Report generation method and system based on multi-source data fusion

Abstract

The present invention provides a report generation method and system for multi-source data fusion, which relates to the field of data processing technology, including: extracting a first filling grid in a first report requirement, the first filling grid having a first filling logic; obtaining a first filling parameter of the first filling grid after a deadline, and introducing a multi-layer verification mechanism in combination with the first filling logic for verification, to obtain a first verification result; when the first verification result reaches a first predetermined constraint, generating a first sequence based on the first filling parameter, including a first horizontal sequence and a first vertical sequence; introducing a multi-parallel verification mechanism to verify the first horizontal sequence or the first vertical sequence, to obtain a second verification result; when the second verification result reaches a second predetermined constraint, generating a target report. The present invention solves the technical problem that it is difficult to unify the standards for data quality and accuracy of different data sources in the prior art, resulting in insufficient data verification, and thus leading to poor accuracy and effectiveness of generated reports.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a report generation method and system for multi-source data fusion. Background Art

[0002] Enterprises often face multiple data from different systems, such as financial systems, sales systems, inventory management systems, etc. Through multi-source data fusion, these heterogeneous data can be unified into structured reports for management to make decisions and analyze. However, differences in format, unit, structure, etc. between different data sources lead to difficulties in data integration. This heterogeneity makes direct integration and generation of unified reports complex and error-prone; moreover, multi-source data often comes from different systems, and the data quality is inconsistent, and there may be problems such as missing values, redundant data, and abnormal data. In the process of report generation, how to clean and verify this data to ensure the accuracy and consistency of the data is a key challenge. If the data is not strictly controlled in terms of quality, the generated report may contain erroneous data, which will affect the effectiveness of decision-making and reporting. Summary of the invention

[0003] This application provides a report generation method and system for multi-source data fusion, aiming to solve the technical problem that it is difficult to unify the standards for data quality and accuracy of different data sources in the prior art, resulting in insufficient data verification, and thus causing poor accuracy and effectiveness of generated reports.

[0004] The first aspect disclosed in the present application provides a report generation method for multi-source data fusion, the method comprising: extracting a first filling cell in a first report requirement, wherein the first filling cell has a first filling logic; obtaining a first filling parameter of the first filling cell after a deadline, and introducing a multi-layer verification mechanism in combination with the first filling logic to verify the first filling parameter to obtain a first verification result; when the first verification result reaches a first predetermined constraint, generating a first sequence based on the first filling parameter, the first sequence comprising a first horizontal sequence and a first vertical sequence; introducing a multi-parallel verification mechanism to verify the first horizontal sequence or the first vertical sequence to obtain a second verification result; when the second verification result reaches a second predetermined constraint, generating a target report of the first report requirement.

[0005] According to a second aspect disclosed in the present application, a report generation system for multi-source data fusion is provided, and the system is used for the above-mentioned report generation method for multi-source data fusion, and the system includes: a filling grid extraction module, the filling grid extraction module is used to extract a first filling grid in a first report requirement, wherein the first filling grid has a first filling logic; a filling parameter verification module, the filling parameter verification module is used to obtain a first filling parameter of the first filling grid after a deadline, and introduce a multi-layer verification mechanism in combination with the first filling logic to verify the first filling parameter to obtain a first verification result; a first sequence generation module, the first sequence generation module is used to generate a first sequence based on the first filling parameter when the first verification result reaches a first predetermined constraint, the first sequence including a first horizontal sequence and a first vertical sequence; a first sequence verification module, the first sequence verification module is used to introduce a multi-parallel verification mechanism to verify the first horizontal sequence or the first vertical sequence to obtain a second verification result; a target report generation module, the target report generation module is used to generate a target report of the first report requirement when the second verification result reaches a second predetermined constraint.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] Identifying and extracting the first filling cell from the first report requirement can clarify the specific location and method of data filling, ensure the automation of the data filling process, avoid manual intervention, and improve filling efficiency and accuracy; by obtaining the first filling parameter after the deadline, ensure the timeliness and accuracy of data collection, provide the latest data according to the deadline, and ensure that the generated report is based on real-time information. By introducing a multi-layer verification mechanism, the filling parameters can be verified in multiple dimensions to ensure the efficiency and accuracy of data processing; automatically generate the first sequence according to the first filling parameter, including horizontal sequence and vertical sequence, to ensure the effective presentation of data in the report. The horizontal and vertical sequences correspond to different dimensions and indicators, respectively, providing a basis for the structured layout of the report. By automatically generating and organizing these sequences, the structured presentation of data is ensured, and it can cope with complex reporting requirements and provide efficient data processing and display methods. Through the multi-parallel verification mechanism, the parameters of the horizontal or vertical sequences can be independently verified at the same time, which greatly improves the efficiency of data processing. Compared with the traditional serial verification method, this parallel method can speed up the verification process, reduce the time delay of report generation, and avoid affecting the final report quality due to problems with a certain dimension or field. When the second verification result meets the predetermined standard, the final target report is generated. This process based on a strict verification mechanism ensures the accuracy and consistency of the report content, avoids report errors caused by data errors, and ensures the quality control of the report.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A flowchart of a report generation method for multi-source data fusion provided in an embodiment of the present application.

