Scene-driven multi-agent data collaborative governance method

Abstract

The application discloses a multi-subject data collaborative management method based on scene driving and relates to the technical field of data processing and information management, including the following steps: collecting the change data of the multi-subjects in the continuous promotion process under the same business scene, arranging according to a unified time sequence, recording the update state of each subject at the corresponding time point in the time sequence, and forming a basic record. The application keeps the multi-subject data continuous expression under the unified time reference by constructing a unified time sequence and identifying the rhythm interruption difference, avoids the interference of data static on the overall judgment, improves the state judgment accuracy, positions the abnormal stagnation position, completes the stop update data, restores the change process continuity, presents the real change trend, and improves the change expression integrity and the adjustment response ability.

Description

Technical Field

[0001] This invention relates to the field of data processing and information management technology, specifically to a scenario-driven multi-entity collaborative data governance method. Background Technology

[0002] Scenario-driven multi-entity data collaborative governance refers to a data application environment involving multiple participants (such as different systems, departments, or organizations). Instead of focusing on a single data source or fixed rules, it revolves around the actual needs of specific business scenarios, unifying the data resources generated and controlled by each entity. By aggregating, associating, and aligning data under the same time benchmark and business context, it identifies the differences and connections between different entities in terms of data content, status changes, and usage goals. Based on this, it coordinates and constrains the data flow methods, processing order, and usage permissions, enabling all entities to form a consistent data cognition and collaborative behavior in the same scenario. This avoids information fragmentation, redundant processing, or decision conflicts, achieving orderly sharing and efficient utilization of data among multiple parties and improving the overall governance effect.

[0003] The existing technology has the following shortcomings: In existing technologies, during multi-entity collaborative data processing, the data update rhythm of each participant often depends on its own operational status and proceeds independently, lacking a constraint mechanism for the consistency of overall changes. When one entity prematurely stops updating data before the actual change has ended, its data may appear to be in a continuously stable state, while the data of other entities remains in a dynamic process of change. In this case, when the system makes an overall state judgment, it will be affected by the "static data" of that entity, easily misjudging local stillness as overall stability, thereby masking the real fluctuations that are still occurring. This leads to a deviation in the judgment of the current scene's changing trend, and may even cause subsequent adjustment strategies to lag or fail.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a scenario-driven, multi-subject data collaborative governance method to solve the problems mentioned in the background.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a scenario-driven multi-subject data collaborative governance method, comprising the following steps: Collect change data of multiple entities in the same business scenario during continuous progress, organize it in a unified time sequence, record the update status of each entity at the corresponding time point in the time series, and form a basic record containing the change rhythm and progress. Based on the basic records, the update status of each subject at adjacent time points is continuously analyzed to identify the time segments in which the subject stops updating but the overall change continues, thus forming the time distribution results of rhythm interruption differences. Based on the temporal distribution of rhythm interruption differences, the stop update position is mapped back to the time series for comparison of changes before and after, the time position that is inconsistent with the overall change direction is determined, and an abnormal stagnation position record is generated. By combining records of abnormal stagnation locations, a retrospective analysis is performed on the change sequence after the corresponding time location to unfold the slowdown process of change caused by the cessation of updates, identify the actual change segments that are masked by local stagnation, and obtain the unfolding results of hidden fluctuations. Based on the unfolding results of the hidden fluctuations, the time range corresponding to the abnormal stagnation position is adjusted. Within this time range, the data of the stopped updating subject is supplemented according to the overall change direction, and the static duration interval is shortened to keep the overall change progressing continuously.

[0007] Preferably, the basic record formation steps are as follows: As the business scenario progresses, data on changes in multiple entities are collected hourly and a unified time identifier is attached. The data on changes are arranged in the order of the time identifiers to form a continuous time series structure. The empty space identifiers of the entities are retained in the time position to maintain the consistency of the entity positions. Based on the continuous time series structure, the update status of each subject's change data at each time position is determined, an update status identifier is added to the time position, and the update status sequence of each subject is recorded along the time series, so that the update status sequence corresponds one-to-one with the change data. Based on the update state sequence, the update state of each subject is expanded segment by segment, the update states between adjacent time positions are connected to form update segments and static segments, and the update segments are bound with the changed data to form segment representations. Based on the segment expression, the time series is uniformly collected and the distribution of each subject's update segment and static segment is summarized to form the subject's rhythm trajectory and record the update continuity, thus obtaining a record of the change rhythm and progress.

