Screening method and device of target data and computer equipment

Abstract

The invention relates to a target data screening method and device, computer equipment, a storage medium and a computer program product. The method comprises the steps of obtaining to-be-screened data and historical index data; determining index data of each dimension of the to-be-screened data by using a preset index model; wherein the index data at least comprises one of model scoring data, type identification and strategy identification; processing the index data of each dimension and the historical index data by utilizing a preset safety coefficient generation strategy to obtain a safety coefficient of the to-be-screened data; wherein the safety coefficient at least comprises one of a group stability index and a change parameter; and comparing the safety coefficient with a preset safety threshold, and screening the to-be-screened data to obtain target data. By adopting the method, the data can be automatically screened, and the data screening efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of big data technology, and in particular to a method, apparatus, computer equipment, storage medium and computer program product for screening target data. Background Art

[0002] With the continuous development of the field of artificial intelligence, artificial intelligence models are being used more and more widely in corporate decision-making. Among them, the corporate scoring model, as a predictive model, plays an important role in all aspects of the enterprise. However, the monitoring means of the corporate scoring model for abnormal situations are not perfect, it is difficult to accurately identify potential problems, and it is impossible to intuitively monitor the overall stability. In the prior art, manual methods are usually used to monitor the scoring results of the corporate scoring model in real time and identify potential problems. However, this method is not only inefficient, but also has a low accuracy rate. Summary of the invention

[0003] Based on this, it is necessary to provide a method, device, computer equipment, computer-readable storage medium and computer program product for screening target data in response to the above technical problems.

[0004] In a first aspect, the present application provides a method for screening target data. The method comprises:

[0005] Obtain the data to be screened and historical indicator data;

[0006] Determine the indicator data of each dimension of the data to be screened by using a preset indicator model; wherein the indicator data includes at least one of the following: model scoring data, category identification, and strategy identification;

[0007] Using a preset safety factor generation strategy, the indicator data of each dimension and the historical indicator data are processed to obtain a safety factor of the data to be screened; wherein the safety factor includes at least one of the following: a group stability index and a change parameter;

[0008] The safety factor is compared with a preset safety threshold, and target data is obtained from the data to be screened.

[0009] In one embodiment, the method for obtaining the safety threshold further includes:

[0010] Obtaining a reference identifier of the data to be screened;

[0011] Determining reference data according to the reference identifier;

[0012] The safety threshold is determined according to the indicator data of each dimension of the reference data.

[0013] In one embodiment, the method of using a preset safety factor generation strategy to process the indicator data of each dimension and the historical indicator data to obtain the safety factor of the data to be screened includes:

[0014] According to the indicator data, determine the proportion of the data to be screened that meets the preset indicator data threshold; and according to the historical indicator data, determine the proportion of the historical data to be screened that meets the preset indicator data threshold;

[0015] The proportion of the to-be-screened data that meets the preset indicator data threshold and the proportion of the historical to-be-screened data that meets the preset indicator data threshold are subjected to difference processing to obtain a first sub-stability index;

[0016] The proportion of the to-be-screened data that meets the preset indicator data threshold and the proportion of the historical to-be-screened data that meets the preset indicator data threshold are processed to obtain an initial second sub-stability index;

[0017] Performing natural logarithm processing on the initial second self-stability index to obtain a second sub-stability index;

[0018] The first sub-stability index and the second sub-stability index are multiplied to obtain the group stability index.

[0019] In one embodiment, the method of using a preset safety factor generation strategy to process the indicator data of each dimension and the historical indicator data to obtain the safety factor of the data to be screened includes:

[0020] Sorting the indicator data to obtain sorted indicator data;

[0021] According to the preset variation coefficient, the indicator data of the rank corresponding to the variation coefficient is obtained from the sorted indicator data and determined as the variation parameter.

[0022] In one embodiment, the step of comparing the safety factor with a preset safety threshold and filtering the target data from the data to be filtered includes:

[0023] Comparing the safety factor of each dimension with the corresponding safety threshold;

[0024] When the safety factors of each dimension meet the safety threshold, the data to be screened is determined as target data.