[0010] Figure 2 A schematic diagram of the structure of a report generation system for multi-source data fusion provided in an embodiment of the present application.

[0011] Description of the accompanying drawings: filling grid extraction module 10, filling parameter verification module 20, first sequence generation module 30, first sequence verification module 40, target report generation module 50. DETAILED DESCRIPTION

[0012] The embodiments of the present application provide a report generation method and system for multi-source data fusion, which solves the technical problem that it is difficult to unify the standards for data quality and accuracy of different data sources in the prior art, resulting in insufficient data verification and poor accuracy and effectiveness of generated reports.

[0013] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically introduced below in conjunction with the accompanying drawings of the specification. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0014] Embodiment 1, as Figure 1 As shown, the embodiment of the present application provides a report generation method for multi-source data fusion, the method comprising:

[0015] A first filling cell in a first report requirement is extracted, wherein the first filling cell has a first filling logic.

[0016] The first report requirement refers to the report generation requirement predefined by the user or system, which includes information such as the report format, data fields, time range, etc. This report requirement is generated in a certain business context and can be obtained through business analysis, user needs, etc.

[0017] The first fill-in cell is extracted from the first report requirement. This is a specific cell in the report template, which contains the data value that needs to be filled in. It can represent a data item in the report, such as the total sales of a certain month, the inventory quantity of a certain product, etc.

[0018] The first fill-in cell has the first fill-in logic, which defines how to obtain the data of the fill-in cell from an external data source or internal calculation and fill it into the cell. The complexity of the fill-in logic varies depending on the type and requirements of the report, such as static fill-in logic, dynamic fill-in logic, formula fill-in logic, etc. In this way, it can be ensured that each fill-in cell in the report can be correctly filled with data according to the predetermined logic.

[0019] A first filling parameter of the first filling grid after the deadline is obtained, and a multi-layer verification mechanism is introduced in combination with the first filling logic to verify the first filling parameter to obtain a first verification result.

[0020] The end date refers to the time when the report is generated or the time range of data statistics. For example, the end date of a monthly report may be the last day of the month. Filling parameters are the data required to fill specific cells in the report, such as sales, cost, inventory, etc., which can be obtained through database query. According to the logic of the first filling cell, the required data type, source and processing method are determined. According to the end date, the matching data is extracted from the corresponding data source to obtain the first filling parameter.

[0021] In order to ensure the validity and accuracy of the filling parameters, a multi-layer verification mechanism is introduced to verify the filling parameters. The verification process is divided into multiple levels. After verification, a first verification result is obtained. If the filling parameters pass all verification levels, the first verification result is generated as passed, indicating that the parameters are determined to be qualified and can be used for report filling; otherwise, if any verification level fails, the first verification result is generated as failed.

[0022] When the first verification result reaches a first predetermined constraint, a first sequence is generated based on the first filling parameter, where the first sequence includes a first horizontal sequence and a first vertical sequence.

