Abnormal accident monitoring method and system for a third-party data platform

Abstract

The present invention provides a method and system for monitoring abnormal accidents of a third-party data platform, and the method includes: collecting data source sub-scores and total scores of the third-party data platform for the data source, and pre-processing the data source based on the data source sub-scores and total scores; determining the monitoring dimension of the data source based on the pre-processing result, and monitoring the data source based on the monitoring dimension, and visualizing the monitoring result; obtaining screening indicators, and screening and locating the visualized data source based on the screening indicators to determine abnormal data accidents. By conducting all-round, multi-angle, and cross-time real-time monitoring of the data source where the accident occurred and the client organization that called the data source score, it is possible to timely understand the causes and effects of abnormal accidents, summarize the rules, and facilitate the prevention of the recurrence of accidents, and minimize the impact on the business, so as to ensure the normal and stable operation of each business line.

Description

Technical Field

[0001] The present invention relates to the technical field of data information monitoring, and in particular to an abnormal accident monitoring method and system for a third-party data platform. Background Art

[0002] At present, as the understanding of big data in various industries continues to deepen, the strategic significance and importance of big data are also constantly being demonstrated, and the completion of the strategy depends greatly on the stability of the product;

[0003] As product diversification gradually emerges, traditional database monitoring will not be able to detect potential problems in the system operation process in a timely manner, which is likely to cause database data loss and bring impact to the enterprise. In addition, the complexity and diversity of data increase the workload of operation and maintenance personnel, resulting in incomplete monitoring and low efficiency.

[0004] In order to better consolidate the stability of the product, it is necessary to import relevant business data into a visual monitoring system, draw corresponding conclusions through different indicators and data, and thus identify and solve problems in advance within a controllable time frame. Summary of the invention

[0005] The present invention provides an abnormal accident monitoring method and system for a third-party data platform, which is used to conduct all-round, multi-angle, and cross-time real-time monitoring of the data source where the accident occurred and the client organization that calls this data source for scoring, so as to timely understand the cause and impact of the abnormal accident, summarize the rules, and prevent the recurrence of the accident, and minimize the impact on the business, so as to ensure the normal and stable operation of each business line.

[0006] The present invention provides an abnormal accident monitoring method for a third-party data platform, comprising:

[0007] Step 1: Collect the data source sub-scores and total scores of the third-party data platform for the data source, and pre-process the data source based on the data source sub-scores and total scores;

[0008] Step 2: Determine the monitoring dimension of the data source based on the preprocessing result, monitor the data source based on the monitoring dimension, and visualize the monitoring result;

[0009] Step 3: Obtain screening indicators, and screen and locate the visualized data source based on the screening indicators to determine abnormal data accidents.

[0010] Preferably, in a method for monitoring abnormal accidents of a third-party data platform, in step 1, collecting the data source sub-scores and the total score of the third-party data platform for the data source includes:

[0011] Acquire a preset data collection task, and analyze the data collection task to obtain data attributes of the data source to be collected and a collection time period for the data source to be collected corresponding to the preset data collection task;

[0012] Determine the target third-party data platform corresponding to the data source to be collected based on the data attribute, and access the database of the target third-party data platform;

[0013] Based on the collection time period and the data attributes, the data source to be collected is matched from the database of the target third-party data platform, and the data source sub-score and the total score of the target third-party data platform for the data source to be collected are extracted.

[0014] Preferably, a method for monitoring abnormal accidents of a third-party data platform, extracting the data source sub-score and the total score of the target third-party data platform for the data source to be collected, includes:

[0015] Acquire the collected data source to be collected, and extract data features of the data source to be collected, wherein the data source to be collected is at least one type;

[0016] Clustering the data sources to be collected based on the data features, and determining a data center for each category based on the clustering results, wherein each category corresponds to a data center;

[0017] The Hamming distance between the data source to be collected and each data center is determined, and the corresponding data centers of the data source to be collected whose Hamming clustering is less than a preset threshold are classified into one category, and the data source sub-score and total score of the data source to be collected are respectively associated with the corresponding categories.

[0018] Preferably, in a method for monitoring abnormal accidents on a third-party data platform, in step 1, preprocessing the data source based on the data source sub-score and the total score includes:

[0019] Obtain the data source sub-score and total score of the data source from the third-party data platform, and at the same time, obtain the benchmark data source sub-score and benchmark total score of the data source from the third-party data platform;

[0020] Performing a difference operation on the data source sub-score and the total score of the data source with the corresponding benchmark data source sub-score and the benchmark total score, respectively, to obtain a target difference value, and comparing the target difference value with a preset difference threshold;

[0021] If there is a target difference greater than the preset difference threshold, it is determined that the data source sub-score and the total score of the current data source are abnormal, and the data source sub-score and the total score of the current data source are filled with preset specified values;

[0022] Otherwise, it is determined that the data source sub-score and the total score of the current data source are normal.

[0023] Preferably, in a method for monitoring abnormal accidents on a third-party data platform, in step 2, the monitoring dimensions of the data source are determined based on the preprocessing results, including: call volume, average share, hit rate, scorable rate, timeout rate, and number of affected institutions.

[0024] Preferably, a method for monitoring abnormal accidents of a third-party data platform, in step 2, determining the monitoring dimension of the data source based on the preprocessing result, monitoring the data source based on the monitoring dimension, and visualizing the monitoring result, includes:

[0025] Acquire the monitoring dimensions of the data source, and monitor the data source in different dimensions based on the monitoring dimensions to obtain the values ​​of the data source under different monitoring dimensions in the data source layer and the total score layer, wherein each monitoring dimension includes two groups of values: medium and small;

[0026] Based on the values, an overview diagram of abnormal accident phenomena and an overview diagram of abnormal accident impacts corresponding to the data source are drawn, and the specific steps include:

[0027] Among them, the overview diagram of abnormal accident phenomena corresponding to the data source is drawn, including:

[0028] Obtaining a data type of a data source, and generating a first visualization overview base map based on the data type, and obtaining values ​​of the data source under different monitoring dimensions in the data source layer and the total score layer respectively;

[0029] Dividing the visualization overview base map into a first visualization image area and a second visualization image area based on the values, and associating the values ​​of the data source in the data source layer and the total score layer with the first visualization image area and the second visualization image area;

[0030] Based on the association results, the values ​​of the data source in the data source layer and the total score layer are respectively converted into corresponding two-dimensional coordinates, and based on the two-dimensional coordinates, line graphs of the data source in different monitoring dimensions in the data source layer and the total score layer are obtained;

