A change detection method, device and equipment for risk control data

Abstract

This specification discloses a method, apparatus, and device for detecting changes in risk control data. The solution may include: calculating the estimated information value of each risk control data point by utilizing operational record information carrying various risk control data provided by a data provider and risk decision results generated from risk control strategies; and generating detection results reflecting whether the risk control data provided by the data provider has been added or modified based on a comparison between the estimated information value and historical estimated information values.

Description

Technical Field

[0001] This application relates to the field of risk control technology, and in particular to a method, apparatus and equipment for detecting changes in risk control data. Background Technology

[0002] To protect the interests of users and businesses, service providers typically develop risk control strategies based on their accumulated industry experience to minimize losses in the event of risk incidents. Currently, service providers often rely on risk control data provided by data providers to train and run these strategies for business risk control. However, it is generally necessary for data providers to proactively inform service providers when the risk control data changes. This allows service providers to adjust their risk control strategies accordingly, ensuring the effectiveness of the strategies.

[0003] Therefore, how to detect and identify changes in risk control data in a timely and accurate manner has become an urgent technical problem to be solved. Summary of the Invention

[0004] The embodiments of this specification provide a method, apparatus, and equipment for detecting changes in risk control data, which can detect and identify changes in risk control data in a timely and accurate manner.

[0005] To solve the above-mentioned technical problems, the embodiments in this specification are implemented as follows:

[0006] This specification provides an embodiment of a method for detecting changes in risk control data, including:

[0007] Obtain the operation record information of the risk control strategy; the operation record information carries various risk control data provided by the data provider, and risk decision results generated by the risk control strategy based on the risk control data of preset types;

[0008] Based on the operation record information, calculate the estimated information value of each of the risk control data at the risk control strategy;

[0009] The historical forecast information value of the risk control data;

[0010] Based on the estimated information value and the historical estimated information value, a change detection result is generated for the risk control data provided by the data provider.

[0011] This specification provides an embodiment of a change detection device for risk control data, comprising:

[0012] The first acquisition module is used to acquire the operation record information of the risk control strategy; the operation record information carries various risk control data provided by the data provider, and risk decision results generated by the risk control strategy based on the risk control data of preset types;

[0013] The estimated information value calculation module is used to calculate the estimated information value of each of the risk control data at the risk control strategy based on the operation record information.

[0014] The second acquisition module is used to acquire the historical estimated information value of the risk control data;

[0015] The detection result generation module is used to generate detection results for changes in the risk control data provided by the data provider based on the estimated information value and the historical estimated information value.

[0016] This specification provides an embodiment of a change detection device for risk control data, comprising:

[0017] At least one processor; and,

[0018] A memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to:

[0020] Obtain the operation record information of the risk control strategy; the operation record information carries various risk control data provided by the data provider, and risk decision results generated by the risk control strategy based on the risk control data of preset types;

[0021] Based on the operation record information, calculate the estimated information value of each of the risk control data at the risk control strategy;

[0022] The historical forecast information value of the risk control data;

[0023] Based on the estimated information value and the historical estimated information value, a change detection result is generated for the risk control data provided by the data provider.

[0024] At least one embodiment provided in this specification can achieve the following beneficial effects:

[0025] By utilizing operational records containing various risk control data and risk decision-making results generated by risk control strategies provided by the data provider, the estimated information value of each risk control data point can be calculated. Based on the comparison between the estimated information value of the risk control data and the historical estimated information value, a relatively accurate detection result can be generated to reflect whether there are any additions or modifications to the risk control data provided by the data provider. Furthermore, since there is no need to wait for the data provider to proactively inform about changes in the risk control data, timely and accurate detection and identification of changes in risk control data can be performed. This allows for timely adjustments to risk control strategies based on changes in the risk control data, improving the effectiveness and reliability of risk control strategies and ensuring the safe operation of business. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 This is a schematic diagram illustrating an application scenario of a method for detecting changes in risk control data, provided in an embodiment of this specification.

