A government affair data monitoring method and device, electronic equipment and storage medium

Abstract

This application discloses a method, apparatus, electronic device, and storage medium for monitoring government data. The method includes: acquiring multi-source government data and determining a consistency score for the multi-source government data based on a pre-defined consistency constraint rule base; calculating a statistical anomaly score for the multi-source government data based on at least one anomaly detection algorithm from an anomaly detection algorithm pool; calculating a comprehensive risk score for the multi-source government data based on the consistency score and the statistical anomaly score; determining the risk level of the multi-source government data based on the comprehensive risk score; and providing graded and frequency-based early warning pushes according to the risk level. This solution can accurately identify anomalies in multi-source government data, reducing the workload of manual discovery and problem investigation.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, apparatus, electronic device and storage medium for monitoring government data. Background Technology

[0002] With the deepening of the construction of "digital government," government departments at all levels have accumulated massive amounts of government data, covering multiple fields such as market supervision, social security, tax collection and administration, and public services. This data is characterized by its large volume, wide range of sources, rapid updates, and strong correlations. To ensure the accuracy, timeliness, security, and availability of government data, timely detection of abnormal data fluctuations (hereinafter referred to as "abnormalities") has become a core aspect of government data governance.

[0003] Currently, government data monitoring mainly adopts the "statistical anomaly + rule verification" approach, which includes the following steps: 1. Data entry and hierarchical statistics: Key business indicators of each department are summarized at different time granularities such as monthly, weekly, and daily, and stored regularly to form time-series statistical data. 2. Anomaly judgment rules: Based on pre-configured static thresholds or simple statistical rules such as time series slopes (e.g., month-on-month growth rate, year-on-year growth rate), the indicator series are scanned and compared by writing rule SQL to identify whether there are significant fluctuations. 3. Alarm notification and manual review: When an anomaly is determined, the system sends an alert notification to relevant personnel via SMS or in-system message; subsequently, manual intervention is carried out to perform secondary comparison and analysis of the data that triggered the anomaly, and a manual analysis report is generated.

[0004] This method has achieved monitoring of fluctuations in government data to a certain extent, but it has obvious limitations in complex government scenarios: (1) The early warning method is simple: it only makes simple judgments based on thresholds or the slope of trends in time series, and fails to integrate deeper automatic analysis and judgment; (2) High false alarms and weak interpretability: fluctuations caused by seasons, holidays, fiscal cycles, etc. are easily misjudged, and it is difficult to locate the cause of the anomaly; (3) There is no cross-table and cross-domain consistency detection scoring, and no reconciliation analysis to effectively suppress false alarms and perform post-event verification during the detection stage; (4) Lack of causal tracing ability: it cannot identify the transmission path and lag period of "upstream-downstream"; (5) Early warning fatigue: it lacks hierarchical frequency division, cooling and sample gating strategies; (6) Incomplete closed loop: in the traditional government field, it only forms system alarms, manual processing and problem solving, and the feedback of the handling cannot be fed back for online learning and parameter adaptation; (7) Insufficient engineering: it lacks templated rules / DSL dual-mode application and gray release, making it difficult to implement in the government field. Summary of the Invention

[0005] This application provides a method, device, electronic device, and storage medium for monitoring government data, which can accurately identify anomalies in multi-source government data and reduce the workload of manual discovery and problem investigation.

[0006] According to one aspect of this application, a method for monitoring government data is provided, the method comprising: Acquire multi-source government data and determine the consistency score of the multi-source government data based on a pre-defined cross-validation constraint rule base; Calculate the statistical anomaly score of the multi-source government data based on at least one anomaly detection algorithm from the anomaly detection algorithm pool. The comprehensive risk score of the multi-source government data is calculated based on the consistency score and the statistical anomaly score. The risk level of the multi-source government data is determined based on the comprehensive risk score, and early warnings are pushed out in a graded and frequency-based manner according to the risk level.

[0007] According to one aspect of this application, a government data monitoring device is provided, the device comprising: The consistency score determination module is used to acquire multi-source government data and determine the consistency score of the multi-source government data based on a pre-set consistency constraint rule library. The statistical anomaly score calculation module is used to calculate the statistical anomaly score of the multi-source government data based on at least one anomaly detection algorithm in the anomaly detection algorithm pool; The comprehensive risk score calculation module is used to calculate the comprehensive risk score of the multi-source government data based on the consistency score and the statistical anomaly score. The early warning push module is used to determine the risk level of the multi-source government data based on the comprehensive risk score, and to push early warnings in a graded and frequencyd manner according to the risk level.

[0008] According to another aspect of this application, an electronic device is provided, the electronic device comprising: At least one processor; and A memory that is communicatively connected to at least one processor; wherein, The memory stores a computer program that can be executed by at least one processor, such that the at least one processor is able to execute the government data monitoring method of any embodiment of this application.

[0009] According to another aspect of this application, a computer-readable storage medium is provided, which stores computer instructions for causing a processor to execute and implement the government data monitoring method of any embodiment of this application.

[0010] According to another aspect of this application, a computer program product is provided, which includes a computer program that, when executed by a processor, implements the government data monitoring method of any embodiment of this application.

