A method for managing and applying process mining steps

Abstract

The application discloses a kind of based on process mining step management and application method;Including the following steps: S1, data acquisition;S2, index establishment;S3, allocation filter;S4, filter selection;S5, configure screening condition;S6, the storage of screening stack;S7, create data view;S8, increase filter;S9, obtain data view;S10, formula save;S11, exit;S12, call object;S13, read all configuration information;The application is compared with traditional data screening processing, and the multi-dimensional, multi-step screening processing of business log type sequence data is carried out, the target demand is reached, the step of screening processing is managed, each newly created processing step can be flexibly sorted, enabled or closed, more flexible for the purpose of process mining exploration, while also supporting the saving and retrieval of processing step sequence, providing continuity and best practices for processing work, greatly saving time and reuse value experience.

Description

Technical Field

[0001] This invention belongs to the field of process mining technology, and more specifically, it relates to a method for process mining step management and application. Background Technology

[0002] Process mining is a data analytics technique that aims to extract process models from enterprise systems (such as Enterprise Resource Planning (ERP), Customer Relationship Management (CRM), and workflow management systems) to better understand, monitor, and optimize business processes. It combines data mining and process modeling techniques, utilizing event logs recorded in enterprise systems to automatically discover, monitor, and improve actually executed business processes.

[0003] As enterprises advance their digital transformation, more and more companies rely on automation systems in their business management, generating massive amounts of data. This data typically includes execution records of business processes, thus providing valuable insights for process optimization.

[0004] However, existing technologies have several problems: Traditionally, Excel is commonly used for data filtering and penetration. After importing raw data, users enable the filtering function in Excel. Then, they click on the fields to be filtered in the header area and perform the necessary filtering operations. However, existing technologies have several issues: the more fields that need to be filtered simultaneously, the more likely users are to forget the previous filtering process; when there are many fields and steps to filter, adjusting a certain filtering condition or status in a previous step requires constant recall and possible trial and error; they cannot support flexible adjustment of the order of filtering steps; the filtering process is a one-time activity, and when the same data is opened again and the same processing is required, each step must be manually executed from the beginning; and for the same project data, if the task is exited midway and then re-entered after a period of time, the previous operation steps may have been forgotten, requiring time and effort to recall before proceeding; the filtering process cannot be replicated and shared as experience and best practices among human groups. Therefore, we propose a method for process mining-based step management and application. Summary of the Invention

[0005] To address the problems existing in the prior art, the purpose of this invention is to provide a method for managing and applying process mining steps. By managing the steps of data filtering operations, the data processing business process becomes more flexible, controllable, and reusable, thereby improving overall business efficiency.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for process mining step management and application, comprising the following steps:

[0007] S1. Data Collection: Create a project and import the business log data that needs to be filtered. The data is stored in the database.

[0008] S2. Create indexes: Create indexes for the data fields using the database based on their characteristics;

[0009] S3. Assignment Filters: Classify data according to its business dimensions and assign it to different filters. Time dimension corresponds to time interval filter, business steps correspond to process flow filter, time dimension and sequence number dimension correspond to performance filter, business steps and sequence number correspond to business tracking filter, and resource dimension and business dimension correspond to object attribute filter.

[0010] S4. Select Filter: Select a filter through the user interface;

[0011] S5. Configure filter conditions: The filter provides corresponding dimension data for users to observe. Users can configure filter conditions by clicking on the interface. After the user confirms and submits, the corresponding filter parameters are obtained.

[0012] S6. Storage of the filter stack: Temporarily store the corresponding filter categories and filter conditions in the filter stack object;

[0013] S7. Create a data view: Generate corresponding SQL based on the filters and filtering conditions in the filter stack, and simultaneously filter the database to create a data view;

[0014] S8. Add a filter: If the user needs to add a filter, repeat to S4. If the user needs to adjust the previously configured filtering conditions or steps, adjust the filter step order and filtering conditions in the filtering stack through the user interface. After the adjustment is completed, repeat to S6.

[0015] S9. Obtain Data View: The user obtains the data view result they need;

[0016] S10, Recipe Saving: The user saves the filtering steps, corresponding filters and filtering conditions in the filtering stack as a recipe, which is saved as a filtering stack object to the corresponding project and saved in JSON data format to the data table corresponding to the project.

[0017] S11, Exit: The user exits the project;

[0018] S12, Calling Object: When the user re-enters the project, they can select the saved filter stack object mentioned above by clicking the load button;

[0019] S13, Read all configuration information: Successfully read all configuration information of the filter stack from the filter stack data table, and then the system executes S7.