[0028] Figure 2 A flowchart illustrating a method for detecting changes in risk control data, provided as an embodiment of this specification;

[0029] Figure 3 The embodiments provided in this specification correspond to Figure 2 A swimlane flowchart illustrating the method for detecting changes in risk control data.

[0030] Figure 4 The embodiments provided in this specification correspond to Figure 2 A schematic diagram of a change detection device for risk control data;

[0031] Figure 5 The embodiments provided in this specification correspond to Figure 2 A schematic diagram of a change detection device for risk control data. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of one or more embodiments of this specification clearer, the technical solutions of one or more embodiments of this specification will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of them. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of one or more embodiments of this specification.

[0033] The technical solutions provided in the various embodiments of this specification are described in detail below with reference to the accompanying drawings.

[0034] In existing technologies, with the increasing prevalence of internet business models, data interaction between various business participants is becoming more frequent. Since service providers have a limited amount of risk control data, they need to call pre-defined interfaces provided by data providers to obtain the risk control data, and then use this data to train and run risk control strategies to ensure the secure operation of the business.

[0035] Currently, when the risk control data provided by the data provider changes, the data provider needs to modify the preset interface. For example, adding field names and value types to the new preset interface to increase the types of risk control data sent to the service provider. In addition, the data provider needs to proactively inform the service provider of the modifications to the preset interface so that the service provider's staff can modify the service provider's parsing rules for the preset interface based on the received changes, adding parsing logic for the newly added risk control data. This process is quite inconvenient.

[0036] Furthermore, because service providers rely on notifications from data providers to be aware of changes in risk control data and update their risk control strategies, they cannot keep track of these changes in real time. This prevents service providers from updating their risk control strategies promptly based on the changes in the risk control data, which also affects the security of business operations.

[0037] Therefore, how to detect and identify changes in risk control data in a timely and accurate manner has become an urgent technical problem to be solved.

[0038] To address the shortcomings of existing technologies, this solution provides the following embodiments:

[0039] Figure 1 This is a schematic diagram illustrating an application scenario of a method for detecting changes in risk control data, provided in an embodiment of this specification.

[0040] like Figure 1 As shown, the risk control device 101 at the service provider can obtain various risk control data provided by the data provider from the device 102 of the data provider, and generate risk decision results based on the obtained preset types of risk control data using risk control strategies. In this way, it generates operation record information containing various risk control data provided by the data provider and the risk decision results for the risk control strategy.

[0041] The risk control device 101 can also calculate the estimated information value of each of the risk control data at the risk control strategy based on the operation record information; by comparing the estimated information value of the risk control data with the historical estimated information value, a change detection result for the risk control data provided by the data provider can be generated.

[0042] Figure 1 In this solution, the service provider's risk control equipment can automatically generate detection results for changes in the risk control data provided by the data provider based on the risk control data obtained in real time from the data provider, without relying on notifications from the data provider to be aware of changes in the risk control data. This allows the service provider to update its risk control strategies in a timely manner based on changes in the risk control data, thereby ensuring the security of business operations.

[0043] Next, a method for detecting changes in risk control data, as provided in the embodiments of the specification, will be described in detail with reference to the accompanying drawings:

[0044] Figure 2 This is a flowchart illustrating a method for detecting changes to risk control data, provided in an embodiment of this specification. From a programming perspective, the entity executing this process can be the risk control device 101, or an application program installed on the risk control device 101. Figure 2 As shown, the process may include the following steps:

[0045] Step 202: Obtain the operation record information of the risk control strategy; the operation record information carries various risk control data provided by the data provider, and risk decision results generated by the risk control strategy based on the preset types of risk control data.

[0046] In this embodiment of the specification, a risk control engine may be deployed at the risk control device 101. During operation, the risk control engine can obtain various risk control data provided by the data provider and run a pre-set risk control strategy to generate a risk decision result based on the preset types of risk control data in the obtained risk control data. In addition, the risk control engine can also generate a risk control strategy operation record information based on the various risk control data provided by the data provider and the generated risk decision result.