[0031] The line graphs under different monitoring dimensions are color-coded based on the first preset RGB color code, and the line graphs are superimposed with the corresponding first visualization image area and second visualization image area based on the setting results to obtain an overview of the abnormal accident phenomenon corresponding to the data source;

[0032] Among them, the overview diagram of abnormal accident impact corresponding to the data source is drawn, including:

[0033] Determine a data group to be displayed corresponding to the abnormal accident based on the values ​​of the data source under different monitoring dimensions in the data source layer and the total score layer, and generate a second visualization overview base map based on the data group to be displayed;

[0034] Extracting a display item type of the to-be-displayed data group, and dividing the second visualization overview base map into a first area, a second area, and a third area based on the display item type;

[0035] Generate an impact overview map of the abnormal accident based on the first area, generate a detailed chart of the impact overview map based on the second area, and generate a detailed report of the impact overview map based on the third area;

[0036] Based on the second preset RGB color number, corresponding color numbers are set for the impact overview map, the detailed chart of the impact overview map, and the detailed report of the impact overview map, respectively, and based on the setting result, they are merged with the second visualization overview base map to obtain the abnormal accident impact overview map;

[0037] The monitoring results are visualized based on the abnormal accident phenomenon overview diagram and the abnormal accident impact overview diagram.

[0038] Preferably, a method for monitoring abnormal accidents of a third-party data platform, in step 3, obtaining screening indicators, and screening and locating the data source after visualization based on the screening indicators to determine abnormal data accidents, includes:

[0039] Obtaining a visualization result after visualizing the monitoring result, and at the same time, obtaining a screening index for the monitoring result, and generating a target filter based on the screening index, wherein the screening index is at least one;

[0040] Determine the data source number and monitoring time length to be analyzed based on the filter, and determine the change trend of the data source sub-score and the total score corresponding to the data source number within the monitoring time length;

[0041] Determining an abnormal image region corresponding to an abnormal change trend in the change trend, and marking the abnormal image region based on a preset marking symbol;

[0042] Determine the abnormal accident phenomenon by determining the abnormal image area based on the marking result;

[0043] Extracting the target data source number of the abnormal data source corresponding to the abnormal accident phenomenon, and analyzing the impact of the abnormal accident through the target data source number based on the target filter to obtain the product code and product version of the abnormal product corresponding to the abnormal accident;

[0044] Determine the customer organization that retrieved the abnormal product within the monitoring time length based on the product code and product version, and determine the organization name of the customer organization;

[0045] Synchronously updating and displaying the product code, product version, client organization and client organization name in the visualization processing result, and highlighting the data source sub-score and total score of the corresponding data source in the visualization processing result based on the display result;

[0046] An abnormal accident monitoring report is generated based on the displayed result, and the monitoring report is transmitted to the management terminal.

[0047] Preferably, a method for monitoring abnormal accidents of a third-party data platform, transmitting the monitoring report to a management terminal, comprises:

[0048] Obtaining a monitoring report, and extracting first data and second data corresponding to abnormal accident phenomena and abnormal accident impacts in the monitoring report;

[0049] Extracting data attributes of the first data and the second data, and determining data types of the first data and the second data based on the data attributes;

[0050] Based on the data type, a historical data set of the same type is matched from a preset database, and the historical data set is trained to obtain an accident law prediction model;

[0051] Inputting the first data and the second data into the accident law prediction model for analysis to obtain the occurrence law of the abnormal accident, and setting an accident self-inspection strategy based on the occurrence law;

[0052] At the same time, the self-checking time interval of the accident self-checking strategy is determined based on the occurrence law of the abnormal accident, and the data source is self-checked according to the self-checking time interval based on the accident self-checking strategy.

[0053] Preferably, a method for monitoring abnormal accidents of a third-party data platform, transmitting the monitoring report to a management terminal, comprises:

[0054] Obtaining the obtained monitoring report, and extracting abnormal parameters corresponding to the abnormal accident described in the monitoring report;

[0055] Comparing the abnormal parameter with a preset threshold, and determining the abnormal level of the abnormal accident based on the comparison result;

[0056] Determining an alarm level for the abnormal accident based on the abnormal level, and determining a target alarm mode corresponding to the alarm level;

[0057] Send an alarm reminder to the management terminal based on the target alarm mode.

[0058] The present invention provides an abnormal accident monitoring system for a third-party data platform, comprising:

[0059] The data source processing module is used to collect the data source sub-scores and total scores of the data source from the third-party data platform, and pre-process the data source based on the data source sub-scores and total scores;

[0060] A visualization display module, used to determine the monitoring dimension of the data source based on the preprocessing result, monitor the data source based on the monitoring dimension, and visualize the monitoring result;

[0061] The positioning module is used to obtain screening indicators, and based on the screening indicators, screen and locate the data source after visualization processing to determine abnormal data accidents.

[0062] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0063] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0065] Figure 1 This is a flow chart of an abnormal accident monitoring method of a third-party data platform in an embodiment of the present invention;

[0066] Figure 2 It is a schematic diagram of an overview of the impact of abnormal accidents in a method for monitoring abnormal accidents on a third-party data platform in an embodiment of the present invention;

[0067] Figure 3 It is a structural diagram of an abnormal accident monitoring system of a third-party data platform in an embodiment of the present invention. DETAILED DESCRIPTION

[0068] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0069] Embodiment 1:

[0070] This embodiment provides a method for monitoring abnormal accidents on a third-party data platform. Figure 1 As shown, including:

[0071] Step 1: Collect the data source sub-scores and total scores of the third-party data platform for the data source, and pre-process the data source based on the data source sub-scores and total scores;

[0072] Step 2: Determine the monitoring dimension of the data source based on the preprocessing result, monitor the data source based on the monitoring dimension, and visualize the monitoring result;

[0073] Step 3: Obtain screening indicators, and screen and locate the visualized data source based on the screening indicators to determine abnormal data accidents.

[0074] In this embodiment, the data source may specifically be different client organizations, scoring products, model versions, and time point information.

[0075] In this embodiment, the data source sub-score and the total score are respectively used to evaluate the quality of the data status of the data source from different perspectives.

[0076] In this embodiment, preprocessing refers to filling in a specified value for an abnormal data source according to the data source sub-division. The abnormal data source may specifically be a data source with a scoring timeout and a customer group that is not hit.

