Report dynamic merging generation method, system and device under multi-dimensional business perspective

Abstract

The application provides a report dynamic merging generation method, system and equipment under a multi-dimensional business perspective, relates to the technical field of report generation, and comprises the following steps: performing multi-dimensional business perspective cross-business time sequence correlation analysis according to an original business event stream, and constructing M cross-business cooperative collection threads; performing periodic data collection by using M time sequence data field characteristics driving based on thread execution time sequence constraints, and obtaining M single-period cross-perspective time sequence data packets; performing multi-dimensional confidence cross-validation according to a correlation key mapping relationship, and obtaining M single-period verification time sequence data fields; and after obtaining a dynamic fusion matrix report, storing the dynamic fusion matrix report to a time sequenced incremental report warehouse for periodic incremental updating. The application solves the technical problem that the prior art lacks sufficient support in the aspects of cooperative work and data interaction between multiple business systems, data is missing, lost or out of date, and the accuracy and timeliness of a report are affected.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of report generation, in particular to a report dynamic merging generation method, system and device under multi-dimensional business perspectives. BACKGROUND

[0002] Enterprises usually need to integrate data from multiple business systems for comprehensive analysis to help decision-makers make accurate judgments, especially in the aspect of cross-business process and cross-departmental data collaborative collection, the traditional data report generation method has many limitations. The existing technology is usually mature in single business system or single perspective data collection, but lacks sufficient support in collaborative work and data interaction between multiple business systems. The data under each business perspective often has different timing requirements and periodic characteristics, especially when cross-business collaboration, the data may have problems such as asynchronization and delay. Due to the lack of accurate timing control and cross-business perspective synchronization mechanism, the existing technology often leads to data loss, missing or outdated, thereby affecting the accuracy and timeliness of the report. SUMMARY

[0003] The present application provides a report dynamic merging generation method, system and device under multi-dimensional business perspectives, aiming to solve the technical problem that the existing technology lacks sufficient support in collaborative work and data interaction between multiple business systems, leading to data loss, missing or outdated, thereby affecting the accuracy and timeliness of the report.

[0004] The first aspect of the present application provides a report dynamic merging generation method under multi-dimensional business perspectives, the method comprising: performing multi-dimensional business perspective cross-business timing correlation analysis according to the original business event stream, and constructing M cross-business collaborative collection threads; based on thread execution timing constraints, using M timing data field characteristics to drive the M cross-business collaborative collection threads to perform periodic data collection of the multi-dimensional business perspective, and obtaining M single-period cross-perspective timing data packets; performing multi-dimensional confidence cross-validation of the M single-period cross-perspective timing data packets according to the association key mapping relationship of the M cross-business collaborative collection threads, and obtaining M single-period validation timing data fields; merging the M single-period validation timing data fields to obtain a dynamic fusion matrix report, and then dynamically storing the dynamic fusion matrix report to a timing incremental report warehouse for periodic incremental update of the timing incremental report warehouse.

[0005] In a second aspect, the application discloses a report dynamic merging generation system under a multi-dimensional business perspective, which is used for the report dynamic merging generation method under the multi-dimensional business perspective. The system comprises a correlation analysis module, which is used for performing cross-business time sequence correlation analysis of the multi-dimensional business perspective according to an original business event stream, and constructing M cross-business cooperative collection threads; a data collection module, which is used for driving the M cross-business cooperative collection threads by using M time sequence data field characteristics based on thread execution time sequence constraints, performing periodic data collection of the multi-dimensional business perspective, and obtaining M single-period cross-perspective time sequence data packets; a cross verification module, which is used for performing multi-dimensional confidence cross verification of the M single-period cross-perspective time sequence data packets according to a correlation key mapping relationship of the M cross-business cooperative collection threads, and obtaining M single-period verification time sequence data fields; and an incremental update module, which is used for merging the M single-period verification time sequence data fields, obtaining a dynamic fusion matrix report, dynamically storing the dynamic fusion matrix report into a time-sequenced incremental report warehouse, and performing periodic incremental update of the time-sequenced incremental report warehouse.

[0006] In a third aspect, the application discloses a computer device, which comprises a memory and a processor. The memory stores a computer program, and the processor executes the computer program to realize the steps of the report dynamic merging generation method under the multi-dimensional business perspective.

[0007] The one or more technical solutions provided in the application have at least the following beneficial effects:

[0008] By analyzing the cross-business time sequence correlation of the original business event stream, the mutual correlation and dependence between multiple businesses can be mined, and by constructing cross-business collaborative collection threads, the data retrieval of each business perspective can be synchronized and mutually non-interfering. This design avoids the limitations of data extraction under the traditional single perspective, so that multiple business perspectives can be processed at the same time, enhancing the comprehensiveness and comprehensiveness of the data; through the multi-thread design based on time sequence constraints, multiple cross-business collaborative collection threads can independently and in parallel collect data, the periodic data collection of each thread effectively avoids the bottleneck problem of a single time point, and can improve the efficiency and stability of data collection, the periodic retrieval mechanism ensures the persistence and real-time of the data, so that the business data can be continuously updated as time goes on; through multi-dimensional cross verification of the data packets obtained by each thread, the accuracy and consistency of the data can be efficiently verified, through the association key mapping relationship, data verification and cross verification are performed to ensure the integrity and accuracy of each data packet, avoiding data redundancy and abnormal problems, and enhancing the reliability of the data after report merging; by merging single-period verification time sequence data fields from different perspectives, a dynamic fusion matrix report is generated, which guarantees the comprehensiveness of the data and the fusion of multiple perspectives, and the dynamic fusion matrix report is dynamically stored in the time sequence incremental report warehouse, further improving the timeliness and real-time of the data. Compared with the traditional static report, this dynamic generation and incremental update mechanism enables the report to be adjusted and updated in a timely manner as the business data changes, ensuring that the final report always matches the current data state.

[0009] The above description is only a summary of the technical solutions of the present application. In order to more clearly understand the technical means of the present application, the content of the specification can be implemented, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described. BRIEF DESCRIPTION OF DRAWINGS

[0010] Figure 1 A multi-dimensional business perspective report dynamic merging generation method flowchart is provided for the embodiments of the present application.