[0023] The first predetermined constraint is the minimum requirement that the first verification result must meet, for example, it is set to 100% correctness of data format and more than 95% matching degree of filling logic. When the first verification result meets the first predetermined constraint, it means that the first filling parameter is reliable and can be used to generate a report. In this case, the first sequence is generated based on the first filling parameter that has passed the verification. The first sequence includes a first horizontal sequence and a first vertical sequence. The first horizontal sequence is the data arranged in the horizontal dimension of the table, which usually corresponds to the change of a specific indicator over time or other classification dimensions; the first vertical sequence is the data arranged in the vertical dimension of the table, which usually corresponds to the performance of multiple indicators at a certain time point or classification dimension. Specifically, the horizontal and vertical arrangement methods are planned according to the row and column structure and dimension definition (such as time, indicator, classification, etc.) of the report, and the parameter values ​​are assigned to appropriate horizontal or vertical positions according to the filling logic, and the parameters are organized into a sequence that conforms to the logic and report structure to generate the first horizontal sequence and the first vertical sequence.

[0024] A multi-parallel verification mechanism is introduced to verify the first horizontal sequence or the first vertical sequence to obtain a second verification result.

[0025] The multi-parallel verification mechanism is a multi-level, parallelized verification process. By running multiple verification logics at the same time, the different characteristics of the sequence data are fully verified. Each verification process independently verifies specific data characteristics. For example, the trend, volatility, consistency, etc. of the time series data are verified for the horizontal sequence, and the correlation between indicators and the integrity of the classified data are verified for the vertical sequence. After verification, if the data meets all the verification rules, the second verification result is generated as passed. On the contrary, if some verifications fail, the second verification result is generated as failed.

[0026] When the second verification result meets the second predetermined constraint, a target report required by the first report is generated.

[0027] The second predetermined constraint is the minimum requirement that the second verification result must meet. Specifically, the data must pass verification at all levels. If the second verification result meets the second predetermined constraint, the report generation stage is entered. Specifically, the data that has passed the verification is gradually filled into the horizontal axis field of the report according to the horizontal dimension, and the data that has passed the verification is gradually filled into the vertical axis field of the report according to the vertical dimension to obtain the target report.

[0028] Further, including:

[0029] Analyze the first filling logic to obtain a first related mechanism group, wherein the first related mechanism group corresponds to a first filling channel group; randomly extract a first mechanism in the first related mechanism group, wherein the first mechanism corresponds to a first channel in the first filling channel group; obtain a first parameter of the first mechanism in the first channel, and obtain a first correlation coefficient in combination with the first filling parameter; determine whether the first correlation coefficient is within a first predetermined correlation coefficient threshold according to the multi-layer verification mechanism, and obtain a first judgment result; analyze the first judgment result to obtain the first verification result.

[0030] The first filling logic defines how to obtain the data of the filling grid from an external data source or internal calculation. According to the filling logic, determine which institutions the data needs to be obtained from. For example, if the first filling logic describes extracting sales data by department, the relevant institution group may include the sales department, marketing department, and finance department. Combined with the database or configuration file, determine which institutions are related to the first filling logic, and summarize all matching institutions into the first relevant institution group. The first relevant institution group corresponds to the first filling channel group. The filling channel refers to the specific data transmission path between the institution and the report filling logic. One relevant institution corresponds to one filling channel, which is used to obtain the required data. For example, the channel of the marketing department can be the marketing department database interface.

[0031] An institution is randomly selected from the first related institution group as the first institution, and a filling channel of the selected first institution, namely the first channel, is found in the system configuration as an input source for subsequent data operations.

[0032] The first parameter is the core data obtained through the first channel corresponding to the selected first institution, which is used to perform correlation calculation with the first filling parameter. For example, "sales of this quarter" is extracted from the marketing department database interface (first channel) as the first parameter. The correlation calculation method can use the Pearson correlation coefficient. After the correlation calculation, the first correlation coefficient is obtained. The correlation coefficient is used to measure the linear relationship between two sets of data, indicating the correlation between the first parameter and the first filling parameter. The value range is [-1,1], where +1 indicates a complete positive correlation, 0 indicates no correlation, and -1 indicates a complete negative correlation.

[0033] The first predetermined correlation coefficient threshold is a standard range for correlation determination, which can be set according to business logic requirements, for example, [0.8, 1.0] indicates that high correlation is qualified. It is determined whether the first correlation coefficient is within the first predetermined correlation coefficient threshold range to obtain a first determination result.