[0011] The technical solution of this application embodiment acquires multi-source government data and determines the consistency score of the multi-source government data based on a pre-set consistency constraint rule library; calculates the statistical anomaly score of the multi-source government data based on at least one anomaly detection algorithm in the anomaly detection algorithm pool; calculates the comprehensive risk score of the multi-source government data based on the consistency score and the statistical anomaly score; determines the risk level of the multi-source government data based on the comprehensive risk score, and pushes graded and frequency-based early warnings according to the risk level. This solution can accurately identify anomalies in multi-source government data, reducing the workload of manual discovery and problem investigation.

[0012] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent from the following description. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 A flowchart illustrating a government data monitoring method provided in this application embodiment; Figure 2 This application provides a schematic diagram of the structure of a government data monitoring system. Figure 3 This is a schematic diagram of the structure of a government data monitoring device provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0015] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0016] It should be noted that the terms "first," "second," "third," "fourth," "actual," "preset," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0017] Figure 1 This is a flowchart illustrating a method for monitoring government data, provided as an embodiment of this application. This embodiment is applicable to situations involving the monitoring of government data. The method can be executed by a government data monitoring device, which can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the method includes: S110. Obtain multi-source government data and determine the consistency score of the multi-source government data based on a pre-set consistency constraint rule library.

[0018] In this embodiment, multi-source government data from departmental business systems, municipal data platforms, and market-based third-party access is acquired through a data access gateway. The access methods for this multi-source government data include, but are not limited to, batch access, streaming access, timed extraction, and event-triggered collection. During the multi-source government data access process, caliber alignment and master data alignment can be performed simultaneously. Calibration alignment refers to uniformly mapping the definition, statistical scope, and calculation method of the same indicator in government data from different sources to ensure semantic consistency. Master data alignment refers to cleaning and standardizing key fields such as unit codes, region codes, and entity identifiers to eliminate homonyms or synonymous names. Through caliber alignment and master data alignment, multi-source heterogeneous government data can be converted into unified standard data that can be used for subsequent anomaly monitoring.

[0019] In this embodiment, a pre-defined reconciliation constraint rule library is obtained. This library contains multiple reconciliation constraint rules. Multi-source government data is analyzed based on these rules to determine the reconciliation consistency score. For example, each reconciliation constraint rule in the library and the multi-source government data can be input into a pre-trained reconciliation consistency evaluation model. The reconciliation consistency score of the multi-source government data is determined based on the output of the model. Optionally, determining the reconciliation consistency score of the multi-source government data based on the pre-defined reconciliation constraint rule library includes: constructing a reconciliation constraint graph corresponding to the multi-source government data based on the pre-defined library; calculating the reconciliation violation degree corresponding to each reconciliation constraint involved in the graph; and determining the reconciliation consistency score of the multi-source government data based on each reconciliation violation degree.

[0020] For example, determining the various indicator nodes involved in multi-source government data. Among them, indicator nodes This represents the instantiated value of an indicator at a specific region, time granularity, period, and statistical version. It is expanded from the indicator library according to the routing matrix. Granularity selection can include daily, weekly, monthly, quarterly, yearly, etc., and the period uses the standard period (e.g., 2024M08). The constraint edges between indicator nodes are determined based on a pre-defined cross-validation constraint rule base. Among them, constraint edges are executable rules formed by connecting several indicator nodes according to the rule of "interrelationship". The elements constituting the constraint edges can include ⟨ , , , >. Among them, : Indicates the type of constraint between indicator nodes, which can include balance, sum / difference, correspondence, ratio, and other constraint relationships. Represents constraint expressions (equality or interval representation). Indicates tolerance (or uses [l,u]: upper and lower bounds of the scale). This indicates the weight of the constraint (affecting the contribution during fusion; for example, the value can be set to 0.5–2, with a default of 1; a value of 1.5 can be used for rules with strong institutional basis, and the corresponding constraint with known unstable caliber can be reduced to 0.8). This indicates the scope of application (region, department, cycle, version), which can be limited to a specific region, department, or cycle for gray-scale trial operation. This indicates the rule version and effective period. Based on the various indicator nodes and the constraints between them, a cross-validation constraint diagram is constructed: .

[0021] It should be noted that the balance constraint type refers to the sum of the total amount and its components, such as... The sum / difference constraint type refers to an addition / subtraction relationship, such as... This constraint type refers to cross-table / cross-period consistency, such as... The "proportion" constraint type refers to a ratio that is within a reasonable range, such as... ,in, Institutional definitions can be used; if missing, historical quantifiers can be used (e.g.) Estimate. Each constraint comes with... (Region, department, scope) is to support grayscale and change tracking.

[0022] In this embodiment, for each reconciliation constraint involved in the reconciliation constraint diagram, the reconciliation violation degree corresponding to the reconciliation constraint is calculated. Reconciliation constraints refer to consistency constraints within and between reports, such as balance, sum / difference, correspondence, and proportion. Specifically, for unified calculation, reconciliation constraints can be transformed into equation residuals. or interval overdistance Among them, equations of the form (balance, sum and difference, correspondence): ,(Target: For example: balanced ; Interval type (proportion): ,(Target: ( ), where when B=0, a safety procedure is performed: skip the calculation and record "failure reason = denominator data too small". This can be used for... Do Methods for handling outliers.