[0011] Figure 2 A multi-dimensional business perspective report dynamic merging generation system structure diagram is provided for the embodiments of the present application.

[0012] Figure 3 A structure diagram of an exemplary computer device is provided for the embodiments of the present application.

[0013] BRIEF DESCRIPTION OF DRAWINGS: Correlation analysis module 10, data collection module 20, cross verification module 30, incremental update module 40, processor 21, memory 22, input device 23, output device 24. DETAILED DESCRIPTION

[0014] To further clarify the technical means and effects taken by the present application to achieve the predetermined inventive purpose, the specific embodiments, structures, features and effects thereof according to the present application are described in detail below in conjunction with the drawings and preferred embodiments.

[0015] Embodiment one, as shown in the present application, provides a report dynamic merging generation method under multi-dimensional business perspective, which comprises: Figure 1

[0016] According to the original business event stream, the cross-business time sequence correlation analysis of the multi-dimensional business perspective is performed, and M cross-business cooperative collection threads are constructed.

[0017] The original business event stream refers to the business data stream generated by different business systems, including multiple dimensions of data such as timestamp, event type, business execution state, data field, etc. Business data can be analyzed from multiple dimensions, such as from business process, time sequence, event type, user behavior, etc. from multiple perspectives. Cross-business time sequence correlation analysis is to find the correlation between different business events, especially their mutual relationship on the time axis, for example, the payment event of an e-commerce order and the distribution event in the logistics system have a time sequence correlation, or have a relationship with the commodity state change in the inventory management system. By analyzing the time sequence data between different business systems, the time sequence relationship of the original business event stream is established, which can be completed using data mining, time series analysis or graph model technology, for example, based on timestamp, the event stream is sorted, and the closely connected business events in time are identified using association rule mining method. On the basis of cross-business time sequence correlation analysis, M cross-business cooperative collection threads are constructed, M is a positive integer, each thread collects data according to the specified time sequence constraint according to the result of time sequence analysis, these threads have a clear working period, and cross collection of data is performed according to business logic.

[0018] Based on the thread execution time sequence constraint, M cross-business cooperative collection threads are driven by M time sequence data field characteristics to perform periodic data collection of the multi-dimensional business perspective, and M single-cycle cross-perspective time sequence data packets are obtained.

[0019] The execution of each collection thread is based on a specific thread execution time sequence constraint, which means that each collection thread is not working at any time, but collecting data according to the preset rules and time interval, for example, a certain collection thread collects data every minute, and another thread collects data every hour.

[0020] ​The data collected by each collection thread has specific time sequence data field characteristics, including timestamp (e.g. order payment time), event type (e.g. order creation, order payment, etc.), user or entity identification (e.g. user ID, product ID), etc., which determine the data collection range and strategy of each collection thread. Based on these characteristics, each thread collects data at a specified time point or time window.

[0021] Periodic data collection means that each cross-business collaborative collection thread collects data within its execution cycle, which can be fixed or dynamically adjusted according to business changes. For example, in the order management thread, relevant order data can be collected every hour, while the payment system thread collects data every minute. The data collected by each thread is organized and stored in the form of single-cycle cross-perspective time sequence data packets, each of which includes relevant data obtained from different business perspectives and is sorted in chronological order.

[0022] According to the association key mapping relationship of the M cross-business collaborative collection threads, multi-dimensional confidence cross-validation of the M single-cycle cross-perspective time sequence data packets is performed to obtain M single-cycle validation time sequence data fields.

[0023] In cross-business collaborative collection threads, data from different business perspectives needs to be associated so that they can be effectively compared and validated in the same time sequence data packet. Association keys are fields that can associate data across business systems, such as order ID, user ID, product ID, etc. Association key mapping relationship is used to identify the mapping relationship between data from different threads, ensuring that data from different business systems can be correctly matched within the same cycle.

[0024] During data collection, data packets from different business systems may not be consistent. To ensure data reliability, cross-validation is needed. Cross-validation is a check of data accuracy in multiple dimensions, such as time dimension, business perspective dimension, etc. Confidence is a measure of data reliability. In multi-dimensional cross-validation, data packets with high confidence can be used as trusted data, while data packets with low confidence need to be further verified or removed.

[0025] After multi-dimensional confidence cross-validation, the data of each thread is verified, and the final output is M single-cycle validation time sequence data fields, which contain data confirmed to be valid within a validation cycle.

[0026] After merging the M single-period validation timing data fields, a dynamic fusion matrix report is obtained, which is dynamically stored in a time-series incremental report warehouse for periodic incremental updating of the time-series incremental report warehouse.

[0027] When all threads have completed data validation, the validated M single-period validation timing data fields are merged. The merging method is to combine validation data from different perspectives according to time sequence, business relevance, etc. to form a unified report structure, and the obtained dynamic fusion matrix report is a table form containing all multi-dimensional data. Each row represents a time period, and the column represents different business perspectives or data fields. The data in the matrix is the validation result from each thread.

[0028] The merged dynamic fusion matrix report is dynamically stored in a time-series incremental report warehouse, which is a database or storage system specially designed to process time series data. It can efficiently store, retrieve and update timestamp-based data. The time-series incremental report warehouse is periodically updated, which means that each update only adds newly generated data, rather than recalculating the entire report. Incremental updates are usually performed through time windows, such as hourly, daily, weekly, etc. This ensures that cross-business perspective data can be efficiently validated and merged within each period, and the data report can be updated at any time to ensure the real-time and accuracy of business decisions.

[0029] Further, the method further comprises:

[0030] According to the cross-business association relationship, the business association key of the multi-dimensional business perspective is identified to obtain a plurality of groups of business association keys. The original business event stream is taken as an input source, and cross-business timing correlation analysis of the plurality of groups of business association keys is performed to construct the M cross-business cooperative collection threads, wherein the M cross-business cooperative collection threads are subject to thread execution timing constraints. According to the M cross-business cooperative collection threads, the original business event stream is aggregated to perform dynamic extraction of data field features, and the M timing data field features are output.