[0025] In one embodiment, comparing the safety factor of each dimension with the corresponding safety threshold includes:

[0026] Comparing the safety factor of each dimension with the corresponding safety threshold;

[0027] In the case where the safety factor does not meet the safety threshold, the data to be screened is determined as risk data, and a warning signal is generated.

[0028] In a second aspect, the present application also provides a device for screening target data. The device comprises:

[0029] Data acquisition module, used to obtain data to be screened and historical indicator data;

[0030] An indicator data determination module, used to determine the indicator data of each dimension of the data to be screened by using a preset indicator model; wherein the indicator data includes at least one of the following: model scoring data, category identification, and strategy identification;

[0031] A safety factor determination module is used to process the indicator data of each dimension and the historical indicator data using a preset safety factor generation strategy to obtain a safety factor of the data to be screened; wherein the safety factor includes at least one of the following: a group stability index and a change parameter;

[0032] The target data screening module is used to compare the safety factor with a preset safety threshold and screen the target data from the data to be screened.

[0033] In one embodiment, the target data screening module further includes:

[0034] A reference identifier acquisition submodule, used to acquire the reference identifier of the data to be screened;

[0035] A reference data acquisition submodule, used to determine the reference data according to the reference identifier;

[0036] The safety threshold determination submodule is used to determine the safety threshold according to the indicator data of each dimension of the reference data.

[0037] In one embodiment, the safety factor determination module includes:

[0038] A proportion acquisition submodule is used to determine the proportion of the data to be screened that meets the preset indicator data threshold according to the indicator data; and to determine the proportion of the historical data to be screened that meets the preset indicator data threshold according to the historical indicator data;

[0039] An indicator determination submodule, used for performing a difference processing on the proportion of the to-be-screened data that meets a preset indicator data threshold and the proportion of the historical to-be-screened data that meets the preset indicator data threshold, to obtain a first sub-stability indicator;

[0040] The indicator determination submodule is further used to perform quotient processing on the proportion of the to-be-screened data that meets the preset indicator data threshold and the proportion of the historical to-be-screened data that meets the preset indicator data threshold, so as to obtain an initial second sub-stability indicator;

[0041] The index determination submodule is further used to perform natural logarithm processing on the initial second self-stability index to obtain a second sub-stability index;

[0042] The indicator determination submodule is further used to perform product processing on the first sub-stability indicator and the second sub-stability indicator to obtain the group stability indicator.

[0043] In one embodiment, the safety factor determination module includes:

[0044] The indicator data sorting submodule is used to sort the indicator data to obtain sorted indicator data;

[0045] The variation parameter determination submodule is used to obtain the indicator data of the rank corresponding to the variation coefficient from the sorted indicator data according to a preset variation coefficient, and determine it as the variation parameter.

[0046] In one embodiment, the target data screening module is further used to:

[0047] Comparing the safety factor of each dimension with the corresponding safety threshold;

[0048] When the safety factors of each dimension meet the safety threshold, the data to be screened is determined as target data.

[0049] In one embodiment, the target data screening module is further used to:

[0050] Comparing the safety factor of each dimension with the corresponding safety threshold;

[0051] In the case where the safety factor does not meet the safety threshold, the data to be screened is determined as risk data, and a warning signal is generated.

[0052] In a third aspect, the present application further provides a computer device, wherein the computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method for processing resource application as described in any one of the embodiments of the present disclosure is implemented.

[0053] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for processing resource application as described in any one of the embodiments of the present disclosure is implemented.

[0054] In a fifth aspect, the present application further provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the method for processing resource application as described in any one of the embodiments of the present disclosure is implemented.