[0047] In practical applications, the preset risk control data can be all or part of the risk control data provided by the data provider, without specific limitations. This is because the risk control data provided by the data provider may include newly added risk control data. In this case, since the risk control strategies at the risk control engine have not been updated based on the training samples containing the newly added risk control data, these risk control strategies typically do not use the newly added risk control data to generate risk control strategies. Therefore, the preset risk control data required by the risk control strategy can be a portion of the data provided by the data provider. Alternatively, the data provider can send the risk control data required by multiple risk control strategies together to the risk control device 101. Since the risk control data required by each risk control strategy may be inconsistent, the preset risk control data required by a single risk control strategy can be a portion of the data provided by the data provider.

[0048] However, since it is necessary to detect changes in all risk control data provided by the data provider, all risk control data provided by the data provider can be included in the operation log information of the risk control strategy. Furthermore, since this embodiment also needs to combine the risk decision results generated by the risk control strategy to evaluate the estimated information value of each risk control data point within that risk control strategy, the risk decision results can also be included in the operation log information of the risk control strategy.

[0049] In addition, the operation record information of the risk control strategy may also carry the unique strategy identifier information of the risk control strategy to represent the risk control strategy that generated the operation record information; and may also carry the record ID, the generation time information of the operation record information, etc., without specific limitations.

[0050] Risk control data can be data that can be used for risk identification, such as basic user data, basic business data, data generated by users during business transactions, or score data that is obtained by processing the above data to characterize user behavior or features, etc., without specific limitations.

[0051] Risk control strategies can be sets of risk control rules and models, formed by arranging and combining risk control rules and models based on target customer groups and business scenarios. The forms of risk control rules and models can vary and are not specifically limited.

[0052] Step 204: Based on the operation record information, calculate the estimated information value of each of the risk control data at the risk control strategy.

[0053] In the embodiments of this specification, the estimated information value (IV) of risk control data at the risk control strategy level can be used to measure the predictive ability of the risk control data variable for the risk decision results generated by the risk control strategy. Therefore, the estimated information value of each risk control data at the risk control strategy level can be calculated based on the various risk control data and risk decision results contained in multiple operation record information.

[0054] Step 206: Obtain the historical estimated value of the risk control data.

[0055] In this embodiment of the specification, in order to accurately identify changes in various risk control data, it is also necessary to obtain the historical estimated information value generated for each type of risk control data during the historical change detection process. That is, the historical estimated information value can be the IV value that was previously generated, and the calculation principle of the historical estimated information value can be consistent with the estimated information value mentioned in step 204 to ensure comparability between the two.

[0056] In the embodiments of this specification, since there is usually no historical forecast information value for the newly added type of risk control data, it is possible to obtain the historical forecast information value of all the risk control data provided by the data provider, or it is possible to obtain only the historical forecast information value of some of the risk control data provided by the data provider. No specific limitation is made in this regard.

[0057] Step 208: Based on the estimated information value and the historical estimated information value, generate a change detection result for the risk control data provided by the data provider.

[0058] In this embodiment of the specification, since the estimated information value of risk control data may differ significantly from its historical estimated information value after the data provider modifies or adds risk control data, a comparison can be made between the estimated information value and the historical estimated information value to generate a change detection result for each risk control data provided by the data provider. In practical applications, this change detection result can be used to reflect whether the risk control data provided by the data provider has been added, modified, or remains unchanged.

[0059] Figure 2 The method described herein utilizes operational records containing various risk control data and risk decision-making results generated by risk control strategies provided by the data provider. This allows for the calculation of the estimated information value of each risk control data point. By comparing the estimated information value with historical estimated information values, a relatively accurate detection result can be generated to reflect whether the risk control data provided by the data provider has been added or modified. Furthermore, since it eliminates the need to wait for the data provider to proactively inform the user of any changes to the risk control data, timely and accurate detection and identification of these changes are possible. This allows for timely adjustments to risk control strategies based on these changes, improving the effectiveness and reliability of risk control strategies and ensuring the safe operation of business operations.