[0034] The multi-layer verification mechanism refers to the parallel and unified analysis and verification of the parameters of the first filling channel group corresponding to all first related mechanism groups. Specifically, the first filling channel group is traversed, and the corresponding parameters are obtained through each channel. The parameters of each channel are respectively correlated with the first filling parameters to obtain multiple correlation coefficients, and each correlation coefficient is judged one by one whether it is within the first predetermined correlation coefficient threshold to obtain the judgment results of all channels.

[0035] The judgment results of all channels are collected and integrated into the final judgment, that is, when the correlation coefficients of all channels are within the first predetermined correlation coefficient threshold, the first judgment result is passed, indicating that the current first filling parameter can be used for report filling; when any channel is not within, the first judgment result is failed. By introducing a multi-layer verification mechanism, it is ensured that the data obtained from all relevant channels are highly correlated with the filling logic to avoid filling errors caused by abnormal data in a certain channel.

[0036] Further, including:

[0037] Obtain a first historical parameter record of the first institution; perform a correlation analysis on the first historical parameter record and the first historical filling parameter record of the first filling grid to obtain a first correlation analysis result; determine the first predetermined correlation coefficient threshold value based on the first maximum correlation coefficient and the first minimum correlation coefficient in the first correlation analysis result.

[0038] A first historical parameter record is obtained, where the first historical parameter record is historical data obtained from a data source of the first institution and is used for comparative analysis of the correlation between the first historical parameter record and the historical data of the first fill-in grid.

[0039] The first historical filling parameter record is the historical data specified in the report requirement of the first filling grid. A correlation analysis is performed on the first historical parameter record and the first historical filling parameter record, for example, using Pearson correlation coefficient calculation, and the calculated correlation coefficient is used as the first correlation analysis result.

[0040] The first predetermined correlation coefficient threshold is used as a standard for determining data correlation, and is used to verify whether the current filling parameters conform to the rules of historical data. The wider the threshold range is set, the larger the allowed variation space is, and the narrower the setting is, the stricter the verification is. The maximum correlation coefficient and the minimum correlation coefficient in the first correlation analysis result are extracted, and the minimum correlation coefficient is used as the lower limit of the first predetermined correlation coefficient threshold, and the maximum correlation coefficient is used as the upper limit of the first predetermined correlation coefficient threshold as the standard for verifying the current filling data.

[0041] Further, including:

[0042] Extract the first adjacent sequence parameter group in the first horizontal sequence, the first adjacent sequence parameter group including the first sequence parameter and the second sequence parameter; determine whether the first sequence parameter and the second sequence parameter belong to the same dimension; if so, perform the multi-parallel verification mechanism on the first horizontal sequence.

[0043] The first horizontal sequence is a horizontal data sequence generated based on report requirements. The first adjacent sequence parameter group is two adjacent data points in the first horizontal sequence, including the first sequence parameter and the second sequence parameter. Specifically, in horizontal order, the two adjacent parameters in the first horizontal sequence are combined into an adjacent parameter group in sequence, and the horizontal sequence is traversed. From all adjacent parameter groups, one group is randomly extracted as the first adjacent sequence parameter group, and one of the parameters in the first adjacent sequence parameter group is used as the first sequence parameter, and the other parameter adjacent to it is the second sequence parameter.

[0044] Dimension refers to the physical quantity or data unit, such as ten thousand yuan, number, percentage. The same dimension requires that the units and data types of the two parameters in the parameter group be consistent. For the first sequence parameters and the second sequence parameters, check their units and data types. If the two parameters represent the same indicator, the dimensions are consistent; if they are different, the dimensions are inconsistent.

[0045] If the two parameters belong to the same dimension, the first adjacent sequence parameter group belonging to the same dimension is verified, and the specific verification process is detailed in the subsequent steps. According to the multi-parallel verification mechanism, all adjacent sequence parameter groups in the first horizontal sequence are traversed, and the above judgment process is repeated. For all adjacent sequence parameter groups belonging to the same dimension, parallel multi-layer verification is performed. Through the multi-parallel verification mechanism, the verification of all adjacent sequence parameter groups runs simultaneously, which improves the verification efficiency.