[0020] Specifically, the business log data in S1 is preprocessed before being stored in the database; the preprocessing is used to clean and transform the data.

[0021] Cleaning is used to remove noise from data, detect anomalies, detect missing values, and fill in missing values; transformation is used to convert data formats.

[0022] Specifically, the noise removal is performed for each data point. Median filtering is performed using windows of different scales, and then the filtering results of different scales are weighted and fused.

[0023] ,

[0024] in, This represents the filtered output data; This indicates the number of scales, i.e., the number of different scales used when performing median filtering, with each scale corresponding to a filter window of a different size; Representing scale The weights are used to perform a weighted average of the filtering results at different scales; Indicated in scale The following is the result of median filtering on the data. Specifically, In scale All data within the window range below, i.e., neighborhood data at this scale; Representative scale The size of the corresponding neighborhood window varies at different scales.

[0025] Specifically, the missing value detection is performed using isna() and isnull(); isna() and isnull() are used to return a Boolean DataFrame or Series with the same data structure as the original data, where True indicates a missing value and False indicates a non-missing value;

[0026] Furthermore, notna() and notnull() are the return methods of isna() and isnull(), respectively. Returning True indicates a non-null value, and False indicates a null value.

[0027] Furthermore, the notna() and notnull() functions, combined with the sum() method, count the missing values ​​in each column or row to determine the number of missing values.

[0028] Furthermore, the Any() and All() methods can be used to help detect whether a column contains any missing values ​​or consists entirely of missing values.

[0029] Specifically, the anomaly detection is calculated as follows:

[0030] The Z-score of a data point is obtained by dividing the difference between the data point and the mean by the standard deviation.

[0031] ,

[0032] in, It is the value of the data. The mean of the data, Z is the standard deviation of the data. Points with a Z value greater than 3 or less than -3 are considered outliers and are removed.

[0033] The missing value imputation is used to fill the gaps left by outlier removal and to fill the gaps left by missing value detection using the mean value. The mean value is calculated as follows:

[0034] ,

[0035] in, For values ​​of non-missing data, This represents the number of non-missing data.

[0036] Specifically, the format conversion scales the data to the range [0, 1] through normalization, and the normalization is calculated as follows:

[0037] ,

[0038] in, For the input data, It is the minimum value in the data. It is the maximum value in the data, scaling the data to the range [0, 1].

[0039] Specifically, the characteristics of the data fields include event ID, resource, timestamp, activity name, and associated case ID;

[0040] The features of the data fields are extracted and calculated as follows:

[0041] ,

[0042] in, This reflects the importance of a feature in a data field; the more frequently it appears, the higher the TF value. Indicates the characteristics of computation, Represented as features The number of times it appears in the data field Used to measure the prevalence of a feature across the entire corpus. The higher the value, The less common a feature is in the corpus, the better. Represented as the total number of texts in the corpus. Represented as containing features The number of texts;

[0043] If a feature appears frequently in the data field that needs to be detected and calculated, but appears less frequently in other text fields, A high value indicates the importance of the feature to the data field that needs to be detected and calculated;

[0044] Conversely, if the feature frequency is low or the feature appears in many texts, then A low value indicates that the feature is not very important to the data field that needs to be detected and calculated.

[0045] Specifically, in S7, the SQL queries data using filter categories and filtering conditions, and generates a data view from the query results;

[0046] The steps for generating the data view are as follows:

[0047] Select view type: Use Tableau to select the appropriate view and generate the view;

[0048] Data processing: SQL queries data using filter categories and filtering conditions, and stores the queried data in a view;

[0049] Publish and share: Display the generated view in the user interface.

[0050] Specifically, the formula in S10 includes a screening step, a corresponding filter, and screening conditions;

[0051] Filtering steps: Filter each filter operation in the filter stack;

[0052] Filters: Filters used in each screening step;

[0053] Filter criteria: Specific conditions under the filter;

[0054] The recipe saves the screening steps, corresponding filters, and screening conditions in JSON data format, and stores the data in the recipe through the data view.

[0055] Specifically, the filtering steps, corresponding filters, and filtering conditions in the JSON data format are associated with multiple keys;

[0056] The key is generated using a hash function, which is calculated as follows:

[0057] The formula for outputting the hash value is expressed as:

[0058] ,

[0059] in: This is the current hash value. It is the initial hash value. This refers to the current filtering step, the corresponding filter, and the filtering conditions. Hash function.