[0031] The cross-business association relationship refers to the data association relationship between different business systems, for example, there is an association key relationship between the order system and the payment system of an e-commerce platform, or there may be an association key relationship between the e-commerce platform and the logistics system. Each business perspective is based on a different business domain perspective to analyze data, such as order management, payment processing, logistics distribution, etc. The business association key is a field used to associate data under different business perspectives, which spans different business systems and is used to identify the same business object in different processing stages of the same business event. The business association key can be an order ID, a user ID, a product ID, a payment ID, a delivery number, etc. By analyzing the data flow in different business systems and combining domain knowledge, it is identified which fields can be associated between different business systems to meet the cross-business association between different business perspectives, for example, an e-commerce platform involves multiple systems, including order management, payment system, logistics system, each system has different association keys, and the combination is multiple sets of business association keys.

[0032] The original business event stream refers to the event data stream obtained from different business systems, and the cross-business time sequence association analysis refers to the analysis of the time sequence dependency relationship between the event data across multiple business systems, for example, the order payment event and the delivery event are time sequence dependent on each other, that is, after the payment event occurs, the delivery event will be triggered. By mapping the association keys of different business systems to the same business object, the time sequence relationship between events in different systems is identified, and through time sequence analysis, it can be determined which event occurs first and which event occurs later, and a basis is provided for subsequent data collection and processing. The cross-business cooperative collection thread is designed according to the time sequence association analysis result, and multiple parallel running collection threads are designed for different business perspectives, each collection thread is responsible for periodically collecting data from a certain business perspective. Each collection thread is subject to a specific thread execution time sequence constraint according to the result of the time sequence association analysis, and has a clear periodic collection task, the execution time sequence and task of each thread will be adjusted based on the cross-business time sequence association, for example, the thread of the payment system is executed according to the time sequence of the occurrence of the payment event, and the thread of the logistics system depends on the time window after the order payment is completed to collect data.

[0033] Each cross-business cooperative collection thread aggregates the business event stream from different business systems from its own business perspective according to the analyzed time sequence association relationship, that is, the data from different source systems is integrated in the corresponding dimension, for example, the data in different business systems is aggregated together through the order ID to form a complete order life cycle view.

[0034] In the original service event stream after aggregation, key data field features are dynamically extracted, which are the core dimensions of business data, including timestamp (representing the occurrence time of the event), status field (such as payment status, order status, etc.), identifier field (such as order ID, payment ID, product ID, etc.), and the extraction of these data field features is flexible, based on the changes of the current data stream and the needs of analysis, the extracted data fields will be dynamically adjusted, for example, in a certain business perspective, more attention is paid to the payment amount, while in another perspective, more attention is paid to the order status change.

[0035] Further, according to the cross-business association relationship, the business association key of the multi-dimensional business perspective is identified, and a plurality of groups of business association keys are obtained, and the method comprises:

[0036] The plurality of groups of business execution keys of the multi-dimensional business perspective are called, and the topology connection of the plurality of groups of business execution keys is constructed according to the cross-business cross relationship of the multi-dimensional business perspective, and an entity interaction topology is obtained; the multi-business system source data in the message format is called; according to the multi-business system source data, the execution key association attribute analysis of the entity interaction topology is carried out, and the topology connection is valued according to the analysis result, and an attribute injection topology is obtained; in the attribute injection topology, the node association centrality is calculated according to the topology connection weighting result, and a plurality of groups of association strength quantization values of the plurality of groups of business execution keys are obtained; a dynamic coverage threshold is preset to traverse the plurality of groups of association strength quantization values, and the plurality of groups of business association keys are mapped and screened from the plurality of groups of business execution keys.

[0037] Multi-dimensional business perspective refers to analyzing business data from multiple different business perspectives, and each business perspective determines key data fields according to its own specific needs, such as order management system focusing on order ID, customer information, product information, etc., and payment system focusing on payment ID, transaction amount, payment time, etc. Business execution key refers to the field that can uniquely identify business operation or event in each business perspective, which is usually the identifier of some specific business activity, such as order ID, user ID, payment ID, product ID, etc.

[0038] Cross-business cross relationship refers to the data association method between different business perspectives, and different business systems are connected by sharing some common fields, for example, order ID in e-commerce system can cross multiple systems. Topology connection refers to the association network between different business perspectives established by business execution keys, each business execution key (such as order ID) represents a node, and different business systems are connected through these business execution keys to form a graph-like topology structure, i.e. entity interaction topology, which represents the mutual dependence and data transmission relationship between different business systems.

[0039] Message format refers to the data extracted from different business systems is usually in some standardized message format, message format can be JSON, XML, CSV, etc., depending on the design of business systems and interfaces, each system will format data according to its own needs and standards. Access to multiple business system data sources to ensure that cross-business perspective data can be fully collected and integrated, these data sources contain data accessed from different business systems.

[0040] Performing key association attribute parsing is to parse the business objects represented by business execution keys and their attribute relationships. Each business execution key is not only a simple identifier, but also contains multiple attributes such as timestamp, status code, and amount. These attributes determine the effectiveness of cross-business system interaction. In the parsing process, these attributes are extracted from multi-business system source data and mapped to the corresponding nodes and edges in the topology. The purpose of topology edge assignment is to ensure that each edge not only represents a simple association relationship, but also has actual business attribute information. Each node represents a business execution key and needs to be assigned relevant attributes based on the parsing results. Each edge represents the relationship between different business perspectives and needs to reflect the relationship between nodes through assignment, such as time delay and dependency strength. The final result of attribute injection topology is an enhanced entity interaction topology, i.e. attribute injection topology, which not only describes the relationship between different business perspectives, but also injects business attributes into each node and edge, providing more information to support subsequent calculation and verification.

[0041] Topology edge weighting result refers to assigning a weight to each topology edge, and the weight is assigned according to different business attributes or relationship strength. Weighting can be based on multiple factors, such as time delay, the delay time between payment and logistics may affect the association strength between the two business systems; based on dependency relationship, some business systems may have stronger dependence on other systems, so the weight of the edge between them will be larger; based on frequency, if a business event (such as payment event) frequently associates with other business events, its association strength should be higher.