[0023] For any cross-reference constraint The degree of reconciliation violation corresponding to reconciliation constraints (Uniform scale, the larger the scale, the more inconsistent the expression), for example, equations (balance, sum, correspondence): ; Interval type (proportion): ,in, The historical residual standard deviation can be represented by the most recent standard deviation. Robust scaling of residuals for periods such as (e.g., day=60, week=26, month=12) (Use MAD / IQR or robust standard deviation); This is a tolerance term used to avoid division by zero and to reflect institutional tolerance. For example, take... Set the tolerance to the 90th percentile of the absolute value of historical residuals—90% of normal fluctuations are considered tolerable, with sensitivity only maintained for the most abnormal 10% (ensuring greater noise resistance and fewer false alarms). If historical data is insufficient, use... ( (Default 1). The advantage of this setting is that it organizes the reconciliation rules such as "balance, sum, correspondence, and proportion" in a factor graph manner, and is oriented towards consistency verification across tables, periods, and calibers. It adopts historical adaptive violation quantification and outputs a unified and comparable consistency score.

[0024] In this embodiment of the application, the consistency score of multi-source government data is determined based on each consistency violation degree. For example, the consistency score can be determined according to the following formula. (The larger the value, the greater the inconsistency): , This is the compression factor (which can be set to 3 by default).

[0025] S120. Calculate the statistical anomaly score of the multi-source government data based on at least one anomaly detection algorithm in the anomaly detection algorithm pool.

[0026] In this embodiment, based on the various anomaly detection algorithms in the anomaly detection algorithm pool, initial anomaly scores for multi-source government data are calculated, and statistical anomaly scores for the multi-source government data are obtained based on these initial anomaly scores. The anomaly detection algorithms can include sentinel-type algorithms, detective-type algorithms, and oracle-type algorithms. Sentinel-type algorithms are mainly used to detect anomalies with obvious statistical regularities or abrupt changes; these algorithms can include SH-ESD, STL+Shewhart control charts, and BOCPD. Detective-type algorithms are mainly used to discover hidden structural or morphological anomalies in the data; these algorithms can include Matrix Profile (discord) and Isolation Forest. Oracle-type algorithms are mainly used to evaluate residual anomalies through prediction models; these algorithms can include SARIMA, Prophet, and LSTNet. Each anomaly detection algorithm in the anomaly detection algorithm pool can declare "usability conditions" (such as lower limit of sample size, presence or absence of seasonality, and whether multidimensional features are required). When an algorithm is unavailable, it is marked as "absent" and annotated in the explanation. Optionally, when determining the initial anomaly score of multi-source government data based on anomaly detection algorithms, the input to the anomaly detection algorithms can include not only the multi-source government data itself, but also its metadata. This metadata can include the SLA verification results, data volume, and cooling status of the multi-source government data. It is understood that, following the above method, an initial anomaly score vector composed of the initial anomaly scores of the multi-source government data determined by each anomaly detection algorithm can be determined. ,in, This represents the initial anomaly score of the multi-source government data determined by the anomaly detection algorithm in the i-th iteration.

[0027] Obtain offline playback metrics of historical government data for anomaly detection algorithms, and determine the credibility weight of the anomaly detection algorithm based on these metrics. For example, the credibility weight is determined based on metrics (AUC-PR (prioritizing positive examples when they are sparse), Recall@fixed alarm rate, and false alarm cost) of the anomaly detection algorithm for offline playback / gray-scale operation over the past W periods (day=60, week=26, month=12). (Upper / lower limits can be configured); the credibility weight corresponding to the newly launched anomaly detection algorithm can be set to... Warm start. When an anomaly detection algorithm is unavailable... The explanation should include the "reason for unavailability" (such as insufficient sample size, seasonal missing data, unavailable features, etc.).

[0028] Based on a preset aggregation strategy, the initial anomaly scores and confidence weights corresponding to each anomaly detection algorithm are aggregated to generate statistical anomaly scores for multi-source government data. For example, the statistical anomaly score can be determined using an unlabeled / weakly labeled sparse aggregation method, calculated according to the following formula. This refers to the weighted median, which is resistant to some overly aggressive algorithms. Since different algorithms perform differently in different scenarios—some have high false positive rates, while others have strong recall—a credibility weight is used. Higher weights can be given to stable and reliable algorithms, while lower weights can be given to algorithms that drift or frequently generate false alarms. For example, when it is desirable to define "high risk only when there is consensus across multiple parties" (soft AND behavior) and to suppress spikes in individual algorithms, an alternative weighted geometric mean can be used to suppress single-point maxima and emphasize the determination of statistical anomaly scores for multi-source government data by multiple algorithms. ,in, It is a pre-set minimum value (e.g.) )prevent An error occurred when taking the logarithm. This indicates that the credibility weights corresponding to each anomaly detection algorithm are normalized to ensure that the statistical anomaly score still falls within [0,1]. The advantage of this setting is that it determines the anomaly score based on a multi-algorithm pool of "sentinel, detective, and prophet" and combines it with dynamic reliable weights for robust aggregation, reducing the interference of a single algorithm spike on the overall judgment.

[0029] Optionally, the statistical anomaly score of the multi-source government data is calculated based on at least one anomaly detection algorithm in the anomaly detection algorithm pool, including: determining the initial anomaly score of the multi-source government data based on each anomaly detection algorithm in the anomaly detection algorithm pool; inputting the initial anomaly score corresponding to each anomaly detection algorithm and the metadata of the multi-source government data into a pre-trained scoring fusion model to obtain the statistical anomaly score of the multi-source government data output by the scoring fusion model.