[0077] In this embodiment, the monitoring dimensions include six major indicators, namely, call volume, average share, hit rate, scorable rate, timeout rate and number of affected institutions. The purpose is to analyze the data source from six dimensions, so as to achieve accurate judgment of abnormal accident data.

[0078] In this embodiment, visualization processing refers to displaying the monitoring values ​​of the data source under different monitoring dimensions through a line chart using different RGB color numbers. Specifically, it can be to visualize the six dimensional indicators through Tableau software.

[0079] In this embodiment, the screening indicators can be used to analyze the changes in the sub-scores and total scores of the data source in different time periods, so as to facilitate the accurate positioning of abnormal data production accidents. Specifically, it can be time, third-party data platform, sub-score, product service code, product version and customer organization.

[0080] In this embodiment, the abnormal data accident refers to a data segment in which the data source sub-score and the total score of the data source change too much.

[0081] The beneficial effect of the above technical solution is: through all-round, multi-angle, and cross-time real-time monitoring of the data source where the accident occurred and the client organization that calls this data source for scoring, it is possible to timely understand the causes and impacts of abnormal accidents, summarize the rules, and prevent the recurrence of accidents, and minimize the impact on the business, so as to ensure the normal and stable operation of each business line.

[0082] Embodiment 2:

[0083] Based on Example 1, this example provides an abnormal accident monitoring method for a third-party data platform. In step 1, the data source sub-scores and the total score of the data source of the third-party data platform are collected, including:

[0084] Acquire a preset data collection task, and analyze the data collection task to obtain data attributes of the data source to be collected and a collection time period for the data source to be collected corresponding to the preset data collection task;

[0085] Determine the target third-party data platform corresponding to the data source to be collected based on the data attribute, and access the database of the target third-party data platform;

[0086] Based on the collection time period and the data attributes, the data source to be collected is matched from the database of the target third-party data platform, and the data source sub-score and the total score of the target third-party data platform for the data source to be collected are extracted.

[0087] In this embodiment, the preset data collection task is set in advance and is used to characterize the type of data to be collected and the corresponding data collection time period.

[0088] In this embodiment, data attributes refer to the types of data of the data source to be collected and the association relationship between data.

[0089] In this embodiment, the collection time period refers to the data source that needs to be collected from the third-party data platform for a predetermined time period.

[0090] In this embodiment, the target third-party data platform may be a third-party data platform that includes the data source to be collected.

[0091] The beneficial effect of the above technical solution is: by analyzing the preset data collection tasks, accurate and effective analysis of the data characteristics of the data source to be collected and the collection time period can be achieved, thereby facilitating accurate and effective collection of the data source to be collected, and providing convenience and guarantee for all-round, multi-angle, and cross-time real-time monitoring, so as to timely understand the causes and impacts of abnormal accidents.

[0092] Embodiment 3:

[0093] Based on Example 2, this example provides a method for monitoring abnormal accidents of a third-party data platform, which extracts the data source sub-scores and the total score of the target third-party data platform for the data source to be collected, including:

[0094] Acquire the collected data source to be collected, and extract data features of the data source to be collected, wherein the data source to be collected is at least one type;

[0095] Clustering the data sources to be collected based on the data features, and determining a data center for each category based on the clustering results, wherein each category corresponds to a data center;

[0096] The Hamming distance between the data source to be collected and each data center is determined, and the corresponding data centers of the data source to be collected whose Hamming clustering is less than a preset threshold are classified into one category, and the data source sub-score and total score of the data source to be collected are respectively associated with the corresponding categories.

[0097] In this embodiment, the data feature may be key data such as the value characteristics of the data that can clearly indicate the data feature.

[0098] In this embodiment, the Hamming distance is used to characterize the distance relationship between different data and the data center. When the Hamming distance is less than a preset threshold, it is determined that the data source to be collected corresponds to the current data center.

[0099] In this embodiment, the preset threshold is set in advance and is used to measure whether the classification requirements are met.

[0100] The beneficial effect of the above technical solution is: by extracting the data features of the data sources to be collected and clustering the data sources to be collected according to the data features, the data sources to be collected can be accurately and effectively classified, thereby facilitating the all-round, multi-angle and cross-time real-time monitoring of the data sources where accidents occur and the client organizations that call this data source for scoring.

[0101] Embodiment 4:

[0102] Based on Example 1, this example provides an abnormal accident monitoring method for a third-party data platform. In step 1, the data source is preprocessed based on the data source sub-score and the total score, including:

[0103] Obtain the data source sub-score and total score of the data source from the third-party data platform, and at the same time, obtain the benchmark data source sub-score and benchmark total score of the data source from the third-party data platform;

[0104] Performing a difference operation on the data source sub-score and the total score of the data source with the corresponding benchmark data source sub-score and the benchmark total score, respectively, to obtain a target difference value, and comparing the target difference value with a preset difference threshold;

[0105] If there is a target difference greater than the preset difference threshold, it is determined that the data source sub-score and the total score of the current data source are abnormal, and the data source sub-score and the total score of the current data source are filled with preset specified values;

[0106] Otherwise, it is determined that the data source sub-score and the total score of the current data source are normal.

[0107] In this embodiment, the benchmark data source sub-score and the benchmark total score refer to the standard data source sub-score and total score given to different customers and different products by the third-party data platform when there are no abnormal circumstances.

[0108] In this embodiment, the target difference refers to the difference between the data source sub-score and the total score of the data source and the corresponding benchmark data source sub-score and the total score.

[0109] In this embodiment, the preset difference threshold is set in advance and is used to measure whether the difference exceeds the maximum allowable range.

[0110] In this embodiment, the preset value is set in advance and is used to replace the abnormal data source sub-score and total score when there are abnormalities in the data source sub-score and the total score.

[0111] The beneficial effect of the above technical solution is: by verifying the data source sub-scores and total scores of the acquired data sources, the abnormal data source sub-scores and total scores can be corrected, thereby improving the accuracy of all-round, multi-angle, and cross-time real-time monitoring of the data source where the accident occurred and the client organization that called this data source score.