[0008] Preferably, the steps for forming the temporal distribution results of rhythm interruption differences are as follows: The time position is expanded point by point along the time series direction, the update status identifier of each subject is extracted and the correspondence between adjacent time positions is established, the update status changes are compared to determine the starting time position of the interruption of the update behavior and form a continuous tracking sequence. The update status of subsequent time positions is continuously read around the starting time position. The update status of the corresponding subject is compared with the update status of other subjects, and the combination is used to form a continuous time interval. The subject identifier and time range are determined to obtain the candidate difference interval. Based on the candidate difference interval, the time range is mapped to the changing data sequence, the time position change data is expanded and the change is compared to determine the correspondence between continuous change and static state, thus forming the rhythm interruption difference interval. The time positions are sequentially arranged according to the rhythm interruption difference intervals, and the associated subject identifiers and time ranges form a continuous distribution structure to obtain the time distribution results of rhythm interruption differences.

[0009] Preferably, the determination of rhythm interruption difference intervals includes continuously comparing the change data at each time position within the candidate difference interval, identifying the time interval where the subject maintains unchanged data at adjacent time positions while other subjects show continuous changes, and recording the subject identification and time range association of the time interval, thereby limiting the formation conditions of rhythm interruption difference intervals.

[0010] Preferably, the steps for generating abnormal stop location records are as follows: Extract the starting time position of the time interval from the time distribution results and map it to the time series. Select the previous and next time positions around the starting time position and extract the change data of each subject to form a continuous data segment. Based on continuous data segments, the changes in the subject that has stopped updating are compared between the previous time position and the current time position. Simultaneously, the changes in the remaining subjects are extracted and arranged in parallel to determine the candidate positions for separation of change relationships and bind the subjects. Starting from the candidate position, select subsequent time positions along the time series, continuously compare the change data of each subject, determine the time position where the subject that stops updating remains unchanged and the other subjects continue to change as the inconsistent time position, and associate it with the continuous time range. The inconsistencies in time and location are sorted sequentially, and the main identifiers, time intervals, and changes are collected to form a record of abnormal stagnation locations.

[0011] Preferably, the change data of the subject that stopped updating is continuously extracted at each time position within a continuous time range corresponding to the inconsistency time position, and the data of the subject at adjacent time positions is compared with the change data of other subjects. The inconsistency time position is further defined based on the distribution of change relationships within the continuous time range.

[0012] Preferably, the steps for obtaining the hidden wave expansion result are as follows: Time series location is performed around the abnormal stagnation location record and the corresponding time position is extracted. Subsequent time positions are selected along the time series and the change data of each subject are extracted. Combined with the subject that stops updating, a continuous change sequence is formed. Based on the continuous change sequence, the change data at each time position is expanded, and the main data that stops updating is compared with the other main data. The time range in the continuous time position where the main data that stops updating remains unchanged and the other main data changes is identified as the change slowdown segment. By mapping the slow-down segments to the time series and performing reverse backtracking, the change data at each time point is extracted and continuously compared to determine the time points where the main body stops updating and remains unchanged while the other main bodies continue to change, thus forming the actual change segments. The actual change segments are sorted sequentially and associated with records of abnormal stagnation locations to form a continuously arranged hidden fluctuation unfolding result.

[0013] Preferably, the steps for adjusting the time range corresponding to the abnormal stop position are as follows: The actual change segments are extracted from the unfolding results of the hidden fluctuations and mapped to the time series. The actual change segments are then bound to the abnormal stagnation positions, and the time adjustment interval is determined. The data changes at each time point are expanded around the time adjustment interval, the data changes of each subject are extracted and arranged side by side to form a continuous comparison sequence that includes the static section of the subject that stops updating and the change trajectory of other subjects. Based on the continuous comparison sequence, the data of the subject that stopped updating was completed point by point at each time position, and a continuous change sequence was formed by combining the data of the previous time position with the changes of other subjects. Based on the completion processing results, the time adjustment intervals are organized, static segments are replaced with continuously changing data and sequentially spliced ​​to form a time series expression of continuous overall change.