[0030] For example, the metadata of multi-source government data is determined, including data volume, seasonal intensity / whether it is a holiday, time axis weight, SLA verification results, cooling-off status, indicator type, regional level, granularity, and other related data. The initial anomaly scores corresponding to each anomaly detection algorithm and the metadata of the multi-source government data are input into a pre-trained score fusion model, i.e., using... Metadata (such as sample size and seasonal intensity) is used as input to the scoring fusion model, enabling the model to analyze and fuse the inputs to output statistical anomaly scores for multi-source government data. Optionally, the scoring fusion model can be a Logistic / GBDT binary classifier. For example, the scoring fusion model uses Logistic (L2), with GBDT as an alternative (depth ≤ 3, leaves ≥ 20); time-block cross-validation and online calibration are used. Output For calibration, use Isotonic if the sample size is ≥100; otherwise, use Platt.

[0031] S130. Calculate the comprehensive risk score of the multi-source government data based on the consistency score and the statistical anomaly score.

[0032] In this embodiment of the application, a comprehensive risk score for multi-source government data is calculated based on the consistency score and the statistical anomaly score. For example, the weighted sum of the consistency score and the statistical anomaly score is used as the comprehensive risk score for multi-source government data.

[0033] S140. Determine the risk level of the multi-source government data based on the comprehensive risk score, and push out graded and frequency-based early warnings according to the risk level.

[0034] In this embodiment, a pre-defined mapping relationship between risk scores and risk levels is obtained, and the risk level of government data is determined based on this mapping relationship and the comprehensive risk score. For example, the risk level may include three levels: high, medium, and low. When the comprehensive risk score... satisfy When the overall risk score is high, it is considered a high-risk level; when the overall risk score is high, it is considered a high- satisfy At that time, it was classified as a medium-risk level; when the comprehensive risk score was... satisfy At any given time, the risk level is considered low. The higher the comprehensive risk score, the higher the corresponding risk level. Early warnings are pushed out in a tiered and frequency-based manner based on the risk level. For example, the recipients of the early warnings are determined according to a preset responsibility routing matrix. This responsibility routing matrix can be a mapping relationship between abnormal government data indicators and responsible entities, defined by at least four dimensions: department, data domain, region, and monitoring period. The target push method and target push frequency are determined based on the risk level, and the early warning information is sent to the recipients according to the target push method and target push frequency. The target push method can include SMS, email, system in-site messages, or other compatible internal communication tools.

[0035] The technical solution of this application embodiment acquires multi-source government data and determines the consistency score of the multi-source government data based on a pre-set consistency constraint rule library; calculates the statistical anomaly score of the multi-source government data based on at least one anomaly detection algorithm in the anomaly detection algorithm pool; calculates the comprehensive risk score of the multi-source government data based on the consistency score and the statistical anomaly score; determines the risk level of the multi-source government data based on the comprehensive risk score, and pushes graded and frequency-based early warnings according to the risk level. This solution can accurately identify anomalies in multi-source government data, reducing the workload of manual discovery and problem investigation.

[0036] Before determining the consistency score of the multi-source government data based on a pre-defined consistency constraint rule base, the method further includes: performing SLA verification on the multi-source government data to obtain the SLA verification result; determining the consistency score of the multi-source government data based on the pre-defined consistency constraint rule base, including: when the SLA verification result is that the SLA verification is passed, determining the consistency score of the multi-source government data based on the pre-defined consistency constraint rule base.

[0037] In this embodiment, SLA verification is performed on multi-source government data to obtain SLA verification results. For example, the arrival delay, integrity, and missing rate of the multi-source government data are determined, and it is judged whether the arrival delay, integrity, missing rate, and anomaly rate meet preset data validity conditions to perform quality and freshness SLA verification on the multi-source government data. For instance, it is judged whether the arrival delay is less than a preset delay threshold, whether the integrity is greater than a preset integrity threshold, whether the missing rate is less than a preset missing rate threshold, and whether the anomaly rate is less than a preset anomaly rate threshold. If all the above conditions are met simultaneously, the SLA verification result of the multi-source government data is determined to be successful; otherwise, the SLA verification result of the multi-source government data is determined to be unsuccessful. The arrival delay can be the difference between the time the multi-source government data is generated and the time it is entered into the database, or the deviation between the actual update time and the agreed update time. Completeness can be the ratio of data subjects that should be reported (such as departments, regions, and units) to the actual reporting subjects. The missing rate can be the proportion of key fields (such as amount, quantity, and code) with null values. The anomaly rate can be the proportion of anomalies in the multi-source government data whose value range, format rules, or enumeration values ​​are clearly unreasonable (such as negative amounts, proportions exceeding 100%, and illegal administrative division codes). When the SLA verification result is successful, it indicates that the acquired multi-source government data is valid. At this point, the consistency score of the multi-source government data is further determined based on a pre-set reconciliation constraint rule base. The advantage of this setup is that it can effectively distinguish between "data quality issues" and "real business anomalies," significantly improving the accuracy and reliability of government data anomaly monitoring.