[0046] Further, if it does not belong, the first longitudinal sequence is verified by the multi-parallel verification mechanism.

[0047] If the first sequence parameters and the second sequence parameters do not belong to the same dimension, it means that the verification cannot be continued on the horizontal sequence. In this case, the first vertical sequence is verified instead. The first vertical sequence is the data arranged by the vertical dimension in the report. When the parameter dimensions of the horizontal sequence are inconsistent, the vertical verification is mainly used to verify data of different dimensions. The multi-parallel verification mechanism for the first vertical sequence is similar to that for the first horizontal sequence. The specific verification process will be detailed in the subsequent steps.

[0048] Further, including:

[0049] Perform energy modal decomposition processing on the first horizontal sequence or the first vertical sequence to obtain a decomposition processing result; perform weight evaluation on multiple energy modal components in the decomposition processing result, and screen to obtain the maximum energy modal component; generate a first parameter sequence of the first mechanism based on the first historical parameter record according to the multi-parallel verification mechanism, and obtain the first maximum energy modal component of the first parameter sequence; obtain a first coverage of the maximum energy modal component and the first maximum energy modal component; analyze the first coverage to obtain the second verification result.

[0050] Energy modal decomposition is a signal processing technique used to extract intrinsic modal components from nonlinear and non-stationary signals. Each modal component represents a different frequency component of the signal. Energy modal decomposition decomposes the signal in an adaptive way and decomposes complex time series data into a set of components called intrinsic mode functions.

[0051] Energy mode decomposition is performed on the first horizontal sequence (such as time series, sales data, etc.) or the first vertical sequence (such as statistical data of multiple dimensions) to extract information of different frequencies or fluctuation modes. Specifically, for the first horizontal sequence or the first vertical sequence, the original sequence is decomposed into several intrinsic mode functions through the empirical mode decomposition algorithm. Each intrinsic mode function represents a frequency component in the signal. The decomposition process generates and calculates envelopes based on local extreme points until intrinsic mode functions with different frequency characteristics are obtained. Finally, multiple energy mode components are output as the decomposition processing results, and each component corresponds to different characteristics of the data. For example, sales data may contain long-term trends and periodic fluctuations on different time scales. Energy mode decomposition can decompose this information and help identify potential patterns.

[0052] For multiple energy modal components in the decomposition processing results, energy evaluation is performed on them respectively. Energy is the sum of the squares of the amplitudes of each energy modal component, which represents the energy contribution of the component in the original signal. By evaluating the energy of each energy modal component, its dominant role in the decomposition result is judged. According to the energy evaluation result of each energy modal component, the maximum energy modal component is screened out, that is, the energy modal component with the largest energy value, which represents the most important or most significant component in the signal.

[0053] The multi-parallel verification mechanism means that multiple verification operations are performed in parallel to ensure efficient data processing. Based on the first historical parameter record, a first parameter sequence is generated. This sequence contains historical data from different times or different sources. For example, if the first historical parameter record is the sales data of the past 4 months, the generated first parameter sequence is the sales data arranged in chronological order. Through the aforementioned energy modal decomposition process, the first parameter sequence is subjected to energy modal decomposition, and according to the decomposition result, the component with the highest energy is selected as the first maximum energy modal component.

[0054] Coverage is an indicator that measures the similarity or overlap of two energy modal components. It describes the similarity between the two components and can be calculated by their correlation or intersection. Compare the maximum energy modal component and the first maximum energy modal component, and use correlation coefficients, overlapping areas and other metrics to calculate the coverage of the two components to obtain the first coverage.

[0055] The first coverage is judged. For example, if the first coverage reaches the set threshold, such as 0.9, it means that the overlap between the two modal components is large, that is, the data is consistent in terms of change trend and fluctuation pattern, so the data quality is high. In this case, the second verification result is judged to be passed; otherwise, it is failed.

[0056] Further, including:

[0057] The first correlation coefficient is used as a weight coefficient to adjust the first coverage to obtain a first effective coverage; the first effective coverage is analyzed to obtain the second verification result.