[0042] Node association centrality is an indicator to measure the relative importance of a node in a topology graph. By calculating the association centrality of each node, it can be determined which business execution key is crucial to the data flow of the entire system. For example, by calculating the number of direct associations of a certain node, assuming that an order ID is associated with many payment IDs and logistics order numbers, the node association centrality of this order ID is high. The association strength quantification value is a value obtained by calculating the node association centrality, which is used to reflect the importance and influence of different business execution keys in the entire business process. The execution key with a high strength quantification value may be the core node of cross-business perspective data flow, with strong dependency or influence.

[0043] The dynamic coverage threshold refers to a flexible threshold set according to different situations when traversing the association strength quantification value, which is used to determine which business execution key is critical or high-impact. Based on the set dynamic coverage threshold, multiple sets of association strength quantification values are traversed to filter out business execution keys with high association strength. The business association key obtained by filtering refers to those business identifiers with high association between cross-business systems. In this process, by analyzing the association strength quantification value, key key values with strong cross-association in multiple business perspectives are filtered out.

[0044] Further, taking the original business event stream as the input source, the cross-business time sequence association analysis of the multiple sets of business association keys is performed to construct the M cross-business cooperative collection threads. The method comprises:

[0045] Based on the spatiotemporal resolution strategy, the key value slicing of the original business event stream is performed to obtain H event spatiotemporal slicing sets. The multiple sets of business association keys are used to traverse the H event spatiotemporal slicing sets to obtain H cross-perspective association key clusters. According to the timestamp ascending order arrangement of the H event spatiotemporal slicing sets, the causal dependence chain identification of the H cross-perspective association key clusters is performed to output H business causal dependence chains. Based on the H business causal dependence chains, the association path pattern mining is performed to generate M association key transfer chains. The M association key transfer chains are used to backtrack the original business event stream to perform high-frequency event time sequence characteristic identification to obtain M strong time sequence constraint paths. The M association key transfer chains and M strong time sequence constraint paths are dynamically topological instantiated to output the M cross-business cooperative collection threads.

[0046] The spatiotemporal analysis strategy refers to a method of splitting and organizing the business event stream according to the time dimension and the space dimension, and the goal is to reasonably segment the original business event stream according to the time sequence (time dimension) and business scope (space dimension), so as to facilitate subsequent cross-business analysis. Each event in the original business event stream usually contains several key values, such as order ID, payment ID, user ID, etc., which play a key role in associating different business systems. Key value fragmentation refers to splitting the original business event stream according to these key values. For example, the original business event stream is split according to a time window, such as every hour, every day, or according to a business perspective, such as order, payment, logistics, etc. The fragmented data is organized into H event spatiotemporal fragment sets, where H is a positive integer, and each event spatiotemporal fragment set represents a collection of events in a certain time period and under a specific business perspective.

[0047] In cross-business perspective analysis, business association keys are the basis for associating data from different business systems. Multiple sets of business association keys are used to traverse the H event spatiotemporal fragment sets. Within each event spatiotemporal fragment set, according to the cross-perspective relationship of the multiple sets of business association keys, related events in different business systems can be identified. For example, within the same time range, an order ID is associated with multiple payment events and logistics events. Cross-perspective association key clusters refer to a set of interrelated event keys identified within each event spatiotemporal fragment set based on business association keys. Each cluster represents associated data under different business perspectives. Cross-perspective means that the data in a cluster comes from different business systems, reflecting the data interaction relationship between systems.

[0048] In cross-business perspective analysis, different events usually have different timestamps, and they occur or exist in time order. Ascending order of timestamps ensures that the causal relationship between events can be correctly identified in the analysis process according to the time order. If the timestamp order of event occurrence is inconsistent, it may lead to incorrect identification of causal relationships.

[0049] Each cross-perspective association key cluster contains a set of associated events occurring within a time window. By analyzing the timestamps, attributes, and state information between these events, the causal dependency relationship between them can be identified. For example, the payment event of an order depends on the order creation event, and the payment event causes the logistics delivery event. Through causal chain identification, it can be understood how each event affects or depends on other events. Ultimately, through the identification of causal relationships in cross-perspective association key clusters, H business causal dependency chains are generated. These business causal dependency chains describe how one event triggers another event or how one business system depends on the events of another business system in the entire business process.

[0050] Correlation path pattern mining refers to correlation analysis based on H business causal dependency chains, identifying common path patterns among them. In this case, the correlation path pattern refers to the common dependency sequence between multiple business systems or events, for example, in an e-commerce system, a common path pattern is order creation to payment to delivery. By mining these common patterns of causal relationships, M correlation key transfer chains can be generated, each of which describes the sequence and dependency of a series of events in time.

[0051] Using the generated M correlation key transfer chains, trace back to the original business event stream, trace back means to trace these correlation key transfer chains in the direction of time sequence, to identify the mutual influence between business events at different time points. Through the trace back, it can be verified whether these events occur according to the predetermined business causal relationship, and whether it conforms to the expected timing pattern. In the process of trace back, the high-frequency event timing characteristics of the event stream are identified, that is, the frequently occurring event patterns or dependency between events in a period of time are found out, which specifically includes identifying which events occur more frequently, and which events have strong timing constraints on other events, for example, in the typical path of order creation to payment to delivery, there is a strong timing constraint between order creation and payment, while the delivery event occurs relatively less frequently. After identifying the timing characteristics of high-frequency events, M strong timing constraint paths are extracted, which represent which events have strong timing dependencies on other events from a cross-business perspective. These strong timing constraint paths define the timing rules that must be followed in the cross-business collaborative data collection process.

[0052] Topology instantiation refers to mapping the M correlation key transfer chains and M strong timing constraint paths that have been identified into a dynamic topology structure through certain rules. The topology structure represents the dependency between different events and data, similar to a graph, where nodes represent events and edges represent the dependency between events. Dynamic topology instantiation refers to instantiating these paths according to real-time business conditions and timing requirements, forming a dynamic and variable business process network, for example, if the execution speed of a certain business event slows down, the topology will adjust the data flow path to respond to the change.

