this time, metadata can be composed of key information indicating whether it is a primary key or a foreign key, a field name indicating what the field is, type information distinguishing whether the field is a case type or a timestamp type, and required information indicating whether the field is required or optional.

And, the system (100) can generate an event map and a sample event log using metadata (230). A detailed description of step 230 will be provided later with reference to FIG. 3.

Additionally, the system (100) can calculate the quality of the sample event log (240).

At step 240, the system (100) can calculate by considering at least one of a case identifier ratio of the sample event log, a trace mutation ratio, an average ratio of unique activities per case, an endpoint activity ratio, and a start-to-finish activity ratio.

More specifically, at step 240, the system (100) can calculate the quality of the event log by referring to <Mathematical Formula 1> and <Mathematical Formula 2> below.

[Mathematical formula 1]

Here,

는 이벤트 로그(

)의 흥미도 이고,

is the event log (

) is also of interest,

는 이벤트 로그(

)의 사례 식별자 비율이고,

is the event log (

) is the case identifier ratio,

는 이벤트 로그(

)의 고유 사례 식별자의 총 수이고,

is the event log (

) is the total number of unique case identifiers,

는 이벤트 로그(

)의 총 이벤트 수이고,

is the event log (

) is the total number of events,

는 이벤트 로그(

)의 추적 변형 비율이고,

is the event log (

) is the tracking deformation ratio,

는 이벤트 로그(

)의 총 추적 변형 수이고,

is the event log (

) is the total number of tracking variations,

는 케이스당 고유 활동의 평균 비율이고,

is the average proportion of unique activities per case,

는 이벤트 로그(

)의 케이스(

)의 총 고유 활동 수 이고,

is the event log (

) case(

) is the total number of unique activities,

는 이벤트 로그(

)의 케이스(

)의 전체 이벤트이고,

is the event log (

) case(

) is the entire event,

는 엔드 포인트 활동 비율이고,

is the endpoint activity rate,

는 이벤트 로그(

)의 고유한 시작 및 종료 활동의 총 수 이고,

is the event log (

) is the total number of unique start and end activities,

는 시작 종료 활동 비율이고,

is the start-end activity ratio,

는 이벤트 로그(

)의 고유한 시작 및 종료 활동의 총 수를 나타냅니다.

is the event log (

) represents the total number of unique start and end activities.

는 이벤트 로그(

)에서 시작 및 종료 활동의 공통 요소의 총 수를 나타냅니다.

is the event log (

) represents the total number of common elements of start and end activities.

는 사례 식별자 비율에 대한 0에서 1 사이의 값을 가지는 가중치이고,

is a weight with a value between 0 and 1 for the case identifier ratio,

는 추적 변형 비율에 대한 0에서 1 사이의 값을 가지는 가중치이고,

is a weight with a value between 0 and 1 for the tracking deformation ratio,

는 케이스당 고유 활동의 평균 비율에 대한 0에서 1 사이의 값을 가지는 가중치이고,

is a weight with a value between 0 and 1 for the average proportion of unique activities per case,

는 엔드 포인트 활동 비율에 대한 0에서 1 사이의 값을 가지는 가중치이고,

is a weight with a value between 0 and 1 for the endpoint activity ratio,

는 시작 종료 활동 비율에 대한 0에서 1 사이의 값을 가지는 가중치이다.

is a weight with a value between 0 and 1 for the start-to-end activity ratio.

[Mathematical formula 2]

는 다중 관점 이벤트 로그 관심도로 이벤트 맵 결과

에서 생성되는 각 이벤트 로그 관점

에 대한 이벤트 로그 관심도의 평균으로, 샘플 이벤트 로그의 품질에 해당하고,Here,

is a multi-perspective event log with an event map result as an interest map.

Each event log perspective generated from

The average of the event log interest in the sample event log, corresponding to the quality of the sample event log,

는 <수학식 1>에서 각 이벤트 로그의 관점

의 이벤트 로그 흥미도를 나타내고,

is the viewpoint of each event log in <Mathematical Formula 1>

Indicates the interest of the event log,

은 이벤트 로그 관점의 총 수 이다.

is the total number of event log views.