[0060] The technical effects and advantages of this invention are as follows:

[0061] Compared with traditional data filtering and processing, this invention performs multi-dimensional and multi-step filtering and processing on business log-type sequence data to achieve target requirements. It manages the filtering and processing steps, and can flexibly sort, enable or disable each newly created processing step, making it more flexible for the purpose of process mining and exploration. At the same time, it also supports the saving and retrieval of processing step sequences, providing continuity and best practices for processing work, greatly saving time and reusing valuable experience.

[0062] During process mining, filters are assigned to query data, and filter conditions are set to obtain corresponding filter parameters. Filter categories and conditions are temporarily stored in a filter stack object, which is then saved as a recipe in JSON format. This allows for easy loading and application of these filter recipes based on project IDs, providing flexibility and enabling easy expansion of different filter conditions and step types. The query results are also displayed in a data view for easy visualization, and the filter stack data table is easily accessible for subsequent retrieval, enabling queries to utilize all configuration information of the filter stack for further queries, thus improving query efficiency.

[0063] Other features and advantages of the invention will become clear from the following detailed description of exemplary embodiments of the invention with reference to the accompanying drawings. Attached Figure Description

[0064] Figure 1 This is a schematic diagram of the steps provided by the present invention. Detailed Implementation

[0065] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without inventive effort are within the scope of protection of this invention.

[0066] like Figure 1 As shown in the figure, an embodiment of the present invention provides a method for process mining step management and application, which includes the following steps:

[0067] S1. Data Collection: Create a project and import the business log data that needs to be filtered. The data is stored in the database.

[0068] S2. Create indexes: Create indexes for the data fields using the database based on their characteristics;

[0069] S3. Assignment Filters: Classify data according to its business dimensions and assign it to different filters. Time dimension corresponds to time interval filter, business steps correspond to process flow filter, time dimension and sequence number dimension correspond to performance filter, business steps and sequence number correspond to business tracking filter, and resource dimension and business dimension correspond to object attribute filter.

[0070] S4. Select Filter: Select a filter through the user interface;

[0071] S5. Configure filter conditions: The filter provides corresponding dimension data for users to observe. Users can configure filter conditions by clicking on the interface. After the user confirms and submits, the corresponding filter parameters are obtained.

[0072] S6. Storage of the filter stack: Temporarily store the corresponding filter categories and filter conditions in the filter stack object;

[0073] S7. Create a data view: Generate corresponding SQL based on the filters and filtering conditions in the filter stack, and simultaneously filter the database to create a data view;

[0074] S8. Add a filter: If the user needs to add a filter, repeat to S4. If the user needs to adjust the previously configured filtering conditions or steps, adjust the filter step order and filtering conditions in the filtering stack through the user interface. After the adjustment is completed, repeat to S6.

[0075] S9. Obtain Data View: The user obtains the data view result they need;

[0076] S10, Recipe Saving: The user saves the filtering steps, corresponding filters and filtering conditions in the filtering stack as a recipe, which is saved as a filtering stack object to the corresponding project and saved in JSON data format to the data table corresponding to the project.

[0077] S11, Exit: The user exits the project;

[0078] S12, Calling Object: When the user re-enters the project, they can select the saved filter stack object mentioned above by clicking the load button;

[0079] S13, Read all configuration information: Successfully read all configuration information of the filter stack from the filter stack data table, and then the system executes S7.

[0080] In this embodiment, preferably, the business log data in S1 is preprocessed before being stored in the database; the preprocessing is used to clean and transform the data.

[0081] Cleaning is used to remove noise from data, perform anomaly detection, missing value detection, and imputation of missing values; transformation is used to convert data formats.

[0082] It should be noted that preprocessing cleans and transforms the data, making it easier to remove noise, detect outliers and missing values, fill in missing values ​​to ensure data integrity, and convert the format to maintain data consistency.

[0083] In this embodiment, preferably, the noise removal is performed for each data point. Median filtering is performed using windows of different scales, and then the filtering results of different scales are weighted and fused.