[0112] Embodiment 5:

[0113] Based on Example 1, this example provides a method for monitoring abnormal accidents on a third-party data platform. Figure 2 As shown, in step 2, the monitoring dimension of the data source is determined based on the preprocessing result, and the data source is monitored based on the monitoring dimension, and the monitoring result is visualized, including:

[0114] Acquire the monitoring dimensions of the data source, and monitor the data source in different dimensions based on the monitoring dimensions to obtain the values ​​of the data source under different monitoring dimensions in the data source layer and the total score layer, wherein each monitoring dimension includes two groups of values: medium and small;

[0115] Based on the values, an overview diagram of abnormal accident phenomena and an overview diagram of abnormal accident impacts corresponding to the data source are drawn, and the specific steps include:

[0116] Among them, the overview diagram of abnormal accident phenomena corresponding to the data source is drawn, including:

[0117] Obtaining a data type of a data source, and generating a first visualization overview base map based on the data type, and obtaining values ​​of the data source under different monitoring dimensions in the data source layer and the total score layer respectively;

[0118] Dividing the visualization overview base map into a first visualization image area and a second visualization image area based on the values, and associating the values ​​of the data source in the data source layer and the total score layer with the first visualization image area and the second visualization image area;

[0119] Based on the association results, the values ​​of the data source in the data source layer and the total score layer are respectively converted into corresponding two-dimensional coordinates, and based on the two-dimensional coordinates, line graphs of the data source in different monitoring dimensions in the data source layer and the total score layer are obtained;

[0120] The line graphs under different monitoring dimensions are color-coded based on the first preset RGB color code, and the line graphs are superimposed with the corresponding first visualization image area and second visualization image area based on the setting results to obtain an overview of the abnormal accident phenomenon corresponding to the data source;

[0121] Among them, the overview diagram of abnormal accident impact corresponding to the data source is drawn, including:

[0122] Determine a data group to be displayed corresponding to the abnormal accident based on the values ​​of the data source under different monitoring dimensions in the data source layer and the total score layer, and generate a second visualization overview base map based on the data group to be displayed;

[0123] Extracting a display item type of the to-be-displayed data group, and dividing the second visualization overview base map into a first area, a second area, and a third area based on the display item type;

[0124] Generate an impact overview map of the abnormal accident based on the first area, generate a detailed chart of the impact overview map based on the second area, and generate a detailed report of the impact overview map based on the third area;

[0125] Based on the second preset RGB color number, corresponding color numbers are set for the impact overview map, the detailed chart of the impact overview map, and the detailed report of the impact overview map, respectively, and based on the setting result, they are merged with the second visualization overview base map to obtain the abnormal accident impact overview map;

[0126] The monitoring results are visualized based on the abnormal accident phenomenon overview diagram and the abnormal accident impact overview diagram.

[0127] In this embodiment, the values ​​of the data source under different monitoring dimensions in the data source layer and the total score layer refer to the call volume, average score, hit rate, scorable rate, timeout rate, number of affected institutions, etc. corresponding to the data source.

[0128] In this embodiment, the overview diagram of abnormal accident phenomena refers to the specific values ​​of the call volume, average score, hit rate, scorable rate, timeout rate, and number of affected institutions corresponding to the data source in the data source layer and the total score layer.

[0129] In this embodiment, the abnormal accident impact overview diagram refers to the number of affected institutions caused by the data source, the total score call volume during the abnormal period, the proportion of the total score during the abnormal period to the current total score call volume, the total score scoring rate during the abnormal period, and the difference between the average total score during the abnormal period and the average total score of the day.

[0130] In this embodiment, the data type may be data such as customer organization, product type, and customer code contained in the data source.

[0131] In this embodiment, the first visualization overview base map refers to the corresponding background image when the abnormal accident phenomenon overview map is generated.

[0132] In this embodiment, the first visualization image area refers to the area used to display the specific value corresponding to the data source in the data source layer.

[0133] In this embodiment, the second visualization image area refers to the specific value corresponding to the tactical data source at the total score layer.

[0134] In this embodiment, the first preset RGB color number is set in advance, and can be specifically 123, 184, 179, 246, 140, 36, 73, 122, 162, 253, 175, 183, 249, 210, 60, 233, 235, 224; 123, 184, 179, 246, 140, 36, 73, 122, 162.

[0135] In this embodiment, the data group to be displayed refers to the type of project that needs to be displayed in the abnormal accident impact overview diagram, which may be the number of affected institutions, the total score call volume during the abnormal period, the proportion of the total score during the abnormal period to the current total score call volume, the total score scoring rate during the abnormal period, and the difference between the average total score during the abnormal period and the average total score of the day.

[0136] In this embodiment, the second visualization overview base map refers to the background image corresponding to the abnormal accident phenomenon overview map, that is, the image template.

[0137] In this embodiment, the display item type refers to the data content that needs to be displayed in the abnormal accident impact overview map, wherein the second visualization overview base map is divided into three parts according to the display item type.

[0138] In this embodiment, the first area, the second area and the third area refer to image areas corresponding to each part after the second visualization overview base map is divided.

[0139] In this embodiment, the impact overview chart refers to the total statistical values ​​corresponding to the number of affected institutions, the total score call volume during the abnormal period, the proportion of the total score during the abnormal period to the current total score call volume, the total score scoring rate during the abnormal period, and the difference between the average total score during the abnormal period and the average total score of the day.

[0140] In this embodiment, the detailed chart refers to the specific value of each display item and the specific difference between the current value and the standard value.

[0141] In this embodiment, the detailed report refers to generating a corresponding table of the total score call volume during the abnormal period, the proportion of the total score during the abnormal period to the current total score call volume, the total score scoring rate during the abnormal period, and the difference between the average total score during the abnormal period and the average total score of the day, so as to achieve accurate statistical recording of the display items.

[0142] In this embodiment, the second preset RGB color number is set in advance, and can be specifically 227, 155, 154, 150, 149, 147, 166, 1, 17, 73, 122, 162; 215, 181, 166, 186, 149, 130, 186, 176, 172, 151, 121, 94; 235, 231, 227, 45, 42, 42, 73, 122, 162.

[0143] The beneficial effect of the above technical solution is: by determining the values ​​of the data under different monitoring dimensions in the data source layer and the total score respectively, an overview diagram of abnormal accident phenomena and an overview diagram of abnormal accident impacts corresponding to the data source are generated, which facilitates an intuitive and effective understanding of the values ​​of the data source under different monitoring dimensions, thereby facilitating an accurate summary of the causes and impacts of abnormal accidents and ensuring the monitoring effect of the data source.