[0060] based on Figure 2 In addition to the method described in the embodiments of this specification, some specific implementation schemes of the method are also provided, which will be described below.

[0061] To facilitate understanding, the method for generating operational record information of risk control strategies will be explained.

[0062] Specifically, before step 202: obtaining the operational record information of the risk control strategy, it may also include:

[0063] Receive risk control data in key-value pair format to be parsed from the data provider.

[0064] The risk control data to be parsed is parsed to obtain various types of risk control data provided by the data provider.

[0065] From the various types of risk control data, determine the preset types of risk control data required for the risk control strategy.

[0066] The risk control strategy is used to process the risk control data of the preset types to obtain the risk decision results.

[0067] Based on the risk decision results and the various types of risk control data, the operation record information of the risk control strategy is generated.

[0068] In the embodiments of this specification, the key-value pair format is widely used because the key can be an index and the value can be data. Key-value pair format data is concise and easy to parse. In practical applications, JSON (JavaScript Object Notation) is a lightweight data exchange format. JSON objects can consist of key-value pairs enclosed in curly braces and separated by commas. Therefore, data providers can use JSON technology to send risk control data in key-value pair format to service providers.

[0069] Based on this, the risk control data to be parsed, in key-value pair format, sent by the data provider, may specifically include:

[0070] Receive a JSON string containing risk control data sent by the data provider.

[0071] Correspondingly, the risk control data to be parsed is parsed to obtain various types of risk control data provided by the data provider, specifically including:

[0072] The JSON string is parsed to obtain risk control data in key-value pair format; wherein the key data of the risk control data is used to reflect the type of the risk control data, and the value data of the risk control data is used to reflect the numerical value of the risk control data.

[0073] In the embodiments of this specification, each key-value pair formatted risk control data may include a key and a value. The key can be a field name set by the data provider for the risk control data, such as score1, score2…scoreN, to represent the type of risk control data. The value can be a field value, such as 10, 2.5…15, to represent the numerical value of the risk control data. In practical applications, the naming convention of the key and the value range of the value can be set according to actual needs and are not specifically limited. As an example, a JSON string carrying risk control data could be {"score1":"10","score2":"2.5"…"scoreN":"15"}.

[0074] In practical applications, since the parsing rules for JSON strings remain consistent regardless of the number of key-value pairs containing risk control data, when a data provider adds or modifies the risk control data it provides to the service provider, the service provider does not need to modify the parsing rules of the data provider's pre-defined interface. This ensures the service provider has the ability to retrieve the changed risk control data quickly and easily. Furthermore, when a data provider needs to add new risk control data, it only needs to add a key-value pair to the JSON string, making the operation quite convenient.

[0075] In the embodiments of this specification, a pre-established correspondence between risk control strategies and the preset categories of risk control data required for their use can be established. Therefore, when running a risk control strategy, the preset categories of risk control data provided by the data provider can be extracted as input parameters based on this correspondence. In practical applications, since the key data contained in risk control data can reflect the category to which the risk control data belongs, a correspondence between risk control strategies and the key data of the required risk control data can be established. For example, assuming a pre-established correspondence between risk control strategies and "score1" and "score2", when the JSON string carrying risk control data is {"score1":"10","score2":"2.5"..."scoreN":"15"}, the value data "10" and "2.5" can be used as input parameters for the risk control strategy to obtain the risk decision result output by the risk control strategy. The risk decision result can be a binary result, such as "yes" or "no", or "1" can be used to represent "yes" and "0" to represent "no"; of course, the risk decision result can also be a multi-class result or other forms of result, without specific limitations.

[0076] In the embodiments described in this specification, the risk control device may not only deploy a risk control engine, but also a system for detecting changes to risk control data. Based on this, the risk control engine can also send its generated operational log information for risk control strategies to this system for execution. Figure 2 The method described above. In practical applications, the operational log information for risk control strategies can also be a JSON string for easy parsing. Of course, it can also be other formats such as tables, text, and messages; there are no specific limitations on this.