[0084] ,

[0085] in, This represents the filtered output data; This indicates the number of scales, i.e., the number of different scales used when performing median filtering, with each scale corresponding to a filter window of a different size; Representing scale The weights are used to perform a weighted average of the filtering results at different scales; Indicated in scale The following is the result of median filtering on the data. Specifically, In scale All data within the window range below, i.e., neighborhood data at this scale; Representative scale The size of the corresponding neighborhood window varies at different scales;

[0086] It should be noted that median filtering is used to filter the data, and the filtered results are weighted and fused to remove noise from the data, preventing noise from causing errors in subsequent data processing and improving the accuracy of the data.

[0087] In this embodiment, preferably, the missing value detection is performed using isna() and isnull(); isna() and isnull() are used to return a Boolean DataFrame or Series with the same data structure as the original data, where True represents a missing value and False represents a non-missing value;

[0088] Furthermore, notna() and notnull() are the return methods of isna() and isnull(), respectively. Returning True indicates a non-null value, and False indicates a null value.

[0089] Furthermore, the notna() and notnull() functions, combined with the sum() method, count the missing values ​​in each column or row to determine the number of missing values.

[0090] Furthermore, the Any() and all() methods can be used to help detect whether a column contains any missing values ​​or consists entirely of missing values;

[0091] It should be noted that missing value detection is used to detect missing values ​​in the data, and all detected missing values ​​are statistically analyzed to facilitate subsequent data filling.

[0092] In this embodiment, preferably, the anomaly detection is calculated as follows:

[0093] The Z-score of a data point is obtained by dividing the difference between the data point and the mean by the standard deviation.

[0094] ,

[0095] in, It is the value of the data. The mean of the data, Z is the standard deviation of the data. Points with a Z value greater than 3 or less than -3 are considered outliers and are removed.

[0096] The missing value imputation is used to fill the gaps left by outlier removal and to fill the gaps left by missing value detection using the mean value. The mean value is calculated as follows:

[0097] ,

[0098] in, For values ​​of non-missing data, The number of non-missing data;

[0099] It should be noted that Z-score is used to detect and remove outliers in the data. The removed outliers and missing values ​​are then filled using the mean, which improves the completeness of the data and maintains its accuracy.

[0100] In this embodiment, preferably, the format conversion scales the data to the range [0, 1] through normalization processing, and the normalization processing is calculated as follows:

[0101] ,

[0102] in, For the input data, It is the minimum value in the data. It is the maximum value in the data, scaling the data to the range [0, 1].

[0103] It should be noted that normalization scaling the data to the range of [0, 1] facilitates subsequent data processing and reduces computational pressure and complexity.

[0104] In this embodiment, preferably, the data field features include event ID, resource, timestamp, activity name, and associated case ID;

[0105] The features of the data fields are extracted and calculated as follows:

[0106] ,

[0107] in, This reflects the importance of a feature in a data field; the more frequently it appears, the higher the TF value. Indicates the characteristics of computation, Represented as features The number of times it appears in the data field Used to measure the prevalence of a feature across the entire corpus. The higher the value, The less common a feature is in the corpus, the better. Represented as the total number of texts in the corpus. Represented as containing features The number of texts;

[0108] If a feature appears frequently in the data field that needs to be detected and calculated, but appears less frequently in other text fields, A high value indicates the importance of the feature to the data field that needs to be detected and calculated;

[0109] Conversely, if the feature frequency is low or the feature appears in many texts, then A low value indicates that the feature is not very important to the data field that needs to be detected and calculated.

[0110] It should be noted that feature extraction of data fields enables the extraction of features from event ID, resource, timestamp, activity name, and associated case ID, effectively obtaining feature information from the data.

[0111] In this embodiment, preferably, the SQL in S7 queries data through filter categories and filtering conditions, and generates a data view from the query results;

[0112] The steps for generating the data view are as follows:

[0113] Select view type: Use Tableau to select the appropriate view and generate the view;

[0114] Data processing: SQL queries data using filter categories and filtering conditions, and stores the queried data in a view;

[0115] Publish and share: Display the generated view in the user interface;

[0116] It should be noted that by creating a view, the queried data is populated and displayed, which facilitates the presentation of the queried data, the filtering steps, corresponding filters, and filtering conditions.

[0117] In this embodiment, preferably, the formula in S10 includes a screening step, a corresponding filter, and screening conditions;

[0118] Filtering steps: Filter each filter operation in the filter stack;

[0119] Filters: Filters used in each screening step;

[0120] Filter criteria: Specific conditions under the filter;

[0121] The recipe saves the screening steps, corresponding filters, and screening conditions in JSON data format, and stores the data in the recipe through the data view;

[0122] It should be noted that the filtering steps, corresponding filters, and filtering conditions are linked through JSON data format, so that the entire filtering step, corresponding filters, and filtering conditions can be retrieved when a query is performed.