[0038] Optionally, if the SLA verification result is that the SLA verification fails, at least one of the following processing mechanisms will be triggered: 1) Degradation processing: Under the premise of allowing degradation, the system automatically uses historical data or benchmark values ​​from the previous period to replace the current data to ensure that the monitoring process is not interrupted; 2) Delay warning: Send delay or quality margin warning information to the data management party to indicate that the data is abnormal but not completely invalid; 3) Circuit breaker processing: When the data quality or freshness is seriously substandard (such as completeness below the lower limit, large-scale missing key fields, etc.), a circuit breaker operation is executed, "this batch of data is invalid", and the batch of data is prevented from entering any subsequent analysis stage. Among them, in the circuit breaker scenario, the system will mark the batch of data as "invalid" and exclude it in the subsequent calculation process to ensure that it will not participate in the calculation of the comprehensive risk score, thereby avoiding "dirty data" from polluting the overall monitoring results.

[0039] In some embodiments, after performing SLA verification on the multi-source government data and obtaining the SLA verification result, the method further includes: determining the data volume of the multi-source government data and the time interval between the target time of receiving the multi-source government data and the last business warning time; determining the business warning threshold based on the SLA verification result, the data volume, and the time interval; and calculating the comprehensive risk score of the multi-source government data based on the consistency score and the statistical anomaly score, including: calculating the comprehensive risk score of the multi-source government data based on the consistency score, the statistical anomaly score, and the business warning threshold.

[0040] In this embodiment of the application, the data volume of multi-source government data is determined. and the time interval between the target time of receiving multi-source government data and the time of the last business warning. , where the time interval The target time for receiving multi-source government data is the time difference between the time of the previous business alert for the same data in the same region and period. The business alert threshold is determined based on SLA verification results, data volume, and time intervals. For example, the business alert threshold can be calculated using the following formula: Among them, SLA compliance refers to the SLA verification result of multi-source government data passing the verification. This represents the pre-set minimum sample size threshold. This indicates the preset cooldown time threshold. Specifically, if the SLA meets the target... , If at least one of them does not satisfy the condition, then A comprehensive risk score for multi-source government data is calculated based on consistency scores, statistical anomaly scores, and business early warning gating. For example, the comprehensive risk score can be calculated using the following formula: ,in, This indicates the score for statistical anomalies. This represents the pre-defined weighting coefficient for statistical anomaly scores. This represents the pre-defined weighting coefficient for the consistency score. Indicates the consistency score. This indicates a business early warning gate. It's understandable that... At this time, it can enter the data quality and delay early warning channel without triggering business early warning, but the statistical results are still retained, only when Business alerts are generated in a timely manner. The advantage of this setup is that data quality issues such as delays, missing data, and duplication can be independently entered into the quality channel closed loop without interfering with the business alert classification, ensuring that "the scope and quality are guaranteed first, and business risks are assessed later."

[0041] In some embodiments, before calculating the comprehensive risk score of the multi-source government data based on the consistency score, the statistical anomaly score, and the business early warning gating, the method further includes: determining the scenario information of the multi-source government data, and determining the time axis gain and the event axis gain based on the scenario information; determining the dual-axis dynamic weights based on the time axis gain and the event axis gain; and calculating the comprehensive risk score of the multi-source government data based on the consistency score, the statistical anomaly score, and the business early warning gating, including: calculating the comprehensive risk score of the multi-source government data based on the consistency score, the statistical anomaly score, the business early warning gating, and the dual-axis dynamic weights.

[0042] In this embodiment, scenario information from multi-source government data is determined. This scenario information includes information related to seasons, fiscal cycles, holidays, hot topics, special campaigns, and emergencies. Time axis gain is determined based on a preset weighting strategy and the scenario information from the multi-source government data. event axis gain Optionally, the preset initial time axis gain and initial event axis gain can be adjusted based on scenario information from multi-source government data. For example, the preset initial time axis gain can be increased before and after holidays, and the initial event axis gain can be increased during special operations or emergencies. The dual-axis dynamic weights are determined based on the time axis gain and event axis gain. For example, the dual-axis dynamic weights can be determined using the following formula: ,in, This indicates a dual-axis dynamic weight (also known as a time / event dual-axis weight). This represents the preset reference gain, which can be set to 1 by default. Indicates the time-axis gain. This represents the event axis gain. A comprehensive risk score for multi-source government data is calculated based on consistency scores, statistical anomaly scores, business early warning gating, and dual-axis dynamic weighting. For example, the comprehensive risk score can be calculated using the following formula: The advantage of this setup is that it allows for hard gating based on the data quality, volume, and cooling status of multi-source government data, and contextual modulation of risk scores through time and event-based dual-axis gain, thereby improving the reliability of judgments in scenarios such as holidays and special governance campaigns.

[0043] In some embodiments, after pushing tiered and frequency-based early warnings according to the risk level, the method further includes: obtaining feedback data from the early warning recipients regarding the received tiered and frequency-based early warning information; and adjusting the related parameters involved in determining the comprehensive risk score based on the feedback data. For example, receiving feedback data from the early warning recipients regarding the received tiered and frequency-based early warning information may include confirmation results of the tiered and frequency-based early warning information, processing methods for multi-source government data, and reasons for anomalies in the multi-source government data. Adjusting the related parameters involved in determining the comprehensive risk score based on the feedback data may include weights, thresholds, and constraint sensitivities, such as tiered thresholds, etc. , reconciliation tolerance Cooling threshold , Parameters such as causal edge confidence thresholds are included. For example, feedback data can be used as weak labels for the scoring fusion model to update and calibrate it, enabling online learning and continuous optimization. This allows for more accurate determination of statistical anomaly scores in subsequent government data based on the updated scoring fusion model. Specifically, after adjusting the correlation parameters, a sliding window and warm-start deployment strategy can be used to deploy the updated correlation parameters. Specifically, the new parameters or model version are first A / B compared in gray-scale groups. Once indicators such as false positive rate and late reporting loss meet the standards, the updated correlation parameters are fully released.