[0014] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention continuously unfolds the change data of multiple subjects under a unified time series and introduces an update status identifier and a rhythm interruption difference identification process, so that each subject can form a unified time reference relationship under different update rhythm conditions. This avoids the interference of data stagnation caused by individual subjects stopping updates on the overall status judgment, and enables the overall change trend to be expressed based on the continuous change process, thereby improving the coordination consistency and status judgment accuracy of multi-subject data in the same business scenario.

[0015] This invention completes the data of the stopped updating subject within a time range by locating abnormal stagnation positions and combining the unfolding results of hidden fluctuations. This replaces the static segments formed by the interruption of updates with a continuous sequence of changes, thereby restoring the continuous advancement relationship of the overall change in the time dimension. This allows the real change process that was hidden by local stagnation to be presented, improves the completeness of the change process expression, and enhances the responsiveness of subsequent adjustment strategies to the change trend. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0017] Figure 1 This is a flowchart of the scenario-driven multi-subject data collaborative governance method of the present invention. Detailed Implementation

[0018] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0019] This invention provides, for example Figure 1 The scenario-driven, multi-entity data collaborative governance method shown includes the following steps: Collect change data of multiple entities in the same business scenario during continuous progress, organize it in a unified time sequence, record the update status of each entity at the corresponding time point in the time series, and form a basic record containing the change rhythm and progress. As multiple entities continuously advance around the same business scenario, the changes in data from each entity are uniformly organized and expressed to form a continuous foundation of records upon which subsequent analysis relies. The specific implementation steps are as follows: Following the continuous advancement of business scenarios, the system collects change data generated by multiple entities during operation on an hourly basis. During collection, a clear time stamp is attached to each data entry, generated using a unified time base. This ensures that data from different entities under varying conditions can be mapped to specific locations on the same timeline. After attaching the time stamps, the data from each entity is arranged sequentially according to the time stamps, forming a single, progressive sequence across time. In this sequence, each time point contains change data records from all entities at that corresponding moment. When an entity fails to generate data at a given time point, its empty slot is retained to maintain consistency in the time series, thus creating a continuous time series structure with time as the main thread and entities unfolding in parallel. Based on the established continuous time series structure, for each subject's change data at each time position, it is determined whether each subject has updated its data at that time position, and an update status identifier is attached to each subject at the corresponding time position. Specifically, when a subject has newly collected data at the current time position, the time position is marked as an update state; when a subject has not generated new change data at the current time position and continues the data content from the previous time position, the time position is marked as no update state. After completing the status identifier for a single time position, the update status of each subject is continuously recorded along the time series sequence, so that each subject forms an update status sequence composed of continuous time positions over the entire time range. This update status sequence corresponds one-to-one with the change data of the corresponding subject at the time position, thus presenting both change data and update status information simultaneously in the same time series. Based on the time series containing changed data and update status identifiers, the update status sequence of each subject is expanded segment by segment throughout the entire time range. The update status changes of each subject between adjacent time positions are connected one by one, so that time positions with continuous updates form continuous update segments, and time positions with continuous non-updates form continuous static segments. After completing the division of update segments and static segments for each subject, the corresponding segments are bound to the changed data within those segments, so that each update segment corresponds to a continuous change data sequence, and each static segment corresponds to a data record that remains unchanged. This forms a structural expression in the time series composed of alternating update segments and static segments, and this structural expression can reflect the change rhythm and progress status of each subject throughout the entire process. After completing the structural representation of the update and static segments of each subject, the entire time series is uniformly organized. The changed data, update status identifiers, and segment information of each time position are uniformly collected, so that each time position in the time series contains complete and consistent data content. Subsequently, the update status sequences of each subject are summarized along the time series direction. The distribution of update segments and static segments of each subject across the entire time range are continuously spliced ​​together to form a complete rhythm trajectory that reflects the subject's progress in the entire business scenario. At the end of the time series, the update continuity of each subject is recorded centrally. Through the unified presentation of the rhythm trajectories of all subjects, a basic record containing the changing rhythm and progress is formed. This basic record is continuously arranged in the time dimension, maintains consistent position in the subject dimension, and maintains a unified correspondence in the status representation, providing a complete and coherent data foundation for subsequent rhythm interruption difference identification and abnormal stagnation location based on the time series.