[0055] The above-mentioned method, device, computer equipment, storage medium and computer program product for screening target data use the safety factor generation strategy to process the historical indicator data and indicator data to obtain the safety factor of the data to be screened, and compare the safety factor with the safety threshold to screen the target data from the data to be screened. By determining the safety factor for the indicator data and the historical indicator data, potential indicator problems can be accurately identified and the risk control capability can be enhanced. By analyzing the indicator data and the historical indicator data, more accurate data analysis can be provided, thereby improving the accuracy of the safety factor. At the same time, by adjusting the safety threshold, the system can be adapted to different environments, thereby increasing the flexibility and adaptability of the method and further improving the accuracy of target data screening. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 is a schematic diagram of a process for screening target data in one embodiment;

[0057] FIG2 is a schematic diagram of a process for determining a safety threshold in one embodiment;

[0058] FIG3 is a schematic diagram of a process for determining a population stability index in one embodiment;

[0059] FIG4 is a schematic diagram of a process for determining a change parameter in one embodiment;

[0060] FIG5 is a schematic diagram of a process for determining target data in one embodiment;

[0061] FIG6 is a schematic diagram of a process for determining risk data in one embodiment;

[0062] FIG7 is a schematic diagram of a flow chart of a method for screening target data in one embodiment;

[0063] FIG8 is a structural block diagram of a device for screening target data in one embodiment;

[0064] FIG. 9 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0065] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0066] In one embodiment, Figure 1 As shown, a method for screening target data is provided. This embodiment takes the method applied to a terminal as an example. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0067] Step S100, obtaining data to be screened and historical indicator data.

[0068] In an exemplary embodiment, the data to be screened may include business data, etc., and the historical indicator data may include indicator data corresponding to the business data predicted by an artificial intelligence model, etc.

[0069] In an exemplary embodiment, the data to be screened may include various types of data, for example, the data to be screened may be that the sales volume of XX company is Y; the sales volume here is the data to be screened, that is, the historical sales volume data may not be Y; the corresponding historical indicator data is the corresponding historical sales volume; specifically, the historical indicator data matches the historical data to be screened, the indicator data matches the data to be screened, and the meaning represented by the historical data to be screened and the data to be screened is the same, for example, the data is sales volume, the data to be screened may be today's sales volume, and the historical data to be screened may be historical sales volume, etc.

[0070] Step S200, using a preset indicator model, determining indicator data of each dimension of the data to be screened; wherein the indicator data includes at least one of the following: model scoring data, category identification, and strategy identification.

[0071] In an exemplary embodiment, the dimensions may include model scoring data, binning of model features where the data is located, and data hitting strategies, etc.; wherein the model scoring data may include the score of the target data predicted by the model, etc.; the binning of the features where the data is located may include the classification of the data in the model features, etc.

[0072] Step S300, using a preset safety factor generation strategy, processes the indicator data of each dimension and the historical indicator data to obtain a safety factor of the data to be screened; wherein the safety factor includes at least one of the following: a group stability index and a change parameter.

[0073] In an exemplary embodiment, the group stability index may include a model scoring stability index, a model feature bin stability index, a single strategy stability index, etc.; specifically, the model scoring stability index can be used to compare the stability of the daily overall proportion of each scoring level; the model feature bin stability can be used to measure the stability of the daily proportion change of each feature bin; the single strategy stability can be used to measure the stability of the daily proportion change of a single strategy hit, etc.; wherein, the model scoring stability index, the model feature bin stability index, the single strategy stability index, etc. can all be obtained through PSI, etc.

[0074] In an exemplary embodiment, the historical indicator data may include yesterday's indicator data; the indicator data may include today's indicator data; the group stability index (PSI) may be obtained in the following manner: PSI = (today's proportion of a certain item - yesterday's proportion of a certain item) × In (today's proportion of a certain item / yesterday's proportion of a certain item); wherein, the larger the PSI value, the more obvious the change, i.e., the more unstable. For example, the following formula (1) may be used to obtain the stability of the model rating level:

[0075] (1)

[0076] Among them, the PSI represents the stability of the model scoring level, d represents the proportion of today's model scoring level, and b represents the proportion of yesterday's model scoring level.