의 관계를 가질 수 있다.At this time, the relationship between the event map and the generated event log perspective is that exactly one event log is generated for each event map perspective.

can have a relationship.

And, the system (100) can check whether the quality value of the sample event log is higher than the reference value (250).

If the quality value of the sample event log as a result of the verification in step 250 is higher than the reference value, the system (100) can collect data from the data source based on the event map to generate a multi-view event log (270).

If the quality value of the sample event log as a result of the verification in step 250 is less than the reference value, the system (100) can generate an event map based on the data quality (260). A detailed description of step 260 will be provided later with reference to FIG. 6.

And, the system (100) can generate a multi-view event log from a data source based on an event map generated based on data quality in step 270.

FIG. 3 is a flowchart illustrating a process of generating an event map and a sample event log using metadata in a system according to an embodiment of the present invention.

Referring to FIG. 3, the system (100) can convert extracted metadata into a standardized entity relationship data model (310).

And, the system (100) can create a data catalog by connecting entity relationship data models (320). A specific description of step 320 will be explained later through FIG. 4.

And, the system (100) can infer the activity concept and the time stamp concept from the data catalog (330). A specific description of step 330 will be explained later through FIG. 5.

Additionally, the system (100) can infer case concepts from the data catalog (340).

In step 340, the system (100) can generate a case concept by inferring the number of times each field of the entity relationship data models included in the data catalog is cited and the rank of the number of times cited. The case concept can include entity relationship data model information, a field name, information on the number of times the field is cited, and information on the rank of the number of times cited.

And, the system (100) can check whether it has discovered all of the activity concept, the time stamp concept, and the case concept (350).

If, as a result of the verification in step 350, none of the activity concept, the time stamp concept, and the case concept are found, that is, if there is at least one concept that is not found among the activity concept, the time stamp concept, and the case concept, the system (100) can proceed to step 260 of FIG. 2 and perform step 260.

If all of the activity concepts, timestamp concepts, and case concepts are found as a result of the verification in step 350, the system (100) can generate an event map by associating the activity concepts, timestamp concepts, and case concepts (360). That is, in step 360, the system (100) can generate an event map to include all of the activity concepts, timestamp concepts, and case concepts.

And, the system (100) can extract sample data from a data source and extract data from the sample data based on an event map to generate a sample event log (370).

And, the system (100) can generate an event map based on data quality if at least one of the activity concept, the time stamp concept, and the case concept is not found (380).

And, the system (100) can generate a multi-view event log from a data source based on an event map generated based on data quality (390).

FIG. 4 is a flowchart illustrating a process of creating a data catalog in a system according to an embodiment of the present invention.

Referring to FIG. 4, the system (100) can check the similarity of field names included in entity relationship data models (410).

Additionally, the system (100) can connect entity relationship data models to make the data catalog acyclic by utilizing the verified similarity (420).

Additionally, the system (100) can remove optional fields that are not essential from each of the entity relationship data models included in the data catalog (430).

FIG. 5 is a flowchart illustrating a process of inferring an activity concept and a time stamp concept from a data catalog in a system according to one embodiment of the present invention.

Referring to FIG. 5, the system (100) can analyze the data type of each field for entity relationship data models included in the data catalog (510).

Additionally, the system (100) can analyze each field name of entity relationship data models using dictionary-based pattern matching (512).

And, the system (100) can infer the activity concept and the time stamp concept through date or time analysis among the fields included in the entity relationship data models (514). At this time, the time stamp concept can include entity relationship data model information, field name information, and information about the life cycle of the corresponding field.

And, the system (100) can remove all timestamp fields except the representative timestamp field from each of the entity relationship data models (516).

Additionally, the system (100) can remove duplicate column names in entity relational data models included in the data catalog (518).

And, the system (100) can calculate the centrality of each entity relationship data model included in the data catalog based on the number of entity relationship data models that depend on each entity relationship data model (520).