[0020] Based on the basic records, the update status of each subject at adjacent time points is continuously analyzed to identify the time segments in which the subject stops updating but the overall change continues, thus forming the time distribution results of rhythm interruption differences. After establishing a basic record containing changes in rhythm and progress, the basic record is expanded time-by-time and linked with continuous states to obtain the distribution of rhythm differences among multiple entities during the advancement process. The specific implementation steps are as follows: Along the time sequence direction in the basic record, each time position is expanded point by point in chronological order. During the expansion process, the update status identifiers corresponding to all subjects at that time position are extracted sequentially, and the current time position is connected with the immediately preceding time position to establish a direct correspondence between the two time positions in time sequence. After establishing this correspondence, the update status of each subject at these two consecutive time positions is compared one by one. When the update status of a subject at the previous time position is "updated" but the update status at the current time position is "not updated", the current time position is determined as the starting time position where the subject's update behavior is interrupted. The update status of the subject after that time position is expanded and recorded along the time sequence, so that the change in the subject's update status can form a continuous tracking sequence starting from the starting time position, thereby providing a clear starting basis for subsequently identifying the time range of the stop update. After obtaining the starting time position of the interruption of each subject's update behavior, starting from each starting time position, the update status of the corresponding subject at each subsequent time position is read one by one along the time sequence direction, and the update status of the subject at these time positions is synchronously compared with the update status of other subjects at the same time positions. During the comparison process, when the subject remains in a state of no update in multiple consecutive time positions, and there is at least one other subject with an update status of update in these same time positions, these consecutive time positions are combined into a continuous time interval, and the start time position and end time position of the continuous time interval are marked one by one, so that the time interval can fully reflect the time range in which a single subject stops updating while other subjects continue to advance. At the same time, the time interval is bound to the corresponding subject to form a candidate difference interval with a clear time range and subject identification. After forming candidate difference intervals, the time range corresponding to each candidate difference interval is mapped back to the change data sequence in the base record. The change data of each subject at each time position within the time range is expanded point by point, so that the change data and the update status form a one-to-one correspondence at the same time position. During the expansion process, the change data at each time position within the candidate difference interval is compared sequentially. When at least one subject in the time range shows continuous change between adjacent time positions, while the corresponding subject that stops updating keeps the data unchanged at all time positions within the time range, the candidate difference interval is determined as the rhythm interruption difference interval. The subject identifier, the start time position, and the end time position of the interval are recorded centrally, so that each rhythm interruption difference interval is completely expressed in a unified structural form. After determining all rhythm interruption difference intervals, all rhythm interruption difference intervals are organized in chronological order. The start and end times of each interval are arranged according to the time sequence, and the corresponding subject identifiers are associated with the interval information one by one, so that all rhythm interruption difference intervals form a continuous distribution structure in the time dimension. In this continuous distribution structure, the original time position relationship of each interval in the basic record is preserved, so that the connection relationship between rhythm interruption difference intervals can be fully presented, thus forming the time distribution result of rhythm interruption difference. This time distribution result can reflect the deviation of the advancement rhythm between multiple subjects due to the cessation of updates by individual subjects within a unified time series framework, and provide continuous and clear time basis for subsequent location of abnormal stagnation positions.