[0077] In an exemplary embodiment, the change parameter may include sorting each data item by size, determining the final change parameter according to a preset change coefficient, etc. For example, there are three indicator data items, 5, 10, and 15, respectively; the change coefficient is 60%; then 10 (median, 50%) + (15-10) × (10% / 50%) = 11 is taken as the change parameter, etc.

[0078] In an exemplary embodiment, the safety factor may also include indicators such as model migration rate, model score stability, feature stability, single strategy monitoring, strategy gradient hit monitoring, and strategy concentration; wherein the model migration rate may include the number or proportion of specific changes in the score, specifically including the number or proportion of data that was X yesterday and Y today; for example, the data being searched may be X yesterday and Y today; it may also be not X yesterday and X today; it may also be X yesterday and not X today, etc.

[0079] Step S400, comparing the safety factor with a preset safety threshold, and filtering out target data from the data to be filtered.

[0080] In an exemplary embodiment, a safety threshold may be set for all safety factors, and when all safety factors satisfy the safety threshold, the data may be determined as target data, etc.; in another exemplary embodiment, the safety factors may be weightedly summed to obtain a target safety factor, and a safety threshold may be set, and when the target safety factor satisfies the safety threshold, the data may be determined as target data, etc.

[0081] In an exemplary embodiment, comparing the safety factor with a preset safety threshold, and filtering out the target data from the data to be filtered may include determining the data to be filtered as the target data when each of the safety factors satisfies the safety threshold; in another exemplary embodiment, comparing the safety factor with a preset safety threshold, and filtering out the target data from the data to be filtered may include using a preset weighting coefficient to perform a weighted summation on each of the safety factors to obtain a target safety factor, and determining the data to be filtered as the target data when the target safety factor satisfies the safety threshold, etc.

[0082] In an exemplary embodiment, the method of filtering the target data from the data to be filtered may also include filtering the risk indicator data from the indicator data, etc. Specifically, the artificial intelligence model may be scored by the proportion of the risk indicator data, wherein the artificial intelligence model is used to determine the model scoring data, etc. for the data to be filtered.

[0083] In an exemplary embodiment, the data to be screened may include data to be screened of different objects, and key objects may be set, and the data of key objects need to be detected in key areas; specifically, for example, data with large changes may be due to hitting certain strategies or changes in their own characteristics, which will lead to large changes in the scoring results. By verifying the changes in the characteristics and name strategies of these key enterprises, it can be verified whether the result changes are in line with expectations, etc.

[0084] In the above-mentioned method for screening target data, the safety factor generation strategy is used to process the historical indicator data and the indicator data to obtain the safety factor of the data to be screened, and the safety factor is compared with the safety threshold to screen the target data from the data to be screened. By determining the safety factor for the indicator data and the historical indicator data, potential indicator problems can be accurately identified and risk control capabilities can be enhanced. By analyzing the indicator data and the historical indicator data, more accurate data analysis can be provided, thereby improving the accuracy of the safety factor. At the same time, by adjusting the safety threshold, the system can adapt to different environments, thereby increasing the flexibility and adaptability of the method and further improving the accuracy of target data screening.

[0085] In one embodiment, as shown in FIG2 , the method for obtaining the safety threshold further includes:

[0086] Step S401: obtaining a reference identifier of the data to be screened.

[0087] In an exemplary embodiment, the reference identifier may include a portion of data that is pre-set as reference data, etc. For example, there are sales of five companies in the data, and we select two of them as reference data, that is, we use the data to be screened of these two reference companies to determine or optimize the safety threshold, etc.

[0088] In an exemplary embodiment, the data to be screened may include data to be screened of different data objects; the reference identifier may include reference identifiers of different data objects, and the reference identifier is used to indicate whether the data object is a reference object, etc.

[0089] Step S402: Determine reference data according to the reference identifier.

[0090] In an exemplary embodiment, the data to be screened may include the same data of different objects, etc.; the reference identifier may be an object identifier, that is, the object identifier is used to determine the data corresponding to the object identifier; and the data is used to determine or optimize the security threshold, etc. It is understandable that the individual data of a single object may be used separately, for example, the sales of Company A may be used as reference data, but the sales amount of Company A is not used as reference data, etc.