[0123] In this embodiment, preferably, the filtering steps, corresponding filters, and filtering conditions in the JSON data format are associated by multiple keys;

[0124] The key is generated using a hash function, which is calculated as follows:

[0125] The formula for outputting the hash value is expressed as:

[0126] ,

[0127] in: This is the current hash value. It is the initial hash value. This refers to the current filtering step, the corresponding filter, and the filtering conditions. Hash function;

[0128] It should be noted that a hash function is used to calculate the correlation value between the filtering steps, the corresponding filters, and the filtering conditions, thereby establishing a correlation between them.

[0129] The specific operating steps of this invention are as follows:

[0130] Step 1: Data Collection: Create a project and import the business log data that needs to be filtered. The data is stored in the database.

[0131] Step 2: Create indexes: Based on the characteristics of the data fields, create indexes for them using the database;

[0132] Step 3: Assign filters: Classify the data according to its business dimensions and assign it to different filters. Time dimension corresponds to time interval filter, business steps correspond to process flow filter, time and sequence number dimensions correspond to performance filter, business steps and sequence number correspond to business tracking filter, and resource and business dimensions correspond to object attribute filter.

[0133] Step 4: Select a filter: Select a filter through the user interface;

[0134] Step 5: Configure filter conditions: The filter provides the corresponding dimension data for the user to observe. The user configures the filter conditions by clicking on the interface. After the user confirms and submits, the corresponding filter parameters are obtained.

[0135] Step 6: Storing the filter stack: Temporarily store the corresponding filter categories and filtering conditions in the filter stack object;

[0136] Step 7: Create a data view: Generate corresponding SQL based on the filters and filtering conditions in the filter stack, and simultaneously filter the database to create a data view;

[0137] Step 8: Add a filter: If the user needs to add a filter, repeat step 4. If the user needs to adjust the previously configured filtering conditions or steps, adjust the filter step order and filtering conditions in the filtering stack through the user interface. After the adjustment is completed, repeat step 6.

[0138] Step 9: Obtain the data view: The user obtains the data view result they need;

[0139] Step 10, Recipe Saving: The user saves the filtering steps, corresponding filters, and filtering conditions in the filter stack as a recipe, which is saved as a filter stack object to the corresponding project and in JSON data format to the data table corresponding to the project.

[0140] Step 11: Exit: The user exits the project;

[0141] Step 12, Call the object: The user re-enters the project and selects the saved filter stack object mentioned above via the load button;

[0142] Step 13: Read all configuration information: Successfully read all configuration information of the filter stack from the filter stack data table, and then the system executes step 7.

[0143] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for process mining step management and application, characterized in that, Includes the following steps: S1. Data Collection: Create a project and import the business log data that needs to be filtered. The data is stored in the database. S2. Create indexes: Create indexes for the data fields using the database based on their characteristics; The data fields are characterized by event ID, resource, timestamp, activity name, and associated case ID; The features of the data fields are extracted and calculated as follows: ， in, This reflects the importance of a feature in a data field; the more frequently it appears, the higher the TF value. Indicates the characteristics of computation, Represented as features The number of times it appears in the data field Used to measure the prevalence of a feature across the entire corpus. The higher the value, The less common a feature is in the corpus, the better. Represented as the total number of texts in the corpus. Represented as containing features The number of texts; If a feature appears frequently in the data field that needs to be detected and calculated, but appears less frequently in other text fields, A high value indicates the importance of the feature to the data field that needs to be detected and calculated; Conversely, if the feature frequency is low or the feature appears in many texts, then A low value indicates that the feature is not very important to the data field that needs to be detected and calculated. S3. Assignment Filters: Classify data according to its business dimensions and assign it to different filters. Time dimension corresponds to time interval filter, business steps correspond to process flow filter, time dimension and sequence number dimension correspond to performance filter, business steps and sequence number correspond to business tracking filter, and resource dimension and business dimension correspond to object attribute filter. S4. Select Filter: Select a filter through the user interface; S5. Configure filter conditions: The filter provides corresponding dimension data for users to observe. Users can configure filter conditions by clicking on the interface. After the user confirms and submits, the corresponding filter parameters are obtained. S6. Storage of the filter stack: Temporarily store the corresponding filter categories and filter conditions in the filter stack object; S7. Create a data view: Generate corresponding SQL based on the filters and filtering conditions in the filter stack, and simultaneously filter the database to create a data view; S8. Add a filter: If the user needs to add a filter, repeat to S4. If the user needs to adjust the previously configured filtering conditions or steps, adjust the filter step order and filtering conditions in the filtering stack through the user interface. After the adjustment is completed, repeat to S6. S9. Obtain Data View: The user obtains the data view result they need; S10, Recipe Saving: The user saves the filtering steps, corresponding filters and filtering conditions in the filtering stack as a recipe, which is saved as a filtering stack object to the corresponding project and saved in JSON data format to the data table corresponding to the project. S11, Exit: The user exits the project; S12, Calling Object: When the user re-enters the project, they can select the saved filter stack object mentioned above by clicking the load button; S13, Read all configuration information: Successfully read all configuration information of the filter stack from the filter stack data table, and then the system executes S7.