[0144] Embodiment 6:

[0145] Based on Example 1, this example provides a method for monitoring abnormal accidents on a third-party data platform. In step 3, a screening index is obtained, and the data source after visualization is screened and located based on the screening index to determine the abnormal data accident, including:

[0146] Obtaining a visualization result after visualizing the monitoring result, and at the same time, obtaining a screening index for the monitoring result, and generating a target filter based on the screening index, wherein the screening index is at least one;

[0147] Determine the data source number and monitoring time length to be analyzed based on the filter, and determine the change trend of the data source sub-score and the total score corresponding to the data source number within the monitoring time length;

[0148] Determining an abnormal image region corresponding to an abnormal change trend in the change trend, and marking the abnormal image region based on a preset marking symbol;

[0149] Determine the abnormal accident phenomenon by determining the abnormal image area based on the marking result;

[0150] Extracting the target data source number of the abnormal data source corresponding to the abnormal accident phenomenon, and analyzing the impact of the abnormal accident through the target data source number based on the target filter, to obtain the product code and product version of the abnormal product corresponding to the abnormal accident;

[0151] Determine the customer organization that retrieved the abnormal product within the monitoring time length based on the product code and product version, and determine the organization name of the customer organization;

[0152] Synchronously updating and displaying the product code, product version, client organization and client organization name in the visualization processing result, and highlighting the data source sub-score and total score of the corresponding data source in the visualization processing result based on the display result;

[0153] An abnormal accident monitoring report is generated based on the displayed result, and the monitoring report is transmitted to the management terminal.

[0154] In this embodiment, the target filter is generated according to the screening index, and is used to set the product code or customer name etc. to be analyzed in the filter, so as to analyze and screen the corresponding data through the target filter.

[0155] In this embodiment, the data source number to be analyzed and the monitoring time length refer to the data source used to determine the data source that needs to be analyzed. Each data source corresponds to a different number, and the values ​​of the data sources corresponding to different time periods are different. The data source number and the monitoring time length are used to quickly determine the data source segment to be analyzed from a large number of data sources.

[0156] In this embodiment, the change trend is used to characterize the value changes of the data source sub-score and the total score corresponding to the data source within the monitoring time period, so as to facilitate the analysis of whether there is any abnormal accident in the data source at the end.

[0157] In this embodiment, the abnormal change trend refers to that the values ​​of the data source sub-score and the total score corresponding to the data source fluctuate greatly during the monitoring period, and the fluctuation amplitude exceeds the preset threshold.

[0158] In this embodiment, the abnormal image value refers to the image area corresponding to the abnormal change trend.

[0159] In this embodiment, the preset marking symbol is set in advance and is used to mark the abnormal image area.

[0160] In this embodiment, the target data source code refers to the coding information corresponding to the data source where the abnormal accident occurred.

[0161] In this embodiment, abnormal products refer to data information called by different customers through a third-party data platform.

[0162] In this embodiment, highlighting may be performed by color marking or symbol marking, and the highlighting method is already set in advance.

[0163] The beneficial effects of the above technical solution are: by analyzing the obtained visualization processing results, the abnormal image area of ​​the data source in the visualization image can be accurately locked; secondly, by analyzing the abnormal image area, the coding information of the data source where the abnormal accident occurred and the number of client institutions called through the third-party data platform during the monitoring period and the name of the client institutions can be accurately and effectively locked; finally, the screening and positioning results are synchronously updated and displayed in the visualization image, realizing all-round, multi-angle, and cross-time real-time monitoring of the data source, ensuring the accuracy and efficiency of the monitoring, and minimizing the impact on the business to ensure the normal and stable operation of each business line.

[0164] Embodiment 7:

[0165] Based on Example 6, this embodiment provides an abnormal accident monitoring method for a third-party data platform, which transmits the monitoring report to a management terminal, including:

[0166] Obtaining a monitoring report, and extracting first data and second data corresponding to abnormal accident phenomena and abnormal accident impacts in the monitoring report;

[0167] Extracting data attributes of the first data and the second data, and determining data types of the first data and the second data based on the data attributes;

[0168] Based on the data type, a historical data set of the same type is matched from a preset database, and the historical data set is trained to obtain an accident law prediction model;

[0169] Inputting the first data and the second data into the accident law prediction model for analysis to obtain the occurrence law of the abnormal accident, and setting an accident self-inspection strategy based on the occurrence law;

[0170] At the same time, the self-checking time interval of the accident self-checking strategy is determined based on the occurrence law of the abnormal accident, and the data source is self-checked according to the self-checking time interval based on the accident self-checking strategy.

[0171] In this embodiment, the first data refers to data information corresponding to the abnormal accident phenomenon.

[0172] In this embodiment, the second data refers to data information corresponding to the impact of the abnormal accident.

[0173] In this embodiment, the data attributes are suitable for characterizing the data characteristics of the first data and the second data, and specifically may be the association relationship between the data and the corresponding values ​​of the data.

[0174] In this embodiment, the historical data set is set in advance and is used to generate an accident pattern prediction model.

[0175] In this embodiment, the accident self-check strategy refers to a method for screening data sources formulated according to the occurrence law of abnormal accidents.

[0176] In this embodiment, the self-check time interval is used to represent the length of the time interval for screening the data source through the accident self-check strategy.

[0177] The beneficial effects of the above technical solution are: by determining the data corresponding to abnormal accident phenomena and the impact of abnormal accidents, an accident law prediction model is obtained by training according to the historical data set, and the data corresponding to the abnormal accident phenomena and the impact of abnormal accidents are analyzed through the accident law prediction model, so as to achieve accurate and effective acquisition of the laws of abnormal accidents, and formulate corresponding accident self-inspection strategies according to the laws of abnormal accidents, realize the strictness of monitoring of data sources and the accuracy of monitoring, and minimize the impact on the business, so as to ensure the normal and stable operation of each business line.

[0178] Embodiment 8:

[0179] Based on Example 6, this embodiment provides an abnormal accident monitoring method for a third-party data platform, which transmits the monitoring report to a management terminal, including:

[0180] Obtaining the obtained monitoring report, and extracting abnormal parameters corresponding to the abnormal accident described in the monitoring report;

[0181] Comparing the abnormal parameter with a preset threshold, and determining the abnormal level of the abnormal accident based on the comparison result;

[0182] Determining an alarm level for the abnormal accident based on the abnormal level, and determining a target alarm mode corresponding to the alarm level;

[0183] Send an alarm reminder to the management terminal based on the target alarm mode.

[0184] In this embodiment, the preset threshold is set in advance and is used to measure the alarm level for abnormal accidents.

[0185] In this embodiment, determining the abnormal level of the abnormal accident based on the comparison result means that when the abnormal parameter is less than a preset threshold, it is a first alarm level, and when the abnormal parameter is greater than or equal to the preset threshold, it is a second alarm level.