And, the system (100) can remove an entity relationship data model that does not have a time stamp concept candidate among the entity relationship data models included in the data catalog and does not have an entity relationship data model that depends on the entity relationship data model (522).

And, the system (100) can remove fields that are not primary keys or foreign keys and fields that do not correspond to timestamps so that each of the entity relational data models included in the data catalog has at most one category field and one numeric field (524).

And, the system (100) can infer the time stamp concept and activity life cycle through text similarity analysis of the time stamp field for each entity relationship data model included in the data catalog (526).

And, the system (100) can remove the date and time pattern from the timestamp concept and infer the activity concept by connecting fields that are not primary keys or foreign keys and fields that do not correspond to timestamps (528).

FIG. 6 is a flowchart illustrating a process of generating an event map based on data quality in a system according to an embodiment of the present invention.

Referring to FIG. 6, the system (100) can calculate a data quality index from a data source (610).

And, the system (100) can calculate data quality dimensions and average values from the data source (620).

At this time, the system (100) can calculate steps 610 and 620 by referring to Andrews, Robert, et al. "Quality-informed semi-automated event log generation for process mining." Decision Support Systems 132 (2020): 113265.

And, the system (100) can infer event concept candidates using a classification algorithm (630). A specific description of step 630 will be described later through FIG. 7.

And, the system (100) can generate a plurality of candidate event maps based on the event concept candidates (640).

Additionally, the system (100) can extract sample data from a data source and generate sample candidate event logs based on each of a plurality of candidate event maps (650).

Additionally, the system (100) can calculate the quality for each sample candidate event log (660).

And, the system (100) can determine an event map among a plurality of candidate event maps based on the quality of the sample candidate event log (670).

In step 670, the system (100) determines an event map corresponding to a sample candidate event log having the highest quality, or provides the user with quality information of sample candidate event logs corresponding to a plurality of candidate event maps, and determines an event map corresponding to a candidate event map selected by the user.

FIG. 7 is a flowchart illustrating a process of inferring event opening candidates using a classification algorithm in a system according to one embodiment of the present invention.

Referring to FIG. 7, the system (100) can predict event concepts for each field of entity relationship data models extracted from a data source as event concept candidates by using a classification model generated by using pre-learned event logs as learning data and as input to a classification algorithm (710).

Additionally, the system (100) can remove fields that are not primary or foreign keys and are not event concept fields from entity relational data models (720).

And, the system (100) can set the average value of the data quality indicators for all event concept fields included in the entity relationship data models as a threshold value, and remove event concept fields whose average value of the data quality indicators of the event concept fields is lower than the threshold value (730).

And, the system (100) can modify fields that have the same field name but different event concept candidates in the event concept fields included in entity relationship data models to have the same event concept candidates (740).

And, the system (100) can remove fields with the same field name included in entity relationship data models except for the field with the highest average value of data quality indicator (750).

The devices described above may be implemented as hardware components, software components, and/or a combination of hardware components and software components. For example, the devices and components described in the embodiments may be implemented using one or more general-purpose computers or special-purpose computers, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing instructions and responding to them. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. In addition, the processing device may access, store, manipulate, process, and generate data in response to the execution of the software. For ease of understanding, the processing device is sometimes described as being used alone, but those skilled in the art will appreciate that the processing device may include multiple processing elements and/or multiple types of processing elements. For example, the processing unit may include multiple processors, or a processor and a controller. Other processing configurations, such as parallel processors, are also possible.

The software may include a computer program, code, instructions, or a combination of one or more of these, which may configure a processing device to perform a desired operation or may, independently or collectively, command the processing device. The software and/or data may be permanently or temporarily embodied in any type of machine, component, physical device, virtual equipment, computer storage medium or device, or transmitted signal waves, for interpretation by the processing device or for providing instructions or data to the processing device. The software may also be distributed over network-connected computer systems, and stored or executed in a distributed manner. The software and data may be stored on one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program commands that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program commands, data files, data structures, etc., alone or in combination. The program commands recorded on the medium may be those specially designed and configured for the embodiment or may be those known to and available to those skilled in the art of computer software. Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and hardware devices specially configured to store and execute program commands such as ROMs, RAMs, flash memories, etc. Examples of the program commands include not only machine language codes generated by a compiler but also high-level language codes that can be executed by a computer using an interpreter, etc. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiment, and vice versa.