[0021] Based on the temporal distribution of rhythm interruption differences, the stop update position is mapped back to the time series for comparison of changes before and after, the time position that is inconsistent with the overall change direction is determined, and an abnormal stagnation position record is generated. After obtaining the temporal distribution results of rhythm interruption differences, the corresponding stop update positions are embedded into the continuous time series in the base record, and the process is carried out layer by layer around the changes in the time series to determine the specific time positions that are inconsistent with the overall change direction. The specific implementation steps are as follows: For each time interval included in the time distribution results of rhythm interruption differences, the starting time position of each time interval is extracted and mapped to the time series in the base record, so that the time position has a unique location in the complete time series. After the location is completed, the immediately preceding time position is selected and the immediately following time position is selected, centered on the starting time position. The change data of all subjects are extracted one by one in these three consecutive time positions. The identifier of the subject that stopped updating at the starting time position and the corresponding change data are retained together. In this way, a continuous data segment containing the previous time position, the current time position and the next time position are constructed, so that the data segment forms a continuous connection in the time dimension, maintains a consistent arrangement in the subject dimension, and has a complete correspondence in data expression, providing a direct basis for subsequent comparison of changes before and after. Based on continuous data segments, the changes in data of the subject that has stopped updating between the previous time position and the current time position are compared item by item. At the same time, the changes in data of other subjects between the same two time positions are extracted synchronously and arranged in parallel according to the subject order, so that the changes of each subject within the same time span form a horizontal comparison relationship. In this comparison relationship, the data changes of each subject between the previous time position and the current time position are continuously observed. When the data of the subject that has stopped updating is consistent between these two time positions, while the data of other subjects changes continuously within the same time span, the starting time position is marked as a candidate position where the change relationship has broken down. This candidate position is then bound to the corresponding subject that has stopped updating, so that the candidate position forms an identifier with subject pointing information in the time series, thereby providing a basis for further determining the inconsistency positions. After obtaining candidate positions where the change relationship breaks down, starting from these candidate positions, multiple subsequent consecutive time positions are selected one by one along the time series direction. Change data of each subject is extracted point by point in these time positions, and the data of the subject that has stopped updating in these time positions is continuously compared with the data of other subjects in the corresponding time positions. This establishes a vertical comparison relationship between the static state of the subject that has stopped updating and the continuous change state of other subjects. During this comparison process, when other subjects' data change continuously while the data of the subject that has stopped updating remains unchanged in multiple consecutive time positions, the candidate position is determined as a time position that is inconsistent with the overall change direction. This time position is then associated with its subsequent consecutive time ranges, so that the inconsistent position not only has a single time point identifier but also has corresponding time range information, thus forming a complete expression with contextual relationships in the time series. After identifying all inconsistent time locations, these time locations are organized chronologically within the time series of the basic records. The subject identifier, time interval, and relationships between preceding and following time locations are collected item by item for each inconsistent time location, ensuring that each inconsistent time location possesses complete time information, subject information, and relationship information. A unified identifier is added to these time locations in the time series, enabling them to be clearly distinguished within the overall time series. By continuously arranging and structurally organizing all inconsistent time locations, an abnormal stagnation record is formed. This abnormal stagnation record maintains continuous expression in the time dimension, corresponding relationships in the subject dimension, and complete presentation in terms of relationships, thus providing clear temporal basis and data support for subsequent identification of hidden fluctuations around this record.

[0022] By combining records of abnormal stagnation locations, a retrospective analysis is performed on the change sequence after the corresponding time location to unfold the slowdown process of change caused by the cessation of updates, identify the actual change segments that are masked by local stagnation, and obtain the unfolding results of hidden fluctuations. After obtaining the records of abnormal stagnation locations, the abnormal stagnation location records are correlated point by point with the continuous time series in the base records. Then, the change sequence after the abnormal stagnation time position is continuously expanded and reversed to restore the true change process that was hidden by local stagnation. The specific implementation steps are as follows: For each abnormal stagnation time position contained in the abnormal stagnation location record, it is located one by one in the time series of the basic record. The time identifier corresponding to the time position is matched one by one with the subject identifier, so that the abnormal stagnation time position forms a clear location point in the time series. After the location is completed, starting from the abnormal stagnation time position, multiple consecutive time positions are selected one by one along the time series direction. In each selected time position, all the change data corresponding to the subject are extracted. The identifier of the subject that stopped updating at the abnormal stagnation time position is continuously bound to its data status in each subsequent time position, so that the abnormal stagnation time position and its multiple subsequent time positions form a complete continuous change sequence. This change sequence maintains sequential progression in the time dimension and maintains a consistent arrangement in the subject dimension, thereby constructing the expression structure of the subsequent change sequence starting from the abnormal stagnation time position. After forming the subsequent change sequence, the change data at each time point is expanded point by point along the time sequence of the change sequence. The data performance of the subject that has stopped updating in the change sequence is presented side by side with the change data of other subjects at the same time point, so that the change situation of all subjects at the same time point can form a horizontal comparison relationship. In this comparison relationship, the data change of each subject between adjacent time points is continuously observed. When at least one subject continuously changes data between adjacent time points in multiple consecutive time points, while the subject that has stopped updating always maintains consistent data content in these time points, these consecutive time points are combined into a change slowdown segment. The start time and end time point of the change slowdown segment are recorded one by one, so that the segment can fully express the overall change progress restricted due to the existence of the subject that has stopped updating. At the same time, the change slowdown segment is associated with the corresponding abnormal stagnation time point. After obtaining the slowdown zone, this slowdown zone is remapped to the complete time series in the base record. The change data at all time points within this zone are then expanded in reverse. That is, starting from the end time point of the slowdown zone, the data is traced back along the time series to the abnormal stagnation time point. During the tracing process, all subject change data at each time point is read point by point, and the data status of the stagnant subject at each time point is continuously compared with the changes of other subjects at the corresponding time points. During this tracing process, when there are cases where other subject data continues to change at multiple consecutive time points, while the stagnant subject does not change at the corresponding time points, these time points are extracted one by one and rearranged in chronological order, so that these time points form independent actual change zone expressions, thereby separating the change process that was originally masked by the static state of the stagnant subject from the overall change sequence. After extracting the actual change segments, all actual change segments are organized chronologically. The start and end times of each actual change segment, along with the relevant subject identifiers, are collected one by one. Each actual change segment is then bound to its corresponding abnormal stagnation time, ensuring that each change segment has a clear time range and source location. Subsequently, all actual change segments are organized continuously according to their order in the time series, forming a continuous structure that connects sequentially in the time dimension and maintains a consistent arrangement with the base records in the subject dimension. This results in the unfolding of hidden fluctuations, which fully presents the actual change process that was obscured by the existence of the stagnant subject. This provides continuous and specific data support for subsequent rhythm adjustments to the abnormal stagnation time range.