[0058] The first correlation coefficient is used as a weight coefficient to adjust the first coverage to make it more consistent with the real relationship between the data and the quality requirements. Through the adjustment, the first effective coverage is obtained. If the first correlation coefficient is high, it means that the relationship between the data is strong, and the adjusted first effective coverage will be higher, indicating that the data quality is more reliable; if the first correlation coefficient is low, the adjusted first effective coverage will be lower, indicating that the data quality is poor. Similarly, the adjusted first effective coverage is judged. If the first effective coverage reaches the set threshold, the second verification result is judged to be passed; otherwise, it is judged to be failed.

[0059] Further, including:

[0060] When the second verification result does not meet the second predetermined constraint, a variation weighted calculation is performed on the first effective coverage after normalization to obtain a first coverage coefficient; a first abnormality coefficient is obtained based on the first coverage coefficient; when the first abnormality coefficient reaches a predetermined abnormality threshold, an abnormal warning is issued for the first filling grid.

[0061] If the second verification result does not meet the set second predetermined constraint, it indicates that the data does not meet expectations and there may be quality problems or logical errors. In this case, the first effective coverage is first normalized. Normalization can map the data to a standard range, usually [0,1], to ensure that the scales of different data sets are consistent. Then, the normalized first effective coverage is weighted by variation, that is, weighted according to the variability of the data, such as fluctuations or range of variation, to adjust the weight of the effective coverage to better reflect the uncertainty of the data. The variability can be calculated through metrics such as the standard deviation and variance of the data. The first coverage coefficient is obtained through variation weighted calculation.

[0062] The first anomaly coefficient is obtained based on the first coverage coefficient. If the first coverage coefficient shows large fluctuations or does not conform to the expected range, a higher first anomaly coefficient will be calculated to identify whether the data has anomalies, where the anomalies may be caused by excessive data fluctuations, poor data quality or other reasons.

[0063] The abnormal threshold can be set according to business needs and the normal fluctuation range of data. For example, the first abnormal coefficient is set to be abnormal if it exceeds 1.5. The setting of the abnormal threshold depends on the distribution of historical data or industry standards, and is used to effectively identify data that deviates from the normal range. If the first abnormal coefficient is greater than or equal to the predetermined abnormal threshold, the data is considered abnormal, and an abnormal warning is triggered, indicating that there is a problem with the data. In this way, potential abnormal data can be efficiently identified and processed, ensuring the accuracy of report generation and data quality.

[0064] In summary, the report generation method of multi-source data fusion provided in the embodiment of the present application has the following technical effects:

[0065] Identifying and extracting the first filling cell from the first report requirement can clarify the specific location and method of data filling, ensure the automation of the data filling process, avoid manual intervention, and improve filling efficiency and accuracy; by obtaining the first filling parameter after the deadline, ensure the timeliness and accuracy of data collection, provide the latest data according to the deadline, and ensure that the generated report is based on real-time information. By introducing a multi-layer verification mechanism, the filling parameters can be verified in multiple dimensions to ensure the efficiency and accuracy of data processing; automatically generate the first sequence according to the first filling parameter, including horizontal sequence and vertical sequence, to ensure the effective presentation of data in the report. The horizontal and vertical sequences correspond to different dimensions and indicators, respectively, providing a basis for the structured layout of the report. By automatically generating and organizing these sequences, the structured presentation of data is ensured, and it can cope with complex reporting requirements and provide efficient data processing and display methods. Through the multi-parallel verification mechanism, the parameters of the horizontal or vertical sequences can be independently verified at the same time, which greatly improves the efficiency of data processing. Compared with the traditional serial verification method, this parallel method can speed up the verification process, reduce the time delay of report generation, and avoid affecting the final report quality due to problems with a certain dimension or field. When the second verification result meets the predetermined standard, the final target report is generated. This process based on a strict verification mechanism ensures the accuracy and consistency of the report content, avoids report errors caused by data errors, and ensures the quality control of the report.