[0053] By instantiating the M correlation key transfer chains and the M strong timing constraint paths, M cross-business collaborative collection threads are created, which will be responsible for performing cross-business data collection tasks in sequence in the system.

[0054] Further, the method further comprises:

[0055] statistically distribute M execution time distributions of the M associated key transfer chains in the original business event stream; time sequence splice the M execution time distributions for task delay analysis, and output rigid execution delay threshold and elastic execution tolerance interval; inject the rigid execution delay threshold and the elastic execution tolerance interval into node links of the M cross-business cooperative collection threads to generate thread execution time sequence constraints.

[0056] The execution time distribution refers to the execution characteristics of the M associated key transfer chains on the time axis. The execution of each associated key transfer chain may not be linear, and there may be different delays, blockages, or concurrent executions. The purpose of statistical execution time distribution is to analyze the timeliness of each associated key transfer chain in the execution process. Specifically, the execution of each associated key transfer chain in the original business event stream is statistically analyzed, mainly to analyze the execution performance of these transfer chains at different time points. For example, some transfer chains execute quickly in some time periods, while other transfer chains may be affected by system load, network delay, or other factors, and take longer to execute. Through the statistical results, the execution time distribution of each associated key transfer chain is obtained, i.e., the time, frequency, and delay of each chain.

[0057] Time sequence splicing refers to merging the execution time distribution of each associated key transfer chain in chronological order. By splicing these execution time distributions, a complete time sequence can be formed to represent the time sequence relationship between different associated chains. Task delay analysis is a further analysis of the spliced execution time distribution to evaluate the delay of tasks in different time periods. Delay not only includes the delay of individual events, but also includes the delay across multiple business perspectives. Key analysis points include: whether the execution time of each associated chain meets expectations; whether there is a large delay fluctuation, or the delay of some chains is too high, which may cause the timeout of the entire task; whether the delay between business events affects the accuracy and timeliness of overall data collection.

[0058] Based on the task delay analysis, the rigid execution delay threshold and the elastic execution tolerance interval are output. The rigid execution delay threshold refers to the time range within which certain key tasks or events must be completed during data collection, otherwise the entire business process will be affected. For example, if the time difference between order creation and payment exceeds the set threshold, the order will be invalidated. The elastic execution tolerance interval refers to the delay tolerance of certain tasks or events, but still needs to be completed within a reasonable time. These tasks are important but have stronger tolerance to delay, allowing a certain delay.

[0059] The rigid execution time delay threshold and the elastic execution tolerance interval are applied to the node links of the cross-business collaborative collection thread, the node links refer to the relationship between each task or data processing step in the thread, each node represents a task, and each link represents the dependency relationship between tasks. The generated thread execution time sequence constraint defines the timeliness requirement between tasks in the data collection process, ensuring that the tasks are completed in a specific time sequence and timeliness, wherein the rigid time sequence constraint forces certain tasks to be completed within a specific time, otherwise the entire task fails or is considered abnormal; the elastic time sequence constraint allows the task to be completed within a certain time tolerance interval, providing the system with a certain flexibility to cope with unpredictable delays.

[0060] Further, according to the association key mapping relationship of the M cross-business collaborative collection threads, multi-dimensional confidence cross-validation of the M single-period cross-perspective time sequence data packets is performed to obtain M single-period verification time sequence data fields, and the method comprises:

[0061] The M association key mapping chains of the M association key transmission chains are collected from the attribute injection topology; the data confidence priority of the M cross-business collaborative collection threads is predefined; cross-business cross-validation is performed on the M single-period cross-perspective time sequence data packets according to the M association key mapping chains, and confidence decision of abnormal data packets is made by using the data confidence priority to obtain the M single-period verification time sequence data fields.

[0062] The association key mapping chain of each association key transmission chain is extracted from the attribute injection topology, and the association key mapping chain describes the conversion process of the association key from one business perspective to another business perspective, for example, in the order and payment business perspectives, the order ID and payment ID are the key fields of the mapping relationship. Through these association key mapping chains, it can be identified how the data of each business perspective is associated and mapped, ensuring the correct matching of multi-dimensional data flow.

[0063] In the cross-business data collection process, the reliability and accuracy of different data sources and business perspectives may differ, in order to ensure the quality of the final generated report, it is necessary to define the data confidence priority of each cross-business collaborative collection thread, the data confidence priority is a weight value representing the credibility of the data source or business perspective in data verification, the data source with high confidence priority represents more reliable data, which should be given priority in data verification. Predefinition refers to setting the data confidence priority of different business perspectives and data sources during the system startup or configuration phase, for example, the data of the core system (such as the payment system and the order system) is set to high priority, while the priority of the peripheral system or external data is lower.

[0064] The accuracy of the data is verified by performing cross-service cross-validation on M single-period cross-view timing data packets. Cross-validation is performed by comparing and verifying timing data from different service perspectives to check whether the data meets the expected logical relationship and timing requirements. During cross-validation, the reliability of each data packet is determined based on predefined data confidence priority. If a data packet comes from a low-priority data source, it will be handled more carefully, such as requiring additional verification or data correction. Confidence decision refers to the comparison of data from multiple data sources and business perspectives. Based on the confidence priority of each data source, the data packet is judged. If the data packet has low credibility, the data is discarded or marked as abnormal. Finally, after cross-validation and confidence decision, M single-period verification timing data fields are output. These fields are verified and confirmed to be abnormal data, which are used for subsequent data merging, dynamic report generation and incremental updating.

[0065] Further, according to the multi-service system source data, the execution key association attribute of the entity interaction topology is parsed, and the topology connection assignment is performed according to the parsing result, and the attribute injection topology is obtained. The method comprises:

[0066] The first timing service association record, the first timing association exception record, the first transmission delay record and the first association frequency record of the first service execution key pair are retrieved from the multi-service system source data. The exception rate of the first timing service association record and the first timing association exception record is calculated, and the first stability quantization value is output. The service dependence strength quantization is performed based on the first timing association exception record and the first association frequency record, and the first dependence quantization value is output. The service timing compliance analysis of the first transmission delay record is performed by calling the preset timing fault tolerance window, and the first timeliness quantization value is output. The first stability quantization value, the first dependence quantization value and the first timeliness quantization value are weighted, and the first topology connection assignment is output.