[0044] In some embodiments, after performing graded and frequency-based early warning pushes based on the risk level, the method further includes: obtaining a time-delay causal graph generated by periodic dynamic learning based on historical government data, and determining each candidate upstream indicator of the multi-source government data in the time-delay causal graph; for each candidate upstream indicator, calculating the causal contribution of the candidate upstream indicator to the existence of risk anomalies in the multi-source government data based on the time-delay causal graph; taking the candidate upstream indicator with the largest absolute value of the causal contribution as the source indicator causing the existence of risk anomalies in the multi-source government data, and determining the node link from the source indicator to the multi-source government data based on the time-delay causal graph, and taking the node link as the risk transmission path causing the existence of risk anomalies in the multi-source government data.

[0045] In this embodiment, a time-delay causal graph is obtained based on periodic dynamic learning of historical government data. This time-delay causal graph is a directed acyclic / cyclic graph with edge lags, used to describe the temporal transmission of causality from upstream to downstream. It is understood that the time-delay causal graph is a causal graph dynamically updated by a periodically executed offline task (e.g., monthly updates) to adapt to structural changes in the macroeconomy or policies. It should be noted that domain expert knowledge can be introduced as prior constraints during the dynamic learning process of the time-delay causal graph to improve the accuracy of its structure. Various candidate upstream indicators from multi-source government data are determined within the time-delay causal graph. In this context, each candidate upstream indicator can be understood as a parent node in the time-delay causal graph that has a direct or indirect relationship with the node containing the multi-source government data. For each candidate upstream indicator, an estimator is used to calculate the candidate upstream indicator based on the time-delay causal graph. Causal contribution of multi-source government data to potential risks and anomalies That is, in his father's collection Under the condition of approximate counterfactual: in, This indicates multi-source government data. Indicates when Take a certain value and When taking certain values, The true distribution probability, Indicates when Force when taking certain values Values hour The expected distribution probability. The estimator can be selected from locally linear, causal forest, or dual robust (DR).

[0046] The candidate upstream indicator with the largest absolute value of causal contribution is taken as the source indicator that causes risk anomalies in multi-source government data, that is, taking... The largest As a source indicator, a time-lag causal graph is used to identify the node links from the source indicator to multi-source government data, and these node links are considered as risk transmission paths leading to anomalies in the multi-source government data. Optionally, the time-lag distribution and confidence level of the source indicator can also be determined based on the time-lag causal graph. Optionally, the downstream impact range of the source indicator can also be determined based on the time-lag causal graph.

[0047] In some embodiments, the inference chain service can also be traced back and monitored, and data fragments, reconciliation formulas and residuals, causal chains, parameter versions, weights and gating decisions can be solidified to support version viewing and provide report query and export.

[0048] This application also provides a government data monitoring system. Figure 2 This is a schematic diagram of the structure of a government data monitoring system provided in an embodiment of this application, such as... Figure 2 As shown, the government data monitoring system includes a data access gateway, a quality and freshness SLA service module, a cross-validation constraint engine module, an anomaly monitoring engine module, a risk fusion and classification service module, an early warning routing and frequency control service module, a closed-loop governance and online learning module, and a backtracking and reasoning chain viewing service module. The system comprises the following modules: a data access gateway for receiving multi-source government data and aligning and standardizing the data sources; a quality and freshness SLA service module for performing SLA verification on multi-source government data and implementing degradation or circuit breaker control; a cross-validation constraint engine module for calculating the cross-validation consistency score of multi-source government data; an anomaly monitoring engine module for calculating the statistical anomaly score of multi-source government data using multiple algorithm pools; a risk fusion and grading service module for merging the cross-validation consistency score and the statistical anomaly score into a comprehensive risk score for multi-source government data and determining the corresponding risk level of the data; an early warning routing and frequency control service module for performing graded and frequency-based early warning pushes based on risk levels; a closed-loop governance and online learning module for receiving feedback and driving adaptive updates of models and rules; and a backtracking and inference chain viewing service module for solidifying and displaying data, cross-validation, causal chains, and decision paths.

[0049] The technical solution provided in this application integrates "statistical anomaly detection", "reconciliation constraint verification" and "causal learning and counterfactual assessment" in a closed loop within the same process, forming a full-chain early warning mechanism from discovery to explanation to root cause to treatment suggestions (quality judgment precedes risk judgment, and integration is then graded).

[0050] Figure 3 This is a schematic diagram of a government data monitoring device provided in an embodiment of this application. This device can execute the government data monitoring method provided in any embodiment of this application, and possesses the corresponding functional modules and beneficial effects for executing the method. Figure 3 As shown, the device includes: The consistency score determination module 310 is used to acquire multi-source government data and determine the consistency score of the multi-source government data based on a pre-set consistency constraint rule library. The statistical anomaly score calculation module 320 is used to calculate the statistical anomaly score of the multi-source government data based on at least one anomaly detection algorithm in the anomaly detection algorithm pool. The comprehensive risk score calculation module 330 is used to calculate the comprehensive risk score of the multi-source government data based on the consistency score and the statistical anomaly score. The early warning push module 340 is used to determine the risk level of the multi-source government data based on the comprehensive risk score, and to push early warnings in a graded and frequencyd manner according to the risk level.