2. The method for process mining step management and application according to claim 1, characterized in that: The business log data in S1 is preprocessed before being stored in the database; preprocessing is used to clean and transform the data. Cleaning is used to remove noise from data, detect anomalies, detect missing values, and fill in missing values; transformation is used to convert data formats.

3. The method for process mining step management and application according to claim 2, characterized in that: The noise removal is for each data point. Median filtering is performed using windows of different scales, and then the filtering results of different scales are weighted and fused. ， in, This represents the filtered output data; This indicates the number of scales, i.e., the number of different scales used when performing median filtering, with each scale corresponding to a filter window of a different size; Representing scale The weights are used to perform a weighted average of the filtering results at different scales; Indicated in scale The following is the result of median filtering on the data. Specifically, In scale All data within the window range below, i.e., neighborhood data at this scale; Representative scale The size of the corresponding neighborhood window varies at different scales.

4. The method for process mining step management and application according to claim 2, characterized in that: The missing value detection is performed using isna() and isnull(); isna() and isnull() are used to return a Boolean DataFrame or Series with the same data structure as the original data, where True indicates a missing value and False indicates a non-missing value; Furthermore, notna() and notnull() are the return methods of isna() and isnull(), respectively. Returning True indicates a non-null value, and False indicates a null value. Furthermore, the notna() and notnull() functions, combined with the sum() method, count the missing values ​​in each column or row to determine the number of missing values. Furthermore, the Any() and All() methods can be used to help detect whether a column contains any missing values ​​or is entirely composed of missing values.

5. The method for process mining step management and application according to claim 2, characterized in that: The anomaly detection is calculated as follows: The Z-score of a data point is obtained by dividing the difference between the data point and the mean by the standard deviation. ， in, It is the value of the data. The mean of the data, Z is the standard deviation of the data. Points with a Z value greater than 3 or less than -3 are considered outliers and are removed. The missing value imputation is used to fill the gaps left by outlier removal and to fill the gaps left by missing value detection using the mean value. The mean value is calculated as follows: ， in, For values ​​of non-missing data, This represents the number of non-missing data.

6. The method for process mining step management and application according to claim 2, characterized in that: The format conversion scales the data to the range [0, 1] through normalization, and the normalization is calculated as follows: ， in, For the input data, It is the minimum value in the data. It is the maximum value in the data, scaling the data to the range [0, 1].

7. The method for process mining step management and application according to claim 1, characterized in that: The SQL in S7 queries data through filter categories and filtering conditions, and generates a data view from the query results; The steps for generating the data view are as follows: Select view type: Use Tableau to select the appropriate view and generate the view; Data processing: SQL queries data using filter categories and filtering conditions, and stores the queried data in a view; Publish and share: Display the generated view in the user interface.

8. The method for process mining step management and application according to claim 7, characterized in that: The formula in S10 includes a screening step, a corresponding filter, and screening conditions; Filtering steps: Each filter operation in the filter stack; Filters: Filters used in each screening step; Filter criteria: Specific conditions under the filter; The recipe saves the screening steps, corresponding filters, and screening conditions in JSON data format, and stores the data in the recipe through the data view.

9. The method for process mining step management and application according to claim 8, characterized in that: The filtering steps, corresponding filters, and filtering conditions in the JSON data format are associated with multiple keys; The key is generated using a hash function, which is calculated as follows: The formula for outputting the hash value is expressed as: ， in: It is the current hash value. It is the initial hash value. This refers to the current filtering step, the corresponding filter, and the filtering conditions. Hash function.

Images