Although the embodiments have been described above by way of limited examples and drawings, those skilled in the art may make various modifications and variations from the above description. For example, appropriate results may be achieved even if the described techniques are performed in a different order than the described method, and/or components of the described system, structure, device, circuit, etc. are combined or combined in a different form than the described method, or are replaced or substituted by other components or equivalents.

Therefore, other implementations, other embodiments, and equivalents to the claims are also included in the scope of the claims described below.

Claims (18)

Step of receiving a data source; A step of extracting metadata from the above data source; A step of generating an event map and a sample event log using the above metadata; a step of calculating the quality of the above sample event log; and If the quality value of the above sample event log is greater than or equal to the reference value, a step of collecting data from the data source based on the above event map to generate a multi-view event log A method of generating a multi-perspective event log that includes: In the first paragraph, The step of generating the event map and the sample event log using the above metadata is: A step of converting the extracted metadata into a standardized entity relationship data model; A step of creating a data catalog by connecting the above entity relationship data models; A step of inferring the activity concept and the timestamp concept from the above data catalog; A step of inferring case concepts from the above data catalog; A step of generating the event map by associating the above activity concept, the above time stamp concept and the above case concept; and A step of extracting sample data from the above data source and extracting data from the sample data based on the above event map to generate the sample event log. A method of generating a multi-perspective event log that includes: In the second paragraph, The steps for creating a data catalog by connecting the above entity relationship data models are: A step of verifying the similarity of field names included in the above entity relationship data models; A step of connecting the entity relationship data models so that the data catalog becomes acyclic by utilizing the verified similarity; and A step of removing optional fields that are not essential from each of the entity relationship data models included in the above data catalog. A method of generating a multi-perspective event log that includes: In the second paragraph, The step of inferring the activity concept and the timestamp concept from the above data catalog is: A step of analyzing the data type of each field for the entity relationship data models included in the above data catalog; A step of analyzing each field name of the above entity relationship data models using dictionary-based pattern matching; A step of inferring the activity concept and the timestamp concept through date or time analysis among the fields included in the above entity relationship data models; A step of removing all timestamp fields except the representative timestamp field from each of the above entity relationship data models; A step of removing duplicate column names in the entity relationship data models included in the above data catalog; A step of calculating the centrality of each of the entity relationship data models included in the data catalog based on the number of entity relationship data models that depend on each of the entity relationship data models; A step of removing an entity-relationship data model that does not have a timestamp concept candidate among the entity-relationship data models included in the data catalog and does not have an entity-relationship data model that depends on the entity-relationship data model; Steps to remove fields that are not primary or foreign keys and fields that are not timestamps so that each of the entity relationship data models included in the above data catalog has at most one category field and one numeric field: A step of inferring the time stamp concept and activity life cycle through text similarity analysis of the time stamp field for each of the entity relationship data models included in the data catalog; and A step of removing the date and time pattern from the above timestamp concept and inferring the activity concept by connecting the fields that are not the primary key or the foreign key and the fields that do not correspond to the timestamp. A method of generating a multi-perspective event log that includes: In the second paragraph, The step of inferring the case concept from the above data catalog is: Generate the case concept by inferring the number of times each field of the entity relationship data models included in the data catalog is cited and the rank of the number of times cited. How to generate a multi-perspective event log. In the second paragraph, If at least one of the above activity concept, the above timestamp concept and the above case concept is not found, a step of generating an event map based on data quality; and A step of generating a multi-view event log from the data source based on an event map generated based on the above data quality. A method of generating a multi-perspective event log that includes more than one viewpoint. In the first paragraph, The steps for calculating the quality of the above sample event log are: Calculated by considering at least one of the case identifier ratio, trace mutation ratio, average ratio of unique activities per case, endpoint activity ratio, and start-end activity ratio of the above sample event log. How to generate a multi-perspective