[0077] In the embodiments of this specification, for ease of understanding, the method of generating the estimated information value of risk control data is explained here.

[0078] Specifically, step 204: Based on the operation record information, calculate the estimated information value of each of the risk control data at the risk control strategy, which may include:

[0079] Based on the risk control data of the target type contained in the operation record information, determine the numerical grouping corresponding to the risk control data of the target type.

[0080] Based on the risk decision results contained in the operation record information and the risk control data of the target type, calculate the evidence weight of the risk control data of the target type under each of the numerical groups.

[0081] Based on the evidence weights under each of the numerical groupings, the estimated information value of the risk control data for the target category at the risk control strategy is generated.

[0082] In the embodiments of this specification, the risk control data of the target type can be any type of risk control data provided by the data provider. Since any type of risk control data can typically have the same key data and multiple value data, the value data of the risk control data can be grouped to obtain multiple numerical groups.

[0083] For example, the risk control data for a target category can be a discrete variable. Suppose the data only contains the values ​​"10" and "20". Then, the number of numerical groups corresponding to the risk control data for that target category can be two, with one group containing the value "10" and the other containing the value "20". Alternatively, the risk control data for a target category can also be a continuous variable, for example, a value between 0 and 30. In this case, the number of numerical groups corresponding to the risk control data for that target category can be set to three. After equal-interval grouping, we can obtain one numerical group corresponding to the value data "0-10", one corresponding to the value data "11-20", and one corresponding to the value data "21-30".

[0084] In the embodiments described in this specification, the weight of evidence (WOE) can be used to measure the correlation between the independent variable and the dependent variable.

[0085] Specifically, the evidence weight woe under the i-th numerical group i It can be calculated using the following formula:

[0086]

[0087] Among them, badi The number of execution record information entries that have a certain type of risk decision outcome (e.g., a risk decision outcome identified as bad) under the i-th numerical group, where bad T This can be the total number of operational log entries with a certain type of risk decision outcome (e.g., a risk decision outcome identified as bad), good i The number of run record information items that have other types of risk decision results (e.g., risk decision results identified as "good") under the i-th numerical group, where "good" represents the number of run record items. T This can be the total number of operational record information that has other types of risk decision results (e.g., risk decision results identified as good).

[0088] To facilitate understanding, an example is provided. Assume that the risk control data for a target category has two numerical groups containing values ​​of "10" and "20" respectively. If there are 100 operational records, 50 of these records contain risk control data for the target category with a value of "10," and 40 of these records have a risk decision result of "Yes," while the other 10 records have a risk decision result of "No." Similarly, 50 records contain risk control data for the target category with a value of "20," and 25 of these records have a risk decision result of "Yes," while the other 25 records have a risk decision result of "No."

[0089] If a risk decision result of "yes" belongs to a risk decision result marked as "bad", then in the numerical group containing the value data "10", Within the numerical group containing the value "20",

[0090] After calculating the evidence weights of the risk control data for the target category under each numerical group, the estimated information value VI of the risk control data for the target category at the risk control strategy can be calculated using the following formula:

[0091]

[0092] Where n is the number of numerical groups corresponding to the risk control data of the target type.

[0093] To facilitate understanding, an example will be provided. Continuing from the previous example, suppose... but

[0094] It can be seen that, generally speaking, the larger the VI value, the stronger the predictive ability of the risk control data of the target category to the risk decision results generated by the risk control strategy. Therefore, it is advisable to consider using the risk control data of the target category as an input parameter of the risk control strategy to improve its accuracy and effectiveness. Conversely, the smaller the VI value, the weaker the predictive ability of the risk control data of the target category to the risk decision results generated by the risk control strategy. Therefore, it is advisable to avoid using the risk control data of the target category as an input parameter of the risk control strategy to improve its operational efficiency.