[0186] In this embodiment, the target alarm mode is set in advance, and different alarm levels correspond to different alarm modes.

[0187] The beneficial effect of the above technical solution is: by comparing the parameters corresponding to the abnormal accident with the preset threshold, the alarm level of the abnormal accident can be accurately and effectively obtained, so as to facilitate the corresponding alarm according to the alarm level, realize the monitoring of the data source, and facilitate the management personnel to timely understand the severity level of the current abnormal accident, so as to minimize the impact on the business and ensure the normal and stable operation of each business line.

[0188] Embodiment 9:

[0189] Based on the above-mentioned embodiment 1, this embodiment provides a method for monitoring abnormal accidents of a third-party data platform, which screens and locates the data source after visualization based on the screening index to determine the abnormal data accident, including:

[0190] Obtaining a preset standard threshold interval, and comparing the data source with the corresponding preset standard threshold interval respectively, and obtaining an abnormal data source whose value of the data source is not within the preset standard threshold interval based on the comparison result;

[0191] Determine the data type of the abnormal data source, and verify the abnormal data source by matching the target data verification strategy from a preset data verification library based on the data type, to obtain the total amount of data of the abnormal data source and the number of incorrectly screened data sources;

[0192] The accuracy of data source screening and positioning is calculated based on the total amount of data of the abnormal data source and the number of incorrectly screened data sources, and a comprehensive evaluation value of the monitoring effect of the data source is calculated based on the accuracy. The specific steps include:

[0193] The accuracy of data source screening and positioning is calculated according to the following formula:

[0194]

[0195] Wherein, η represents the accuracy of data source screening and positioning, and its value range is (0, 1); μ represents the error factor, and its value range is (0.02, 0.05); M represents the total amount of data of the abnormal data source; m represents the number of incorrectly screened data sources in the abnormal data source, and its value is less than M; s represents the number of accurate data sources obtained after data source screening; K represents the sum of the data volumes of all data sources;

[0196] Obtaining the screening speed of each type of data source and the total amount of data of each type of data source, and performing dimensionless processing on the screening speed and the total amount of data of each type of data source to obtain a screening speed parameter and a total amount of data parameter of each type of data source;

[0197] The comprehensive evaluation value of the monitoring effect on the data source is calculated according to the following formula:

[0198]

[0199] in, Comprehensive evaluation value of the monitoring effect of the data source; i represents the data source of the current category, and its value range is [1, n]; n represents the total number of data source categories; α represents the weight value corresponding to the data source screening accuracy; η represents the accuracy of data source screening and positioning, and its value range is (0, 1); β represents the weight value of the data source screening speed parameter; v i represents the screening speed parameter for the i-th type of data source screening, dimensionless; ρ represents the weight value of the total amount of data source data parameter; Q i It represents the total amount of data of the i-th data source, dimensionless;

[0200] Compare the calculated comprehensive evaluation value of monitoring effect with the preset evaluation threshold; i

[0201] If the comprehensive evaluation value of the monitoring effect is greater than or equal to the preset evaluation threshold, it is determined that the monitoring effect on the data source is qualified;

[0202] Otherwise, it is determined that the monitoring effect on the data source is unqualified, and the data source is screened and relocated until the comprehensive evaluation value of the monitoring effect is greater than or equal to the preset evaluation threshold.

[0203] In this embodiment, the preset standard threshold interval is set in advance and is used to represent a normal value corresponding to a normal data source.

[0204] In this embodiment, an abnormal data source refers to a data source whose value is not within a preset standard threshold range.

[0205] In this embodiment, the preset data verification library is set in advance and is used to store different data verification strategies.

[0206] In this embodiment, the target data verification strategy refers to a verification strategy suitable for performing a secondary verification on the current abnormal data source, the purpose of which is to determine whether there is an erroneously filtered data source in the obtained abnormal data source.

[0207] In this embodiment, the erroneously screened data source refers to determining a normal data source as an abnormal data source.

[0208] In this embodiment, the comprehensive evaluation value of monitoring effect is used to characterize the monitoring effect on the data source. The larger the value, the better the monitoring intensity and monitoring effect on the data source.

[0209] In this embodiment, the preset evaluation threshold is set in advance and is used to measure whether the comprehensive evaluation value of the monitoring effect meets the expected requirements.

[0210] The beneficial effect of the above technical solution is: by calculating the screening accuracy of the data source, the comprehensive evaluation value of the monitoring effect of the data source can be effectively calculated, so that the user can grasp the monitoring effect of the data source in real time according to the comprehensive evaluation value of the monitoring effect of the data source, so as to facilitate timely checking whether there are any abnormal accidents, thereby improving the timeliness and accuracy of monitoring.

[0211] Embodiment 10:

[0212] This embodiment provides an abnormal accident monitoring system for a third-party data platform, such as Figure 3 As shown, including:

[0213] The data source processing module is used to collect the data source sub-scores and total scores of the data source from the third-party data platform, and pre-process the data source based on the data source sub-scores and total scores;

[0214] A visualization display module, used to determine the monitoring dimension of the data source based on the preprocessing result, monitor the data source based on the monitoring dimension, and visualize the monitoring result;

[0215] The positioning module is used to obtain screening indicators, and based on the screening indicators, screen and locate the data source after visualization processing to determine abnormal data accidents.

[0216] The beneficial effect of the above technical solution is: through all-round, multi-angle, and cross-time real-time monitoring of the data source where the accident occurred and the client organization that calls this data source for scoring, it is possible to timely understand the causes and impacts of abnormal accidents, summarize the rules, and prevent the recurrence of accidents, and minimize the impact on the business, so as to ensure the normal and stable operation of each business line.