[0023] Based on the unfolding results of the hidden fluctuations, the time range corresponding to the abnormal stagnation position is adjusted. Within this time range, the data of the main body that has stopped updating is completed according to the overall direction of change, and the static duration interval is shortened so that the overall change continues to advance continuously. After obtaining the unfolded results of the hidden fluctuations, the unfolded results are fused point by point with the continuous time series in the base record. Continuous unfolding and directional adjustment are then performed around the time range corresponding to the abnormal stagnation position to restore the continuity of changes of multiple subjects within that time range. The specific implementation steps are as follows: For each actual change segment in the unfolded result of hidden fluctuations, the start and end time positions of the segment are extracted one by one, and the time range is mapped to the time series in the base record, so that the segment forms a complete coverage relationship in the time series. After the mapping is completed, each actual change segment is bound to the corresponding abnormal stagnation time position, so that the abnormal stagnation time position forms a continuous connection relationship with the subsequent actual change segments in the time series. Then, taking the abnormal stagnation time position as the starting point and the end time position of the corresponding actual change segment as the ending point, each time position in the time series is determined one by one, so that all time positions within the time range are clearly marked, thereby forming a complete time adjustment interval covering the static segment of the main stop-update body. After determining the time adjustment interval, the data changes of each subject at each time point are expanded point by point along the time sequence of the time adjustment interval. During the expansion process, all subject data at each time point are extracted one by one, and the data content of the subject that has stopped updating at that time point is arranged side by side with the data changes of other subjects at the same time point, so that the changes of all subjects at the same time point form a horizontal correspondence. On this basis, the data changes of other subjects between adjacent time points are continuously sorted out, so that the change trajectory of each subject within the time adjustment interval can be continuously expressed at each time point. At the same time, the original data of the subject that has stopped updating within the time adjustment interval is fully recorded, so that the static segment of the subject within the interval is marked point by point in the time series, thus forming a continuous comparison sequence containing a complete time range, a complete set of subjects, and complete change relationships. After forming a continuous control sequence, each time position within the time adjustment interval is used as a processing unit. The data of the subject that has stopped updating at that time position is then completed. During this process, the data content of the subject at the previous time position is first read, and combined with the changes in other subjects at that time position compared to the previous time position, the direction of advancement of the current time position within the overall change sequence is determined. Subsequently, the data of the subject that has stopped updating at the current time position is adjusted according to this direction of advancement, ensuring that the data content of the subject at the current time position forms a continuous connection with the data at the previous time position, and is consistent with the change trends of other subjects at the same time position. After completing the data completion for the current time position, the completed data is written to the corresponding position in the time sequence, and the same process is repeated for the next time position. This ensures that the subject that has stopped updating forms a continuous change sequence at each time position throughout the time adjustment interval, thereby gradually transforming the original static segment into a continuously advancing segment. After completing the data completion process for all time positions within the time adjustment interval, the time series within this interval is comprehensively organized. The completed data of the stagnant main body is recombine with the change data of other main bodies at their corresponding time positions, ensuring a continuous connection between the changes of each main body within this time range. During this organization process, the original continuous, unchanged time positions are replaced one by one with the completed change data, compressing the area occupied by the original static segments in the time series and filling it with continuously changing data, thus creating a continuous progression of overall change within the time adjustment interval. Subsequently, all adjusted time intervals are spliced ​​together in chronological order, creating a unified rhythm of change across multiple main bodies. Finally, the adjustment process for the time range corresponding to the abnormal stagnation positions is completed, and through the continuous completion of the data of the stagnant main body and the compression of static segments, the overall change maintains a continuous progression in the time dimension.