[0091] Step S403: determining the safety threshold according to the indicator data of each dimension of the reference data.

[0092] In an exemplary embodiment, the indicator data of each dimension of the reference data can be used to optimize the historical safety threshold to obtain the optimized safety threshold, etc.; in another exemplary embodiment, the indicator data of each dimension of the reference data can also be used to determine the safety threshold, etc.

[0093] In an exemplary embodiment, a preset time period can be used to obtain indicator data within the time period; safety thresholds, etc. can be determined based on the indicator data within the time period; specifically, after collecting indicator fluctuations within the time period and analyzing the causes of the fluctuations, the safety thresholds, etc. of each indicator can be ultimately determined.

[0094] In this embodiment, the reference data is determined based on the reference identification of the data to be screened, and the safety threshold is determined using the indicator data of the reference data. By calculating the safety threshold based on the reference data, data that does not meet the standards can be effectively identified, thereby improving data quality and the effectiveness of decision-making. This method not only enhances the transparency of data analysis, but also provides a solid foundation for subsequent risk management and compliance.

[0095] In one embodiment, as shown in FIG3 , the use of a preset safety factor generation strategy to process the indicator data of each dimension and the historical indicator data to obtain the safety factor of the data to be screened includes:

[0096] Step S301, determining the proportion of the data to be screened that meets the preset indicator data threshold according to the indicator data; and determining the proportion of the historical data to be screened that meets the preset indicator data threshold according to the historical indicator data.

[0097] Step S302, performing a difference processing on the proportion of the to-be-screened data that meets the preset indicator data threshold and the proportion of the historical to-be-screened data that meets the preset indicator data threshold, to obtain a first sub-stability indicator.

[0098] Step S303, taking the quotient of the proportion of the to-be-screened data that meets the preset indicator data threshold and the proportion of the historical to-be-screened data that meets the preset indicator data threshold, to obtain an initial second sub-stability indicator.

[0099] Step S304, performing natural logarithm processing on the initial second self-stability index to obtain a second sub-stability index.

[0100] Step S305, performing product processing on the first sub-stability index and the second sub-stability index to obtain the group stability index.

[0101] In an exemplary embodiment, the following formula (2) may be used to obtain the population stability index:

[0102] (2)

[0103] Among them, PSI represents the group stability index, d represents the proportion of the indicator data of the data to be screened to the indicator data of the data to be screened; b represents the proportion of the indicator data of the historically screened data to the historical indicator data; specifically, db is the first sub-stability index, d / b is the initial second sub-stability index, and In (d / b) is the second sub-stability index; specifically, the data to be screened that meets the preset indicator data threshold may include setting a target data threshold, screening indicator data greater than the target data threshold from the data indicators, etc.; for example, there are data with model scoring levels of 1, 2, and 3 in the data to be screened; if the preset indicator data threshold is set to 3, then the data to be screened that meets the preset indicator data threshold is the data with a model scoring level of 3, etc.

[0104] In this embodiment, by utilizing the proportion of data to be screened that meets the index data threshold, and the proportion of historical data to be screened, and using the proportion to respectively determine the first sub-stability index and the second sub-stability index, the group stability index is finally determined. It is possible to achieve a systematic way to evaluate the stability of data by quantifying the difference in the proportion of data to be screened and historical data. First, by calculating the proportion of data that meets the preset threshold, the compliance and quality of the current data can be effectively identified. Secondly, the comparative analysis of historical data makes the evaluation more comprehensive and can reveal the trend and potential risks of data changes. Finally, through mathematical operations such as difference, quotient, and natural logarithm processing, the group stability index obtained provides a scientific basis for decision-making, helping organizations to adjust strategies in a timely manner to cope with data fluctuations, thereby improving the efficiency and reliability of overall data management.

[0105] In one embodiment, as shown in FIG4 , the use of a preset safety factor generation strategy to process the indicator data of each dimension and the historical indicator data to obtain the safety factor of the data to be screened includes:

[0106] Step S311, sorting the indicator data to obtain sorted indicator data.