[0217] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for monitoring abnormal accidents on a third-party data platform, characterized in that: include: Step 1: Collect the data source sub-scores and total scores of the third-party data platform for the data source, and pre-process the data source based on the data source sub-scores and total scores; Step 2: Determine the monitoring dimension of the data source based on the preprocessing result, monitor the data source based on the monitoring dimension, and visualize the monitoring result; Step 3: Obtain screening indicators, and screen and locate the data source after visualization based on the screening indicators to determine abnormal data accidents; Among them, the data source after visualization processing is screened and located based on the screening indicators to determine abnormal data accidents, including: Obtaining a preset standard threshold interval, and comparing the data source with the corresponding preset standard threshold interval respectively, and obtaining an abnormal data source whose value of the data source is not within the preset standard threshold interval based on the comparison result; Determine the data type of the abnormal data source, and verify the abnormal data source by matching the target data verification strategy from a preset data verification library based on the data type, to obtain the total amount of data of the abnormal data source and the number of incorrectly screened data sources; The accuracy of data source screening and positioning is calculated based on the total amount of data of the abnormal data source and the number of incorrectly screened data sources, and a comprehensive evaluation value of the monitoring effect of the data source is calculated based on the accuracy. The specific steps include: The accuracy of data source screening and positioning is calculated according to the following formula: ； in, Indicates the accuracy of data source screening and positioning, and the value range is (0, 1); Represents the error factor, and its value range is (0.02, 0.05); Indicates the total amount of data from the abnormal data source; Indicates the number of incorrectly filtered data sources in the abnormal data source, and the value is less than ; Indicates the number of accurate data sources obtained after filtering the data sources; Indicates the sum of the data volumes of all data sources; Obtaining the screening speed of each type of data source and the total amount of data of each type of data source, and performing dimensionless processing on the screening speed and the total amount of data of each type of data source to obtain a screening speed parameter and a total amount of data parameter of each type of data source; The comprehensive evaluation value of the monitoring effect on the data source is calculated according to the following formula: ； in, Comprehensive evaluation value of monitoring effect on data sources; Indicates the category of the current data source, and its value range is [1, ]; Indicates the total number of categories in the data source; Indicates the weight value corresponding to the accuracy of data source screening; Indicates the accuracy of data source screening and positioning, and the value range is (0, 1); Indicates the weight value of the data source filtering speed parameter; Indicates The filtering speed parameter of the class data source filtering, dimensionless; Indicates the weight value of the total amount of data in the data source; Indicates The total amount of data in the class data source, dimensionless; Compare the calculated comprehensive evaluation value of monitoring effect with the preset evaluation threshold; If the comprehensive evaluation value of the monitoring effect is greater than or equal to the preset evaluation threshold, it is determined that the monitoring effect on the data source is qualified; Otherwise, it is determined that the monitoring effect on the data source is unqualified, and the data source is screened and relocated until the comprehensive evaluation value of the monitoring effect is greater than or equal to the preset evaluation threshold.

2. According to claim 1, a method for monitoring abnormal accidents of a third-party data platform is characterized in that: In step 1, collect the data source sub-scores and total scores of the third-party data platform for the data source, including: Acquire a preset data collection task, and analyze the data collection task to obtain data attributes of the data source to be collected and a collection time period for the data source to be collected corresponding to the preset data collection task; Determine the target third-party data platform corresponding to the data source to be collected based on the data attribute, and access the database of the target third-party data platform; Based on the collection time period and the data attributes, the data source to be collected is matched from the database of the target third-party data platform, and the data source sub-score and the total score of the target third-party data platform for the data source to be collected are extracted.

3. The abnormal accident monitoring method of a third-party data platform according to claim 2 is characterized in that: Extract the data source sub-score and total score of the target third-party data platform for the data source to be collected, including: Acquire the collected data source to be collected, and extract data features of the data source to be collected, wherein the data source to be collected is at least one type; Clustering the data sources to be collected based on the data features, and determining a data center for each category based on the clustering results, wherein each category corresponds to a data center; The Hamming distance between the data source to be collected and each data center is determined, and the corresponding data center of the data source to be collected whose Hamming distance is less than a preset threshold is classified into one category, and the data source sub-score and the total score of the data source to be collected are respectively associated with the corresponding category.

4. The abnormal accident monitoring method of a third-party data platform according to claim 1 is characterized in that: In step 1, the data source is preprocessed based on the data source sub-score and the total score, including: Obtain the data source sub-score and total score of the data source from the third-party data platform, and at the same time, obtain the benchmark data source sub-score and benchmark total score of the data source from the third-party data platform; Performing a difference operation on the data source sub-score and the total score of the data source with the corresponding benchmark data source sub-score and the benchmark total score, respectively, to obtain a target difference value, and comparing the target difference value with a preset difference threshold; If there is a target difference greater than the preset difference threshold, it is determined that the data source sub-score and the total score of the current data source are abnormal, and the data source sub-score and the total score of the current data source are filled with preset specified values; Otherwise, it is determined that the data source sub-score and the total score of the current data source are normal.

5. The abnormal accident monitoring method of a third-party data platform according to claim 1 is characterized in that: In step 2, the monitoring dimensions of the data source are determined based on the preprocessing results, including: call volume, average share, hit rate, scorable rate, timeout rate, and number of affected institutions.

6. The abnormal accident monitoring method of a third-party data platform according to claim 1 is characterized in that: In step 2, the monitoring dimension of the data source is determined based on the preprocessing result, the data source is monitored based on the monitoring dimension, and the monitoring result is visualized, including: Acquire the monitoring dimensions of the data source, and monitor the data source in different dimensions based on the monitoring dimensions to obtain the values ​​of the data source under different monitoring dimensions in the data source layer and the total score layer, wherein each monitoring dimension includes two groups of values: medium and small; Based on the values, an overview diagram of abnormal accident phenomena and an overview diagram of abnormal accident impacts corresponding to the data source are drawn, and the specific steps include: Among them, the overview diagram of abnormal accident phenomena corresponding to the data source is drawn, including: Obtaining a data type of a data source, and generating a first visualization overview base map based on the data type, and obtaining values ​​of the data source under different monitoring dimensions in the data source layer and the total score layer respectively; Dividing the visualization overview base map into a first visualization image area and a second visualization image area based on the values, and associating the values ​​of the data source in the data source layer and the total score layer with the first visualization image area and the second visualization image area; Based on the association results, the values ​​of the data source in the data source layer and the total score layer are respectively converted into corresponding two-dimensional coordinates, and based on the two-dimensional coordinates, line graphs of the data source in different monitoring dimensions in the data source layer and the total score layer are obtained; The line graphs under different monitoring dimensions are color-coded based on the first preset RGB color code, and the line graphs are superimposed with the corresponding first visualization image area and second visualization image area based on the setting results to obtain an overview of the abnormal accident phenomenon corresponding to the data source; Among them, the overview diagram of abnormal accident impact corresponding to the data source is drawn, including: Determine a data group to be displayed corresponding to the abnormal accident based on the values ​​of the data source under different monitoring dimensions in the data source layer and the total score layer, and generate a second visualization overview base map based on the data group to be displayed; Extracting a display item type of the to-be-displayed data group, and dividing the second visualization overview base map into a first area, a second area, and a third area based on the display item type; Generate an impact overview map of the abnormal accident based on the first area, generate a detailed chart of the impact overview map based on the second area, and generate a detailed report of the impact overview map based on the third area; Based on the second preset RGB color number, corresponding color numbers are set for the impact overview map, the detailed chart of the impact overview map, and the detailed report of the impact overview map, respectively, and based on the setting result, they are merged with the second visualization overview base map to obtain the abnormal accident impact overview map; The monitoring results are visualized based on the abnormal accident phenomenon overview diagram and the abnormal accident impact overview diagram.