[0051] Optional, the consistency score determination module is used for: Construct a cross-reference constraint diagram corresponding to the multi-source government data based on a pre-defined cross-reference constraint rule base; For each constraint involved in the cross-validation diagram, calculate the cross-validation violation degree corresponding to the cross-validation constraint. The consistency score of the multi-source government data is determined based on each of the aforementioned consistency violation rates.

[0052] Optional, a statistical anomaly score calculation module is used for: For each anomaly detection algorithm in the anomaly detection algorithm pool, the initial anomaly score of the multi-source government data is determined based on the anomaly detection algorithm, and the credibility weight of the anomaly detection algorithm is determined based on the offline playback index of historical government data by the anomaly detection algorithm. Based on a preset aggregation strategy, the initial anomaly scores and credibility weights corresponding to each anomaly detection algorithm are aggregated to generate statistical anomaly scores for the multi-source government data.

[0053] Optional, a statistical anomaly score calculation module is used for: For each anomaly detection algorithm in the anomaly detection algorithm pool, the initial anomaly score of the multi-source government data is determined based on the anomaly detection algorithm. The initial anomaly scores corresponding to each of the anomaly detection algorithms and the metadata of the multi-source government data are input into a pre-trained scoring fusion model to obtain the statistical anomaly scores of the multi-source government data output by the scoring fusion model.

[0054] Optional, also includes: The SLA verification result determination module is used to perform SLA verification on the multi-source government data before determining the consistency score of the multi-source government data based on a pre-set reconciliation constraint rule base, and obtain the SLA verification result. The consistency score determination module is used for: When the SLA verification result is that the SLA verification is passed, the consistency score of the multi-source government data is determined based on the pre-set reconciliation constraint rule base.

[0055] Optional, also includes: The time interval determination module is used to determine the amount of data in the multi-source government data and the time interval between the target time of receiving the multi-source government data and the last business warning time after performing SLA verification on the multi-source government data and obtaining the SLA verification result. The business early warning threshold determination module is used to determine the business early warning threshold based on the SLA verification result, the data volume, and the time interval. The comprehensive risk scoring calculation module includes: The comprehensive risk scoring calculation unit is used to calculate the comprehensive risk score of the multi-source government data based on the consistency score, the statistical anomaly score, and the business early warning gate.

[0056] Optional, also includes: The gain determination module is used to determine the scenario information of the multi-source government data before calculating the comprehensive risk score of the multi-source government data based on the consistency score, the statistical anomaly score and the business early warning gate, and to determine the time axis gain and event axis gain based on the scenario information. A dual-axis dynamic weight determination module is used to determine dual-axis dynamic weights based on the time axis gain and the event axis gain. The comprehensive risk scoring calculation unit is used for: The comprehensive risk score of the multi-source government data is calculated based on the consistency score, the statistical anomaly score, the business early warning gate, and the dual-axis dynamic weight.

[0057] Optional, also includes: The feedback data acquisition module is used to acquire feedback data from the warning recipients regarding the received tiered and frequency-based warning information after the risk level is classified and frequency-based warning is pushed out. The correlation parameter adjustment module is used to adjust the correlation parameters involved in determining the comprehensive risk score based on the feedback data.

[0058] Optional, also includes: The candidate upstream indicator determination module is used to obtain a time-delay causal graph generated by periodic dynamic learning based on historical government data after the risk level is graded and frequency-based early warning push, and to determine each candidate upstream indicator of the multi-source government data in the time-delay causal graph. The causal contribution calculation module is used to calculate the causal contribution of each candidate upstream indicator to the risk anomaly of the multi-source government data based on the time-delay causal graph. The risk transmission path determination module is used to take the candidate upstream indicator with the largest absolute value of the causal contribution as the source indicator that causes the multi-source government data to have risk anomalies, and to determine the node link from the source indicator to the multi-source government data based on the time-delay causal graph, and to take the node link as the risk transmission path that causes the multi-source government data to have risk anomalies.

[0059] The government data monitoring device provided in this application can execute a government data monitoring method provided in any embodiment of this application, and has the corresponding functional modules and beneficial effects of the method execution.

[0060] Figure 3 A schematic diagram of an electronic device 10, which can be used to implement embodiments of this application, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the application described and / or claimed herein.

[0061] like Figure 3 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0062] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of monitors, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0063] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as government data monitoring methods.

[0064] In some embodiments, the government data monitoring method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the government data monitoring method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the government data monitoring method by any other suitable means (e.g., by means of firmware).

[0065] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0066] Computer programs used to implement the methods of this application may be written in any combination of one or more programming languages. These computer programs may be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable government data monitoring device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0067] In the context of this application, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0068] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0069] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0070] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0071] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the government data monitoring method provided in any embodiment of this application.

[0072] In implementing the computer program product, computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or it can be connected to an external computer (e.g., via the Internet using an Internet service provider). It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired information of the technical solution of this application can be achieved, and this is not limited herein.

[0073] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for monitoring government data, characterized in that, The method includes: Acquire multi-source government data and determine the consistency score of the multi-source government data based on a pre-defined cross-validation constraint rule base; Calculate the statistical anomaly score of the multi-source government data based on at least one anomaly detection algorithm from the anomaly detection algorithm pool. The comprehensive risk score of the multi-source government data is calculated based on the consistency score and the statistical anomaly score. The risk level of the multi-source government data is determined based on the comprehensive risk score, and early warnings are pushed out in a graded and frequency-based manner according to the risk level.