[0024] This invention continuously unfolds the change data of multiple subjects under a unified time series and introduces an update status identifier and a rhythm interruption difference identification process, so that each subject can form a unified time reference relationship under different update rhythm conditions. This avoids the interference of data stagnation caused by individual subjects stopping updates on the overall status judgment, and enables the overall change trend to be expressed based on the continuous change process, thereby improving the coordination consistency and status judgment accuracy of multi-subject data in the same business scenario.

[0025] This invention completes the data of the stopped updating subject within a time range by locating abnormal stagnation positions and combining the unfolding results of hidden fluctuations. This replaces the static segments formed by the interruption of updates with a continuous sequence of changes, thereby restoring the continuous advancement relationship of the overall change in the time dimension. This allows the real change process that was hidden by local stagnation to be presented, improves the completeness of the change process expression, and enhances the responsiveness of subsequent adjustment strategies to the change trend.

[0026] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A scenario-driven, multi-entity data collaborative governance method, characterized in that, Includes the following steps: Collect change data from multiple entities continuously advancing in the same business scenario, organize the data according to a unified time sequence, record the update status of each entity at the corresponding time point in the time series, and form a basic record; Based on the basic records, the update status of each subject at adjacent time points is continuously analyzed to identify the time segments in which the subject stops updating but the overall change continues, thus forming the time distribution results of rhythm interruption differences. Based on the temporal distribution of rhythm interruption differences, the stop update position is mapped back to the time series for comparison of changes before and after, the time position that is inconsistent with the overall change direction is determined, and an abnormal stagnation position record is generated. By combining records of abnormal stagnation locations, a retrospective analysis is performed on the change sequence after the corresponding time location to unfold the slowdown process of change caused by the cessation of updates, identify the actual change segments that are masked by local stagnation, and obtain the unfolding results of hidden fluctuations. Based on the unfolding results of the hidden fluctuations, the time range corresponding to the abnormal stagnation position is adjusted. Within this time range, the data of the main body that has stopped updating is supplemented according to the overall direction of change, and the duration of stagnation is shortened.

2. The scenario-driven multi-subject data collaborative governance method according to claim 1, characterized in that, The basic record creation steps are as follows: As the business scenario progresses, data on changes in multiple entities are collected hourly and a unified time identifier is attached. The data on changes are arranged in the order of the time identifiers to form a continuous time series structure. The empty space identifiers of the entities are retained in the time position to maintain the consistency of the entity positions. Based on the continuous time series structure, the update status of each subject's change data at each time position is determined, an update status identifier is added to the time position, and the update status sequence of each subject is recorded along the time series, so that the update status sequence corresponds one-to-one with the change data. Based on the update state sequence, the update state of each subject is expanded segment by segment, the update states between adjacent time positions are connected to form update segments and static segments, and the update segments are bound with the changed data to form segment representations. Based on the segment expression, the time series is uniformly collected and the distribution of each subject's update segment and static segment is summarized to form the subject's rhythm trajectory and record the update continuity, thus obtaining a record of the change rhythm and progress.