[0066] Embodiment 2, based on the same inventive concept as the report generation method of multi-source data fusion in the above embodiment, Figure 2 As shown, the embodiment of the present application provides a report generation system for multi-source data fusion, the system comprising:

[0067] A filling grid extraction module 10, the filling grid extraction module 10 is used to extract the first filling grid in the first report requirement, wherein the first filling grid has a first filling logic; a filling parameter verification module 20, the filling parameter verification module 20 is used to obtain the first filling parameter of the first filling grid after the deadline, and introduce a multi-layer verification mechanism in combination with the first filling logic to verify the first filling parameter to obtain a first verification result; a first sequence generation module 30, the first sequence generation module 30 is used to generate a first sequence based on the first filling parameter when the first verification result reaches a first predetermined constraint, the first sequence including a first horizontal sequence and a first vertical sequence; a first sequence verification module 40, the first sequence verification module 40 is used to introduce a multi-parallel verification mechanism to verify the first horizontal sequence or the first vertical sequence to obtain a second verification result; a target report generation module 50, the target report generation module 50 is used to generate a target report of the first report requirement when the second verification result reaches a second predetermined constraint.

[0068] Furthermore, the system further includes a first verification result acquisition module to perform the following operation steps:

[0069] Analyze the first filling logic to obtain a first related mechanism group, wherein the first related mechanism group corresponds to a first filling channel group; randomly extract a first mechanism in the first related mechanism group, wherein the first mechanism corresponds to a first channel in the first filling channel group; obtain a first parameter of the first mechanism in the first channel, and obtain a first correlation coefficient in combination with the first filling parameter; determine whether the first correlation coefficient is within a first predetermined correlation coefficient threshold according to the multi-layer verification mechanism, and obtain a first judgment result; analyze the first judgment result to obtain the first verification result.

[0070] Furthermore, the system further includes a threshold acquisition module to perform the following operation steps:

[0071] Obtain a first historical parameter record of the first institution; perform a correlation analysis on the first historical parameter record and the first historical filling parameter record of the first filling grid to obtain a first correlation analysis result; determine the first predetermined correlation coefficient threshold value based on the first maximum correlation coefficient and the first minimum correlation coefficient in the first correlation analysis result.

[0072] Furthermore, the system further includes a verification module to perform the following operation steps:

[0073] Extract the first adjacent sequence parameter group in the first horizontal sequence, the first adjacent sequence parameter group including the first sequence parameter and the second sequence parameter; determine whether the first sequence parameter and the second sequence parameter belong to the same dimension; if so, perform the multi-parallel verification mechanism on the first horizontal sequence.

[0074] Further, if it does not belong, the first longitudinal sequence is verified by the multi-parallel verification mechanism.

[0075] Furthermore, the system further includes a second verification result acquisition module to perform the following operation steps:

[0076] Perform energy modal decomposition processing on the first horizontal sequence or the first vertical sequence to obtain a decomposition processing result; perform weight evaluation on multiple energy modal components in the decomposition processing result, and screen to obtain the maximum energy modal component; generate a first parameter sequence of the first mechanism based on the first historical parameter record according to the multi-parallel verification mechanism, and obtain the first maximum energy modal component of the first parameter sequence; obtain a first coverage of the maximum energy modal component and the first maximum energy modal component; analyze the first coverage to obtain the second verification result.

[0077] Furthermore, the second verification result module further includes the following operation steps:

[0078] The first correlation coefficient is used as a weight coefficient to adjust the first coverage to obtain a first effective coverage; the first effective coverage is analyzed to obtain the second verification result.

[0079] Furthermore, the system also includes an abnormal warning module to perform the following operation steps:

[0080] When the second verification result does not meet the second predetermined constraint, a variation weighted calculation is performed on the first effective coverage after normalization to obtain a first coverage coefficient; a first abnormality coefficient is obtained based on the first coverage coefficient; when the first abnormality coefficient reaches a predetermined abnormality threshold, an abnormal warning is issued for the first filling grid.