[0067] The first business execution key pair related data is retrieved from the multi-service system source data, and the business execution key pair refers to key identifiers in different business perspectives, which are used to associate cross-service data. Specifically, the first time-series business association record records the time sequence association between different business systems, which is usually the relationship of related data in two or more business perspectives within a certain time period. For example, the order creation event in the order system and the payment completion event in the payment system are associated in time. The first time-series association exception record refers to the data exception found in the cross-service data association process. These exceptions usually manifest as time inconsistency, data loss, field mismatch, etc. Such records identify errors or inconsistencies in the cross-service data chain, which may cause errors in subsequent report generation. The first transmission delay record reflects the possible delay in the cross-service system data transmission process. For example, after payment is completed, there is a delay in data transmission to the order system, which will affect the time sequence consistency of the data. The first association frequency record describes the frequency of a certain association within a certain time range. For example, the association of a certain order ID and payment ID is once per second or once per minute. Changes in frequency may affect data quality and system performance.

[0068] The first time-series business association record and the first time-series association exception record are subjected to an exception rate calculation. The exception rate refers to the proportion of abnormal events within a certain time range. For example, if 10 out of 100 association records are abnormal, the exception rate is 10%. The first stability quantitative value is a measure of the stability of time-series data. A stable business association data stream should have a low exception rate and no significant fluctuations in data transmission and processing. Through the calculation of the exception rate, the first stability quantitative value can be obtained, representing the stability of a certain business system or cross-service data stream.

[0069] Business dependency strength quantification is a measure of the mutual dependency between different business perspectives or data sources. The association between some business systems may be closer, such as the association between order and payment systems, while the association between some systems may be looser. The first time-series association exception record reflects the frequency of abnormal association between businesses, and the first association frequency record reflects the normal association frequency within a certain time. By combining these two records, the first dependency quantitative value is calculated, which represents the degree to which a certain business system depends on the data of other business systems. Specifically, if the association frequency of a certain business system is high and the association exception is less, the dependency strength between it and other business systems is high. Conversely, if the association frequency is low or the exception record is more, the dependency strength is lower. The first dependency quantitative value represents the dependency relationship between different business systems. High dependency means that the data relationship between cross-service processes is relatively close.

[0070] The preset timing fault tolerance window is a maximum time deviation interval allowed in the data transmission process. During the cross-system data flow process, the timeliness of data transmission and processing may be affected due to different processing speeds of different business systems. The preset timing fault tolerance window helps determine which time deviations are acceptable and which are abnormal beyond expectations. The first transmission delay record reflects the time delay experienced by the data during transmission from one system to another. This delay can be caused by multiple factors such as network delay, system processing capacity difference, etc. By analyzing the first transmission delay record, it is determined whether the data is within the preset timing fault tolerance window. If the transmission delay of the data exceeds the fault tolerance window, it is considered that the data violates the business timing compliance, which may cause problems in subsequent processing. The first timeliness quantitative value measures whether the data transmission meets the timing requirements. If the data completes transmission within the compliant timing fault tolerance window, the first timeliness quantitative value is higher, otherwise it is lower.

[0071] The first stability quantitative value, the first dependency quantitative value, and the first timeliness quantitative value are weighted to calculate a comprehensive evaluation index, and finally obtain the first topology link assignment. The first topology link assignment defines the data flow relationship between business systems, not only reflecting the dependency between each system, but also reflecting their relative importance and timeliness in timing. Through the weighted processing of these quantitative values, the relationship between data flow and business systems can be evaluated in multiple dimensions according to different indicators. The weight is set according to business requirements, data importance, or system performance goals. For example, if business timeliness is crucial, the first timeliness quantitative value can be assigned a higher weight; if data stability is the focus, the weight of the first stability quantitative value can be higher.

[0072] Further, each cross-business collaborative collection thread includes an associated key transfer chain driven based on a strong timing constraint path.

[0073] In cross-business collaborative data acquisition threads, the core objective of each thread is to collect and process data from multiple business perspectives, ensuring that this data meets temporal requirements. The execution of each thread is driven by strongly time-constrained paths, meaning the order and timing of data acquisition are planned according to the business's temporal requirements. Strongly time-constrained paths refer to business processes and data paths with strict time requirements. For example, an order must be shipped within a certain timeframe after completion; this process has strong temporal constraints. The data transmission order is determined based on these strongly time-constrained paths to ensure that data is collected and processed at the correct time. Association key transmission chains refer to the association paths between key business identifiers from different business perspectives. Each chain involves different business systems and data fields. Through these association key transmission chains, the flow of data between multiple systems can be tracked. Under the association key transmission chain driven by strongly time-constrained paths, each cross-business collaborative data acquisition thread ensures the order and integrity of data based on these association key transmission chains, thereby achieving seamless data integration and synchronization within the cross-business collaborative data acquisition threads.

[0074] Example 2, based on the same inventive concept as the method for dynamically merging and generating reports from a multi-dimensional business perspective in the aforementioned examples, such as... Figure 2 As shown in the embodiments of this application, a system for dynamically merging and generating reports from a multi-dimensional business perspective is provided. The system includes:

[0075] The correlation analysis module 10 is used to perform cross-business temporal correlation analysis from a multi-dimensional business perspective based on the original business event flow, and construct M cross-business collaborative acquisition threads; the data acquisition module 20 is used to drive the M cross-business collaborative acquisition threads based on thread execution temporal constraints, using M temporal data field features to perform periodic data acquisition from the multi-dimensional business perspective, and obtain M single-period cross-period time-series data packets; the cross-validation module 30 is used to perform multi-dimensional confidence cross-validation of the M single-period cross-period time-series data packets based on the association key mapping relationship of the M cross-business collaborative acquisition threads, and obtain M single-period validation time-series data fields; the incremental update module 40 is used to merge the M single-period validation time-series data fields to obtain a dynamic fusion matrix report, and then dynamically store the dynamic fusion matrix report to the time-series incremental report warehouse, and perform periodic incremental updates of the time-series incremental report warehouse.