3. The scenario-driven multi-subject data collaborative governance method according to claim 2, characterized in that, The steps for generating the temporal distribution results of rhythm interruption differences are as follows: The time position is expanded point by point along the time series direction, the update status identifier of each subject is extracted and the correspondence between adjacent time positions is established, the update status changes are compared to determine the starting time position of the interruption of the update behavior and form a continuous tracking sequence. The update status of subsequent time positions is continuously read around the starting time position. The update status of the corresponding subject is compared with the update status of other subjects, and the combination is used to form a continuous time interval. The subject identifier and time range are determined to obtain the candidate difference interval. Based on the candidate difference interval, the time range is mapped to the changing data sequence, the time position change data is expanded and the change is compared to determine the correspondence between continuous change and static state, thus forming the rhythm interruption difference interval. The time positions are sequentially arranged according to the rhythm interruption difference intervals, and the associated subject identifiers and time ranges form a continuous distribution structure to obtain the time distribution results of rhythm interruption differences.

4. The scenario-driven multi-subject data collaborative governance method according to claim 3, characterized in that, Determining the rhythm interruption difference interval involves continuously comparing the data changes at each time position within the candidate difference interval, identifying the time interval where the subject maintains unchanged data at adjacent time positions while other subjects show continuous changes, and recording the subject identification and time range association for the time interval, thereby limiting the formation conditions of the rhythm interruption difference interval.

5. The scenario-driven multi-subject data collaborative governance method according to claim 4, characterized in that, The steps for generating abnormal stationary location records are as follows: Extract the starting time position of the time interval from the time distribution results and map it to the time series. Select the previous and next time positions around the starting time position and extract the change data of each subject to form a continuous data segment. Based on continuous data segments, the changes in the subject that has stopped updating are compared between the previous time position and the current time position. Simultaneously, the changes in the remaining subjects are extracted and arranged in parallel to determine the candidate positions for separation of change relationships and bind the subjects. Starting from the candidate position, select subsequent time positions along the time series, continuously compare the change data of each subject, determine the time position where the subject that stops updating remains unchanged and the other subjects continue to change as the inconsistent time position, and associate it with the continuous time range. The inconsistencies in time and location are sorted sequentially, and the main identifiers, time intervals, and changes are collected to form a record of abnormal stagnation locations.

6. The scenario-driven multi-subject data collaborative governance method according to claim 5, characterized in that, By continuously extracting the change data of the subject that stopped updating at each time point within the continuous time range corresponding to the inconsistency time point, and comparing the data of the subject at adjacent time points with the change data of other subjects, the inconsistency time point is further defined based on the distribution of change relationships within the continuous time range.

7. The scenario-driven multi-subject data collaborative governance method according to claim 5, characterized in that, The steps to obtain the hidden wave expansion result are as follows: Time series location is performed around the abnormal stagnation location record and the corresponding time position is extracted. Subsequent time positions are selected along the time series and the change data of each subject are extracted. Combined with the subject that stops updating, a continuous change sequence is formed. Based on the continuous change sequence, the change data at each time position is expanded, and the main data that stops updating is compared with the other main data. The time range in the continuous time position where the main data that stops updating remains unchanged and the other main data changes is identified as the change slowdown segment. By mapping the slow-down segments to the time series and performing reverse backtracking, the change data at each time point is extracted and continuously compared to determine the time points where the main body stops updating and remains unchanged while the other main bodies continue to change, thus forming the actual change segments. The actual change segments are sorted sequentially and associated with records of abnormal stagnation locations to form a continuously arranged hidden fluctuation unfolding result.

8. The scenario-driven multi-subject data collaborative governance method according to claim 7, characterized in that, The steps for adjusting the time range corresponding to the abnormal stop location are as follows: The actual change segments are extracted from the unfolding results of the hidden fluctuations and mapped to the time series. The actual change segments are then bound to the abnormal stagnation positions, and the time adjustment interval is determined. The data changes at each time point are expanded around the time adjustment interval, the data changes of each subject are extracted and arranged side by side to form a continuous comparison sequence that includes the static section of the subject that stops updating and the change trajectory of other subjects. Based on the continuous comparison sequence, the data of the subject that stopped updating was completed point by point at each time position, and a continuous change sequence was formed by combining the data of the previous time position with the changes of other subjects. Based on the completion processing results, the time adjustment intervals are organized, static segments are replaced with continuously changing data and sequentially spliced ​​to form a time series expression of continuous overall change.

Images