[0107] Step S312: According to a preset variation coefficient, obtain the indicator data of the rank corresponding to the variation coefficient from the sorted indicator data, and determine it as a variation parameter.

[0108] In an exemplary embodiment, the variation coefficient may include the rank in the indicator data, etc. The variation parameter may include the quantile value of the indicator data; for example, there are three indicator data, 5, 10, and 15 respectively; the variation coefficient is 60%; then 10 (median, 50%) + (15-10) × (10% / 50%) = 11 is taken as the variation parameter, etc.

[0109] In this embodiment, the variation parameter is determined by the variation coefficient, so that the variation parameter can be accurately obtained. The variation coefficient is used to provide a data basis for determining the variation parameter, thereby further improving the accuracy of the variation parameter.

[0110] In one embodiment, as shown in FIG5 , the step of comparing the safety factor with a preset safety threshold and filtering the target data from the data to be filtered includes:

[0111] Step S411, comparing the safety factor of each dimension with the corresponding safety threshold.

[0112] Step S412: when the safety factor of each dimension meets the safety threshold, the data to be screened is determined as target data.

[0113] In an exemplary embodiment, the comparison of the safety factor with the safety threshold may include that the safety factor of each dimension meets the safety threshold; it may also include using a preset weight ratio to perform a weighted summation of the safety factors of each dimension, and comparing the result with the safety threshold, etc.

[0114] In an exemplary embodiment, the target data is data that meets a safety threshold, and the target data can be used to determine corresponding indicator data, display the indicator data, and so on.

[0115] In this embodiment, by determining data whose safety factor meets the safety threshold as target data, the required target data can be accurately screened out; by comparing with the safety threshold, a data basis is provided for the screening of target data, further improving the accuracy of target data screening.

[0116] In one embodiment, as shown in FIG6 , comparing the safety factor of each dimension with the corresponding safety threshold includes:

[0117] Step S421, comparing the safety factor of each dimension with the corresponding safety threshold.

[0118] Step S422: When the safety factor does not meet the safety threshold, the data to be screened is determined as risk data, and a warning signal is generated.

[0119] In an exemplary embodiment, the safety factor not meeting the safety threshold may include that the safety factor of one dimension does not meet the safety threshold; it may also include that the weighted sum of the safety factors of each dimension does not meet the safety threshold, etc.

[0120] In an exemplary embodiment, after the warning signal is generated, risk detection can be performed on risk data in a preset manner; specifically, business personnel can be informed through a warning to specifically analyze indicator data that has not passed the threshold verification.

[0121] In this embodiment, by comparing the safety factor of each dimension with the safety threshold, potential risk data can be identified in a timely manner. This real-time monitoring mechanism ensures the security and compliance of the data, and can quickly generate early warning signals when the safety factor does not meet the threshold, thereby providing timely risk warnings to decision makers. In this way, organizations can proactively respond to risks, reduce potential losses, and optimize resource allocation. In addition, the early warning mechanism also enhances the transparency and responsiveness of data management, helping to improve overall operational efficiency and safety levels.

[0122] In an exemplary embodiment, the flowchart shown in FIG. 7 may be used for implementation; specifically, the process includes:

[0123] Step S901, obtaining the scoring model result; obtaining upstream data, using the artificial intelligence model to clean and predict the upstream data, and obtaining the scoring model result. Specifically, the cleaning prediction task can be performed on the big data platform on a daily basis based on the date, and the scoring model result is stored in the server of the big data platform;

[0124] Step S902, according to the scoring model results, determine the data of each indicator; according to today's scoring model results and yesterday's scoring model results, from the model's scoring results, the binning of the model features where the data is located and the hit strategy, calculate the model migration rate, model scoring stability, feature stability, single strategy monitoring, strategy gradient hit monitoring, strategy concentration and other indicators, and store them in the database;