7. The abnormal accident monitoring method of a third-party data platform according to claim 1 is characterized in that: In step 3, the screening index is obtained, and the data source after visualization is screened and located based on the screening index to determine the abnormal data accident, including: Obtaining a visualization result after visualizing the monitoring result, and at the same time, obtaining a screening index for the monitoring result, and generating a target filter based on the screening index, wherein the screening index is at least one; Determine the data source number and monitoring time length to be analyzed based on the filter, and determine the change trend of the data source sub-score and the total score corresponding to the data source number within the monitoring time length; Determining an abnormal image region corresponding to an abnormal change trend in the change trend, and marking the abnormal image region based on a preset marking symbol; Determine the abnormal accident phenomenon by determining the abnormal image area based on the marking result; Extracting the target data source number of the abnormal data source corresponding to the abnormal accident phenomenon, and analyzing the impact of the abnormal accident through the target data source number based on the target filter to obtain the product code and product version of the abnormal product corresponding to the abnormal accident; Determine the customer organization that retrieved the abnormal product within the monitoring time length based on the product code and product version, and determine the organization name of the customer organization; Synchronously updating and displaying the product code, product version, client organization and client organization name in the visualization processing result, and highlighting the data source sub-score and total score of the corresponding data source in the visualization processing result based on the display result; An abnormal accident monitoring report is generated based on the displayed result, and the monitoring report is transmitted to the management terminal.

8. The abnormal accident monitoring method of a third-party data platform according to claim 7 is characterized in that: Transmitting the monitoring report to the management terminal includes: Obtaining a monitoring report, and extracting first data and second data corresponding to abnormal accident phenomena and abnormal accident impacts in the monitoring report; Extracting data attributes of the first data and the second data, and determining data types of the first data and the second data based on the data attributes; Based on the data type, a historical data set of the same type is matched from a preset database, and the historical data set is trained to obtain an accident law prediction model; Inputting the first data and the second data into the accident law prediction model for analysis to obtain the occurrence law of the abnormal accident, and setting an accident self-inspection strategy based on the occurrence law; At the same time, the self-checking time interval of the accident self-checking strategy is determined based on the occurrence law of the abnormal accident, and the data source is self-checked according to the self-checking time interval based on the accident self-checking strategy.

9. The abnormal accident monitoring method of a third-party data platform according to claim 7 is characterized in that: Transmitting the monitoring report to the management terminal includes: Obtaining the obtained monitoring report, and extracting abnormal parameters corresponding to the abnormal accident described in the monitoring report; Comparing the abnormal parameter with a preset threshold, and determining the abnormal level of the abnormal accident based on the comparison result; Determining an alarm level for the abnormal accident based on the abnormal level, and determining a target alarm mode corresponding to the alarm level; Send an alarm reminder to the management terminal based on the target alarm mode.

10. An abnormal accident monitoring system for a third-party data platform, characterized in that: include: The data source processing module is used to collect the data source sub-scores and total scores of the data source from the third-party data platform, and pre-process the data source based on the data source sub-scores and total scores; A visualization display module, used to determine the monitoring dimension of the data source based on the preprocessing result, monitor the data source based on the monitoring dimension, and visualize the monitoring result; The positioning module is used to obtain screening indicators and screen and locate the data source after visualization based on the screening indicators to determine abnormal data accidents; Among them, the positioning module includes: Based on the screening indicators, the visualized data sources are screened and located to identify abnormal data incidents, including: Obtaining a preset standard threshold interval, and comparing the data source with the corresponding preset standard threshold interval respectively, and obtaining an abnormal data source whose value of the data source is not within the preset standard threshold interval based on the comparison result; Determine the data type of the abnormal data source, and verify the abnormal data source by matching the target data verification strategy from a preset data verification library based on the data type, to obtain the total amount of data of the abnormal data source and the number of incorrectly screened data sources; The accuracy of data source screening and positioning is calculated based on the total amount of data of the abnormal data source and the number of incorrectly screened data sources, and a comprehensive evaluation value of the monitoring effect of the data source is calculated based on the accuracy. The specific steps include: The accuracy of data source screening and positioning is calculated according to the following formula: ； in, Indicates the accuracy of data source screening and positioning, and the value range is (0, 1); Represents the error factor, and its value range is (0.02, 0.05); Indicates the total amount of data from the abnormal data source; Indicates the number of incorrectly filtered data sources in the abnormal data source, and the value is less than ; Indicates the number of accurate data sources obtained after filtering the data sources; Indicates the sum of the data volumes of all data sources; Obtaining the screening speed of each type of data source and the total amount of data of each type of data source, and performing dimensionless processing on the screening speed and the total amount of data of each type of data source to obtain a screening speed parameter and a total amount of data parameter of each type of data source; The comprehensive evaluation value of the monitoring effect on the data source is calculated according to the following formula: ； in, Comprehensive evaluation value of monitoring effect on data sources; Indicates the category of the current data source, and its value range is [1, ]; Indicates the total number of categories in the data source; Indicates the weight value corresponding to the accuracy of data source screening; Indicates the accuracy of data source screening and positioning, and the value range is (0, 1); Indicates the weight value of the data source filtering speed parameter; Indicates The filtering speed parameter of the class data source filtering, dimensionless; Indicates the weight value of the total amount of data in the data source; Indicates The total amount of data in the class data source, dimensionless; Compare the calculated comprehensive evaluation value of monitoring effect with the preset evaluation threshold; If the comprehensive evaluation value of the monitoring effect is greater than or equal to the preset evaluation threshold, it is determined that the monitoring effect on the data source is qualified; Otherwise, it is determined that the monitoring effect on the data source is unqualified, and the data source is screened and relocated until the comprehensive evaluation value of the monitoring effect is greater than or equal to the preset evaluation threshold.

Images