event log. In the first paragraph, If the quality value of the above sample event log is below the reference value, a step of generating an event map based on data quality; and A step of generating a multi-view event log from the data source based on an event map generated based on the above data quality. A method of generating a multi-perspective event log that includes more than one viewpoint. In Article 8, The steps for generating an event map based on the above data quality are: A step of calculating data quality indicators from the above data source; A step of calculating data quality dimensions and average values from the above data sources; A step of inferring event concept candidates using a classification algorithm; A step of generating a plurality of candidate event maps based on the above event concept candidates; A step of extracting sample data from the above data source and generating a sample candidate event log based on each of the plurality of candidate event maps; A step of calculating the quality for each of the above sample candidate event logs; and A step of determining the event map among the plurality of candidate event maps based on the quality of the sample candidate event logs. A method of generating a multi-perspective event log that includes: In Article 9, The step of inferring the event concept candidates using the above classification algorithm is: A step of predicting an event concept for each field of entity relationship data models extracted from the data source as an event concept candidate using a classification model generated by using pre-learned event logs as learning data and inputting the classification algorithm; A step of removing fields that are neither primary key nor foreign key and are not event concept fields from the above entity-relationship data models; A step of setting an average value of data quality indicators for all event concept fields included in the entity relationship data models as a threshold value, and removing an event concept field whose average value of data quality indicators is lower than the threshold value; A step for modifying fields whose field names are the same but whose event concept candidates are different in the event concept fields included in the above entity relationship data models to have the same event concept candidates; and Step of removing fields with the same field name included in the above entity relationship data models, except for the field with the highest average value of data quality indicator. A method of generating a multi-perspective event log that includes: In Article 9, The step of determining the event map among the plurality of candidate event maps based on the quality of the sample candidate event logs is: Determine the event map corresponding to the candidate event log of the sample candidate with the highest quality. How to generate a multi-perspective event log. In Article 9, The step of determining the event map among the plurality of candidate event maps based on the quality of the sample candidate event logs is: Providing the user with quality information of the sample candidate event log corresponding to the above plurality of candidate event maps, and determining the event map based on the candidate event map selected by the user. How to generate a multi-perspective event log. A computer-readable recording medium characterized by having recorded thereon a program for executing the method of any one of claims 1 to 12. Memory that stores the data source; and Processor Including, The above processor, Extract metadata from the above data sources, Using the above metadata, create an event map and sample event logs, Calculate the quality of the above sample event log, If the quality value of the above sample event log is higher than the reference value, data is collected from the data source based on the above event map to generate a multi-view event log. A system that generates multi-perspective event logs. In Article 14, The above processor, When generating the event map and sample event log using the above metadata, Convert the above extracted metadata into a standardized entity relationship data model, Create a data catalog by connecting the above entity relationship data models, Infer the activity concept and timestamp concept from the above data catalog, Infer case concepts from the above data catalog, Generate the event map by associating the above activity concept, the above timestamp concept and the above case concept, Extract sample data from the above data source, and extract data from the sample data based on the above event map to generate the sample event log. A system that generates multi-perspective event logs. In Article 15, The above processor, If at least one of the above activity concepts, timestamp concepts and case concepts is not found, an event map is generated based on data quality, Generate multi-view event logs from the above data sources based on the event map generated based on the above data quality. A system that generates multi-perspective event logs. In Article 14, The above processor, When calculating the quality of the above sample event log, Calculated by considering at least one of the case identifier ratio, trace mutation ratio, average ratio of unique activities per case, endpoint activity ratio, and start-end activity ratio of the above sample event log. A system that generates multi-perspective event logs. In Article 14, The above processor, If the quality value of the above sample event log is below the reference value, an event map is generated based on the data quality. Generate multi-view event logs from the above data sources based on the event map generated based on the above data quality. A system that generates multi-perspective event logs.

Images