[0125] Step S903, confirmation of group stability index and quantile value; analysis and aggregation of various indexes for display, and calculation of PSI and quantile values ​​of various dimensions to measure the stability of the monitoring model;

[0126] Step S904, key object detection; a scheduled script can be written to use the key objects as sample pools (i.e., reference objects) every day, and store the object data with large indicator changes in the database for summary, which can be used to derive and verify whether the changes of individual samples meet expectations;

[0127] Step S905, data push and analysis of indicators that have not passed the threshold; according to various monitoring indicators, collect, observe and analyze the fluctuation of indicators in the preset time period and analyze the reasons for the fluctuation, reasonably and scientifically set the threshold of each indicator, configure the monitoring threshold on the big data platform, and automatically compare the indicators with the threshold after the offline task automatically throws out the indicators every day. When all indicators meet the expected threshold, the data push task is automatically executed to update the object score of the day. If it does not meet the expected threshold, the push will not be executed, and the business personnel will be informed by telephone warning, and then the indicators that have not passed the threshold test will be specifically analyzed.

[0128] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0129] Based on the same inventive concept, the embodiment of the present application also provides a device for screening target data for implementing the method for screening target data involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the method above, so the specific limitations in the embodiments of one or more target data screening devices provided below can refer to the limitations of the method for screening target data above, and will not be repeated here.

[0130] In one embodiment, as shown in FIG8 , a target data screening device 100 is provided, comprising: a data acquisition module 101, an indicator data determination module 102, a safety factor determination module 103 and a target data screening module 104, wherein:

[0131] Data acquisition module, used to obtain data to be screened and historical indicator data;

[0132] An indicator data determination module, used to determine the indicator data of each dimension of the data to be screened by using a preset indicator model; wherein the indicator data includes at least one of the following: model scoring data, category identification, and strategy identification;

[0133] A safety factor determination module is used to process the indicator data of each dimension and the historical indicator data using a preset safety factor generation strategy to obtain a safety factor of the data to be screened; wherein the safety factor includes at least one of the following: a group stability index and a change parameter;

[0134] The target data screening module is used to compare the safety factor with a preset safety threshold and screen the target data from the data to be screened.

[0135] In one embodiment, the target data screening module further includes:

[0136] A reference identifier acquisition submodule, used to acquire the reference identifier of the data to be screened;

[0137] A reference data acquisition submodule, used to determine the reference data according to the reference identifier;

[0138] The safety threshold determination submodule is used to determine the safety threshold according to the indicator data of each dimension of the reference data.

[0139] In one embodiment, the safety factor determination module includes:

[0140] A proportion acquisition submodule is used to determine the proportion of the data to be screened that meets the preset indicator data threshold according to the indicator data; and to determine the proportion of the historical data to be screened that meets the preset indicator data threshold according to the historical indicator data;

[0141] An indicator determination submodule, used for performing a difference processing on the proportion of the to-be-screened data that meets a preset indicator data threshold and the proportion of the historical to-be-screened data that meets the preset indicator data threshold, to obtain a first sub-stability indicator;

[0142] The indicator determination submodule is further used to perform quotient processing on the proportion of the to-be-screened data that meets the preset indicator data threshold and the proportion of the historical to-be-screened data that meets the preset indicator data threshold, so as to obtain an initial second sub-stability indicator;

[0143] The index determination submodule is further used to perform natural logarithm processing on the initial second self-stability index to obtain a second sub-stability index;

[0144] The indicator determination submodule is further used to perform product processing on the first sub-stability indicator and the second sub-stability indicator to obtain the group stability indicator.

[0145] In one embodiment, the safety factor determination module includes:

[0146] The indicator data sorting submodule is used to sort the indicator data to obtain sorted indicator data;

[0147] The variation parameter determination submodule is used to obtain the indicator data of the rank corresponding to the variation coefficient from the sorted indicator data according to a preset variation coefficient, and determine it as the variation parameter.

[0148] In one embodiment, the target data screening module is further used to:

[0149] Comparing the safety factor of each dimension with the corresponding safety threshold;

[0150] When the safety factors of each dimension meet the safety threshold, the data to be screened is determined as target data.