[0076] Furthermore, the data acquisition module 20 is used to perform the following operation steps:

[0077] According to the cross-service association relationship, the business association key identification of the multi-dimensional business perspective is performed to obtain a plurality of groups of business association keys; the cross-service time sequence association analysis of the plurality of groups of business association keys is performed with the original business event stream as an input source, and the M cross-service cooperative collection threads are constructed, wherein the M cross-service cooperative collection threads are subject to thread execution time sequence constraints; the original business event stream is aggregated according to the M cross-service cooperative collection threads, data field feature dynamic extraction is performed, and the M time sequence data field features are output.

[0078] Further, the data collection module 20 is configured to perform the following operation steps:

[0079] The plurality of groups of business execution keys of the multi-dimensional business perspective are called, and the topology connection of the plurality of groups of business execution keys is constructed according to the cross-service cross relationship of the multi-dimensional business perspective, to obtain an entity interaction topology; the multi-business system source data in a message format are called; the execution key association attribute analysis of the entity interaction topology is performed according to the multi-business system source data, and the topology connection is valued according to the analysis result, to obtain an attribute injection topology; in the attribute injection topology, the node association centrality is calculated according to the topology connection weighting result, to obtain a plurality of groups of correlation strength quantization values of the plurality of groups of business execution keys; the plurality of groups of correlation strength quantization values are traversed by a preset dynamic coverage threshold, and the plurality of groups of business association keys are mapped and filtered from the plurality of groups of business execution keys.

[0080] Further, the data collection module 20 is configured to perform the following operation steps:

[0081] The key value fragmentation of the original business event stream is performed based on a space-time analysis strategy, to obtain H event space-time fragmentation sets; the H event space-time fragmentation sets are traversed by the plurality of groups of business association keys, to obtain H cross-perspective association key clusters; the H cross-perspective association key clusters are identified according to the timestamp ascending sequence arrangement of the H event space-time fragmentation sets, to output H business causal dependence chains; the H business causal dependence chains are used for association path pattern mining, to generate M association key transmission chains; the original business event stream is traced back by the M association key transmission chains, to perform high-frequency event time sequence characteristic identification, to obtain M strong time sequence constraint paths; the M association key transmission chains and the M strong time sequence constraint paths are dynamically topologically instantiated, to output the M cross-service cooperative collection threads.

[0082] Further, the data collection module 20 is configured to perform the following operation steps:

[0083] statistically distribute M execution time of the M associated key transfer chains in the original service event stream; time sequence splicing the M execution time for task delay analysis, output rigid execution time delay threshold and elastic execution tolerance interval; inject the rigid execution time delay threshold and elastic execution tolerance interval into the node link of the M cross-service cooperative collection threads, generate thread execution time sequence constraints.

[0084] Further, the cross-validation module 30 is configured to perform the following operation steps:

[0085] Collect M associated key mapping chains of the M associated key transfer chains from the attribute injection topology; predefine data confidence priority of the M cross-service cooperative collection threads; perform cross-service cross-validation on the M single-cycle cross-perspective time sequence data packets according to the M associated key mapping chains, and perform confidence decision of abnormal data packets by using the data confidence priority, to obtain the M single-cycle verification time sequence data fields.

[0086] Further, the data collection module 20 is configured to perform the following operation steps:

[0087] Retrieving first time sequence service association records, first time sequence association abnormal records, first transmission delay records and first association frequency records of a first service execution key pair from the multi-service system source data; performing abnormal rate calculation on the first time sequence service association records and the first time sequence association abnormal records, and quantitatively outputting a first stability quantitative value; performing service dependence strength quantification based on the first time sequence association abnormal records and the first association frequency records, and outputting a first dependence quantitative value; calling a preset time sequence fault tolerance window to perform service time sequence compliance analysis on the first transmission delay records, and outputting a first timeliness quantitative value; weighting the first stability quantitative value, the first dependence quantitative value and the first timeliness quantitative value, and outputting a first topology link assignment.

[0088] Further, each cross-service cooperative collection thread includes an associated key transfer chain driven based on a strong time sequence constraint path.

[0089] Embodiment three, as Figure 3 shown, is a structural schematic diagram of an exemplary computer device of the present application. Figure 3 The computer device shown is only an example, and should not impose any limitation on the function and use range of the embodiments of the present application. As Figure 3 shown, the computer device includes a processor 21, a memory 22, an input device 23 and an output device 24; the number of processors 21 in the computer device can be one or more, Figure 3 for example, the processor 21 in the computer device, the processor 21, the memory 22, the input device 23 and the output device 24 in the computer device can be connected through a bus or other means,Figure 3 The connection by the bus is taken as an example.

[0090] The above is only the preferred embodiment of the present application, and does not limit the present application in any form. Although the present application has been disclosed as above with the preferred embodiment, it is not intended to limit the present application. Any person skilled in the art can make some changes or modifications to the above disclosed technical content to obtain equivalent embodiments with equivalent changes, without departing from the technical solution of the present application. Any modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application still belong to the scope of the technical solution of the present application.

Claims

1. A method for report dynamic merging generation under multi-dimensional business perspective, characterized in that, The method comprises: According to the original business event stream, the cross-business time sequence correlation analysis of the multi-dimensional business perspective is performed, and M cross-business cooperative collection threads are constructed; Based on the thread execution time sequence constraint, the M cross-business cooperative collection threads are driven by M time sequence data field characteristics, the periodic data collection of the multi-dimensional business perspective is performed, and M single-period cross-perspective time sequence data packets are obtained; According to the mapping relationship of the associated keys of the M cross-business cooperative collection threads, multi-dimensional confidence cross-verification of the M single-period cross-perspective time sequence data packets is performed, and M single-period verification time sequence data fields are obtained; After the M single-period verification time sequence data fields are merged, a dynamic fusion matrix report is obtained, the dynamic fusion matrix report is stored into a time-sequenced incremental report warehouse dynamically, and periodic incremental updating of the time-sequenced incremental report warehouse is performed.