2. The method according to claim 1, characterized in that, Based on a pre-defined reconciliation constraint rule base, the reconciliation consistency score of the multi-source government data is determined, including: Construct a cross-reference constraint diagram corresponding to the multi-source government data based on a pre-defined cross-reference constraint rule base; For each constraint involved in the cross-validation diagram, calculate the cross-validation violation degree corresponding to the cross-validation constraint. The consistency score of the multi-source government data is determined based on each of the aforementioned consistency violation rates.

3. The method according to claim 1, characterized in that, Based on at least one anomaly detection algorithm from the anomaly detection algorithm pool, the statistical anomaly score of the multi-source government data is calculated, including: For each anomaly detection algorithm in the anomaly detection algorithm pool, the initial anomaly score of the multi-source government data is determined based on the anomaly detection algorithm, and the credibility weight of the anomaly detection algorithm is determined based on the offline playback index of historical government data by the anomaly detection algorithm. Based on a preset aggregation strategy, the initial anomaly scores and credibility weights corresponding to each anomaly detection algorithm are aggregated to generate statistical anomaly scores for the multi-source government data.

4. The method according to claim 1, characterized in that, Based on at least one anomaly detection algorithm from the anomaly detection algorithm pool, the statistical anomaly score of the multi-source government data is calculated, including: For each anomaly detection algorithm in the anomaly detection algorithm pool, the initial anomaly score of the multi-source government data is determined based on the anomaly detection algorithm. The initial anomaly scores corresponding to each of the anomaly detection algorithms and the metadata of the multi-source government data are input into a pre-trained scoring fusion model to obtain the statistical anomaly scores of the multi-source government data output by the scoring fusion model.

5. The method according to claim 1, characterized in that, Before determining the consistency score of the multi-source government data based on a pre-defined consistency constraint rule base, the process also includes: SLA verification is performed on the multi-source government data to obtain the SLA verification results; Based on a pre-defined reconciliation constraint rule base, the reconciliation consistency score of the multi-source government data is determined, including: When the SLA verification result is that the SLA verification is passed, the consistency score of the multi-source government data is determined based on the pre-set reconciliation constraint rule base.

6. The method according to claim 5, characterized in that, After performing SLA verification on the multi-source government data and obtaining the SLA verification result, the process also includes: Determine the amount of multi-source government data and the time interval between the target time of receiving the multi-source government data and the time of the last business warning; The business early warning threshold is determined based on the SLA verification result, the data volume, and the time interval. The comprehensive risk score of the multi-source government data is calculated based on the consistency score and the statistical anomaly score, including: The comprehensive risk score of the multi-source government data is calculated based on the consistency score, the statistical anomaly score, and the business early warning gating.

7. The method according to claim 6, characterized in that, Before calculating the comprehensive risk score of the multi-source government data based on the consistency score, the statistical anomaly score, and the business early warning gating, the following steps are also included: Determine the scenario information of the multi-source government data, and determine the time axis gain and event axis gain based on the scenario information; The dual-axis dynamic weights are determined based on the time axis gain and the event axis gain. Based on the consistency score, the statistical anomaly score, and the business early warning gating, a comprehensive risk score for the multi-source government data is calculated, including: The comprehensive risk score of the multi-source government data is calculated based on the consistency score, the statistical anomaly score, the business early warning gate, and the dual-axis dynamic weight.

8. The method according to any one of claims 1-7, characterized in that, After issuing graded and frequency-based early warnings based on the aforementioned risk levels, the system also includes: Obtain feedback data from the receiving entities regarding the received graded and frequency-based early warning information; The relevant parameters involved in determining the comprehensive risk score are adjusted based on the feedback data.

9. The method according to claim 1, characterized in that, After issuing graded and frequency-based early warnings based on the aforementioned risk levels, the system also includes: Obtain a time-delay causal graph generated by periodic dynamic learning based on historical government data, and determine each candidate upstream indicator of the multi-source government data in the time-delay causal graph; For each candidate upstream indicator, the causal contribution of the candidate upstream indicator to the existence of risk anomalies in the multi-source government data is calculated based on the time-delay causal graph. The candidate upstream indicator with the largest absolute value of the causal contribution is taken as the source indicator that causes the multi-source government data to have risk anomalies. Based on the time-delay causal graph, the node link from the source indicator to the multi-source government data is determined, and the node link is taken as the risk transmission path that causes the multi-source government data to have risk anomalies.

10. A government data monitoring device, characterized in that, include: The consistency score determination module is used to acquire multi-source government data and determine the consistency score of the multi-source government data based on a pre-set consistency constraint rule library. The statistical anomaly score calculation module is used to calculate the statistical anomaly score of the multi-source government data based on at least one anomaly detection algorithm in the anomaly detection algorithm pool; The comprehensive risk score calculation module is used to calculate the comprehensive risk score of the multi-source government data based on the consistency score and the statistical anomaly score. The early warning push module is used to determine the risk level of the multi-source government data based on the comprehensive risk score, and to push early warnings in a graded and frequencyd manner according to the risk level.

11. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the government data monitoring method according to any one of claims 1-9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the government data monitoring method according to any one of claims 1-9.

Images