[0151] In one embodiment, the target data screening module is further used to:

[0152] Comparing the safety factor of each dimension with the corresponding safety threshold;

[0153] In the case where the safety factor does not meet the safety threshold, the data to be screened is determined as risk data, and a warning signal is generated.

[0154] Each module in the above-mentioned target data screening device can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each module.

[0155] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be shown in FIG9. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, the memory, and the input / output interface are connected via a system bus, and the communication interface, the display unit, and the input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a method for screening target data is implemented. The display unit of the computer device is used to form a visually visible picture, which may be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0156] Those skilled in the art will understand that the structure shown in FIG. 9 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0157] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0158] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0159] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0160] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for screening target data, characterized in that: The method comprises: Obtain the data to be screened and historical indicator data; Determine the indicator data of each dimension of the data to be screened by using a preset indicator model; wherein the indicator data includes at least one of the following: model scoring data, category identification, and strategy identification; Using a preset safety factor generation strategy, the indicator data of each dimension and the historical indicator data are processed to obtain a safety factor of the data to be screened; wherein the safety factor includes at least one of the following: a group stability index and a change parameter; The safety factor is compared with a preset safety threshold, and target data is obtained from the data to be screened.

2. The method according to claim 1, characterized in that The method for obtaining the safety threshold further includes: Obtaining a reference identifier of the data to be screened; Determining reference data according to the reference identifier; The safety threshold is determined according to the indicator data of each dimension of the reference data.

3. The method according to claim 1, characterized in that The method of using a preset safety factor generation strategy to process the indicator data of each dimension and the historical indicator data to obtain the safety factor of the data to be screened includes: According to the indicator data, determine the proportion of the data to be screened that meets the preset indicator data threshold; and according to the historical indicator data, determine the proportion of the historical data to be screened that meets the preset indicator data threshold; The proportion of the to-be-screened data that meets the preset indicator data threshold and the proportion of the historical to-be-screened data that meets the preset indicator data threshold are subjected to difference processing to obtain a first sub-stability index; The proportion of the to-be-screened data that meets the preset indicator data threshold and the proportion of the historical to-be-screened data that meets the preset indicator data threshold are processed to obtain an initial second sub-stability index; Performing natural logarithm processing on the initial second self-stability index to obtain a second sub-stability index; The first sub-stability index and the second sub-stability index are multiplied to obtain the group stability index.

4. The method according to claim 1, characterized in that The method of using a preset safety factor generation strategy to process the indicator data of each dimension and the historical indicator data to obtain the safety factor of the data to be screened includes: Sorting the indicator data to obtain sorted indicator data; According to the preset variation coefficient, the indicator data of the rank corresponding to the variation coefficient is obtained from the sorted indicator data and determined as the variation parameter.

5. The method according to claim 1, characterized in that The step of comparing the safety factor with a preset safety threshold and filtering the target data from the data to be filtered includes: Comparing the safety factor of each dimension with the corresponding safety threshold; When the safety factors of each dimension meet the safety threshold, the data to be screened is determined as target data.

6. The method according to claim 5, characterized in that The safety factor of each dimension is compared with the corresponding safety threshold, including: Comparing the safety factor of each dimension with the corresponding safety threshold; In the case where the safety factor does not meet the safety threshold, the data to be screened is determined as risk data, and a warning signal is generated.

7. A device for screening target data, characterized in that: The device comprises: Data acquisition module, used to obtain data to be screened and historical indicator data; An indicator data determination module, used to determine the indicator data of each dimension of the data to be screened by using a preset indicator model; wherein the indicator data includes at least one of the following: model scoring data, category identification, and strategy identification; A safety factor determination module is used to process the indicator data of each dimension and the historical indicator data using a preset safety factor generation strategy to obtain a safety factor of the data to be screened; wherein the safety factor includes at least one of the following: a group stability index and a change parameter; The target data screening module is used to compare the safety factor with a preset safety threshold and screen the target data from the data to be screened.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images