2. The method of claim 1, wherein the multi-dimensional business perspective report is dynamically merged and generated. The method further comprises: According to the cross-business correlation relationship, business associated keys of the multi-dimensional business perspective are identified, and a plurality of groups of business associated keys are obtained; Taking the original business event stream as an input source, cross-business time sequence correlation analysis of the plurality of groups of business associated keys is performed, and the M cross-business cooperative collection threads are constructed, wherein the M cross-business cooperative collection threads are subject to thread execution time sequence constraints; According to the M cross-business cooperative collection threads, the original business event stream is aggregated, data field characteristics are dynamically extracted, and the M time sequence data field characteristics are output.

3. The method of claim 2, wherein the multi-dimensional business perspective report is dynamically merged and generated by: According to the cross-business correlation relationship, business associated keys of the multi-dimensional business perspective are identified, and a plurality of groups of business associated keys are obtained, and the method comprises: The plurality of groups of business execution keys of the multi-dimensional business perspective are called; According to the cross-business cross-relationship of the multi-dimensional business perspective, a topology connection of the plurality of groups of business execution keys is constructed, and an entity interaction topology is obtained; The multi-business system source data in the message format is called; According to the multi-business system source data, an execution key correlation attribute of the entity interaction topology is parsed, and topology connection assignment is performed according to the parsing result, and an attribute injection topology is obtained; In the attribute injection topology, node correlation centrality calculation is performed according to a topology connection weighting result, and a plurality of groups of correlation strength quantization values of the plurality of groups of business execution keys are obtained; A preset dynamic coverage threshold value traverses the plurality of groups of correlation strength quantization values, and the plurality of groups of business associated keys are mapped and filtered from the plurality of groups of business execution keys.

4. The method of claim 3, wherein the report is dynamically merged and generated under the multi-dimensional business perspective. Taking the original business event stream as an input source, cross-business time sequence correlation analysis of the plurality of groups of business associated keys is performed, and the M cross-business cooperative collection threads are constructed, and the method comprises: According to the time and space analysis strategy, key value fragmentation of the original business event stream is performed, and H event time and space fragmentation sets are obtained; The H event time and space fragmentation sets are traversed by using the plurality of groups of business associated keys, and H cross-perspective associated key clusters are obtained; According to time stamp ascending arrangement of the H event time and space fragmentation sets, causal dependence chain identification of the H cross-perspective associated key clusters is performed, and H business causal dependence chains are output; According to the H business causal dependence chains, an associated path mode is mined, and M associated key transmission chains are generated; Backtracking the original business event stream by using the M associated key transmission chains, performing high-frequency event timing characteristic identification, and obtaining M strong timing constraint paths; Instantiating the M associated key transmission chains and the M strong timing constraint paths by using a dynamic topology, and outputting the M cross-business cooperative collection threads.

5. The method of claim 4, wherein the multi-dimensional business perspective report is dynamically merged and generated by: The method further includes: Counting M execution time distribution of the M associated key transmission chains in the original business event stream; Timing splicing the M execution time distribution to perform task time delay analysis, and outputting a rigid execution time delay threshold and an elastic execution tolerance interval; Injecting the rigid execution time delay threshold and the elastic execution tolerance interval into a node link of the M cross-business cooperative collection threads, and generating thread execution timing constraints.

6. The method of claim 4, wherein the multi-dimensional business perspective report is dynamically merged and generated by: According to the associated key mapping relationship of the M cross-business cooperative collection threads, performing multi-dimensional confidence cross-validation on the M single-period cross-perspective timing data packets, and obtaining M single-period verification timing data fields, the method includes: Collecting M associated key mapping chains of the M associated key transmission chains from the attribute injection topology; Predefining data confidence priority of the M cross-business cooperative collection threads; According to the M associated key mapping chains, performing cross-business cross-validation on the M single-period cross-perspective timing data packets, and performing confidence arbitration on abnormal data packets by using the data confidence priority, to obtain the M single-period verification timing data fields.

7. The method of claim 3, wherein the report is dynamically merged and generated under a multi-dimensional business perspective. According to the multi-business system source data, performing execution key associated attribute analysis of the entity interaction topology, and assigning values to the topology according to the analysis result, to obtain an attribute injection topology, the method includes: Retrieving first timing business association records, first timing association abnormal records, first transmission delay records, and first association frequency records of a first business execution key pair from the multi-business system source data; Performing abnormal rate calculation on the first timing business association records and the first timing association abnormal records, and quantitatively outputting a first stability quantitative value; Quantifying business dependency strength based on the first timing association abnormal records and the first association frequency records, and outputting a first dependency quantitative value; Retrieving a preset timing fault tolerance window to perform business timing compliance analysis on the first transmission delay records, and outputting a first timeliness quantitative value; Weighting the first stability quantitative value, the first dependency quantitative value, and the first timeliness quantitative value, and outputting a first topology link assignment value.

8. The method of claim 2, wherein the multi-dimensional business perspective report is dynamically merged and generated. Each cross-business cooperative collection thread includes an associated key transmission chain driven based on a strong timing constraint path.

9. A report dynamic merging generation system under multi-dimensional business perspective, characterized in that, The system is used to implement the multi-dimensional business perspective report dynamic merging generation method of any one of claims 1-8, and the system includes: An association analysis module configured to perform cross-business timing association analysis of a multi-dimensional business perspective based on an original business event stream, and construct M cross-business cooperative collection threads; A data collection module configured to drive the M cross-business cooperative collection threads by using M timing data field features based on thread execution timing constraints, perform periodic data collection of the multi-dimensional business perspective, and obtain M single-period cross-perspective timing data packets. A cross-validation module is configured to perform multi-dimensional confidence cross-validation of the M single-period cross-perspective time-series data packets according to the association key mapping relationship of the M cross-business collaborative collection threads, and obtain M single-period verification time-series data fields. An incremental update module is configured to merge the M single-period verification time-series data fields, obtain a dynamic fusion matrix report, and store the dynamic fusion matrix report to a time-series incremental report warehouse for periodic incremental update of the time-series incremental report warehouse. 10.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-9. The processor executes the computer program to implement the steps of the multi-dimensional business perspective report dynamic merging generation method in any one of claims 1 to 8.

Images