[0081] Through the above detailed description of the report generation method of multi-source data fusion in this specification, those skilled in the art can clearly understand the report generation system of multi-source data fusion in this embodiment. Since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0082] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A report generation method based on multi-source data fusion, characterized in that: The method comprises: Extracting a first filling cell in a first report requirement, wherein the first filling cell has a first filling logic; Acquire a first filling parameter of the first filling grid after the deadline, and introduce a multi-layer verification mechanism in combination with the first filling logic to verify the first filling parameter, to obtain a first verification result; When the first verification result reaches a first predetermined constraint, generating a first sequence based on the first filling parameter, the first sequence comprising a first horizontal sequence and a first vertical sequence; Introducing a multi-parallel verification mechanism to verify the first horizontal sequence or the first vertical sequence to obtain a second verification result; When the second verification result meets the second predetermined constraint, generating a target report required by the first report; Among them, include: Analyze the first filling logic to obtain a first related mechanism group, wherein the first related mechanism group corresponds to a first filling channel group; Randomly extracting a first mechanism from the first related mechanism group, the first mechanism corresponding to a first channel from the first filling channel group; Acquire a first parameter of the first mechanism in the first channel, and obtain a first correlation coefficient in combination with the first filling parameter; Determine whether the first correlation coefficient is within a first predetermined correlation coefficient threshold according to the multi-layer verification mechanism, and obtain a first determination result; Analyzing the first judgment result to obtain the first verification result; Among them, include: Performing energy modal decomposition processing on the first horizontal sequence or the first vertical sequence to obtain a decomposition processing result; Performing weight evaluation on multiple energy modal components in the decomposition processing result, and screening to obtain the maximum energy modal component; According to the multi-parallel verification mechanism, generating a first parameter sequence of the first mechanism based on a first historical parameter record, and obtaining a first maximum energy modal component of the first parameter sequence; Acquire a first coverage of the maximum energy modal component and the first maximum energy modal component; The first coverage is analyzed to obtain the second verification result.

2. The report generation method of multi-source data fusion according to claim 1 is characterized in that: include: Obtaining a first historical parameter record of the first mechanism; Performing a correlation analysis on the first historical parameter record and the first historical filling parameter record of the first filling cell to obtain a first correlation analysis result; The first predetermined correlation coefficient threshold is determined according to a first maximum correlation coefficient and a first minimum correlation coefficient in the first correlation analysis result.

3. The report generation method of multi-source data fusion according to claim 1 is characterized in that: include: Extracting a first adjacent sequence parameter group in the first horizontal sequence, where the first adjacent sequence parameter group includes a first sequence parameter and a second sequence parameter; Determining whether the first sequence parameter and the second sequence parameter belong to the same dimension; If it does, the first horizontal sequence is verified using the multi-parallel verification mechanism.

4. The report generation method of multi-source data fusion according to claim 3 is characterized in that: If not, the first longitudinal sequence is verified using the multi-parallel verification mechanism.

5. The report generation method of multi-source data fusion according to claim 1 is characterized in that: include: Using the first correlation coefficient as a weight coefficient, adjusting the first coverage to obtain a first effective coverage; The first effective coverage is analyzed to obtain the second verification result.

6. The report generation method of multi-source data fusion according to claim 5 is characterized in that: include: When the second verification result does not meet the second predetermined constraint, performing variation weighted calculation on the first effective coverage after normalization to obtain a first coverage coefficient; Obtaining a first anomaly coefficient based on the first coverage coefficient; When the first abnormal coefficient reaches a predetermined abnormal threshold, an abnormal warning is issued for the first filling grid.

7. The report generation system based on multi-source data fusion is characterized by: A report generation method for implementing the multi-source data fusion according to any one of claims 1 to 6, the system comprising: A filling cell extraction module, the filling cell extraction module is used to extract a first filling cell in the first report requirement, wherein the first filling cell has a first filling logic; A filling parameter verification module, the filling parameter verification module is used to obtain a first filling parameter of the first filling grid after the deadline, and introduce a multi-layer verification mechanism in combination with the first filling logic to verify the first filling parameter to obtain a first verification result; a first sequence generating module, the first sequence generating module being configured to generate a first sequence based on the first padding parameter when the first verification result reaches a first predetermined constraint, the first sequence comprising a first horizontal sequence and a first vertical sequence; A first sequence verification module, the first sequence verification module is used to introduce a multi-parallel verification mechanism to verify the first horizontal sequence or the first vertical sequence to obtain a second verification result; A target report generation module is used to generate a target report required by the first report when the second verification result meets a second predetermined constraint.

Images