[0095] In this embodiment of the specification, since the historical estimated value of risk control data and the calculation principle of the aforementioned estimated value can be the same, step 206: obtaining the historical estimated value of the risk control data may specifically include:

[0096] The historical estimated information value generated during the historical change detection process for a specified type of historical risk control data is obtained, and the historical estimated information value corresponding to the specified type of risk control data is obtained; the historical estimated information value is generated based on the evidence weight of the specified type of historical risk control data under each preset numerical group.

[0097] In the embodiments described in this specification, the historical change detection process may be performed before step 202. Figure 2 The process of the plan. Among them, historical risk control data can be risk control data that the risk control equipment has obtained from the data provider.

[0098] Since data providers may modify the risk control data they provide, the types of historical risk control data and the risk control data mentioned in step 202 may or may not be the same; no specific limitation is made. However, for historical risk control data of the same type, the preset value groupings corresponding to the risk control data should generally be consistent to ensure the comparability between the historical estimated information value and the estimated information value generated based on the historical risk control data and the risk control data, respectively. This, in turn, ensures the accuracy of the subsequent risk control data change detection results. This will not be elaborated further.

[0099] In the embodiments of this specification, for ease of understanding, the method for generating the detection results of changes in risk control data is explained here.

[0100] Specifically, step 208: Based on the estimated information value and the historical estimated information value, generate a change detection result for the risk control data provided by the data provider, which may include:

[0101] Determine whether the historical estimated information value corresponding to the risk control data of the target type has been obtained, and obtain the first determination result.

[0102] If the first judgment result indicates that the historical estimated information value corresponding to the risk control data of the target type has not been obtained, then a change detection result is generated to reflect the change of the risk control data of the target type provided by the data provider to the risk control data of a newly added type.

[0103] In the embodiments of this specification, since the risk control device typically has not acquired the newly added risk control data during the historical change detection process when there is newly added risk control data, it may not possess historical estimated information value for the newly added risk control data, thus failing to acquire such historical estimated information value. Therefore, when the first judgment result indicates that the historical estimated information value corresponding to the target type of risk control data has not been acquired, the risk control data of the target type can be determined to be newly added risk control data, ensuring timely and accurate detection and identification of changes in risk control data.

[0104] In addition, since data providers may also modify risk control data, such as changing the way the risk control data is generated, the meaning and function of the risk control data often change. Therefore, it is necessary to detect any modifications to the risk control data.

[0105] Specifically, step 208: Based on the estimated information value and the historical estimated information value, generate a change detection result for the risk control data provided by the data provider, which may further include:

[0106] If the first judgment result indicates that the historical estimated information value corresponding to the risk control data of the target type has been obtained, then the difference between the historical estimated information value corresponding to the risk control data of the target type and the estimated information value of the risk control data of the target type is calculated.

[0107] Determine whether the absolute value of the difference is greater than the second threshold to obtain the third determination result.

[0108] If the third judgment result indicates that the absolute value of the difference is greater than the second threshold, then a change detection result is generated to reflect that the risk control data of the target type provided by the data provider is risk control data that has been modified.

[0109] In this embodiment, the estimated information value / historical estimated information value can be used to measure the predictive ability of risk control data as a variable to the risk decision results generated by the risk control strategy. Generally, if the data provider has not modified the existing risk control data, the predictive ability of any type of risk control data to the risk decision results generated by the risk control strategy should be relatively stable; that is, the absolute value of the difference between the historical estimated information value and the actual estimated information value of the risk control data should be small. However, if the data provider has modified the existing risk control data, the predictive ability of that type of risk control data to the risk decision results generated by the risk control strategy will often change significantly; that is, the absolute value of the difference between the historical estimated information value and the actual estimated information value of the risk control data should be large. Based on this, whether the absolute value of the difference between the historical estimated information value and the actual estimated information value of the risk control data is greater than a second threshold can be used to detect whether the data provider has modified the risk control data. The second threshold can be set according to actual needs, for example, it can be 0.1, 0.2, etc., and is not specifically limited thereto.

[0110] In practical applications, if the third judgment result indicates that the absolute value of the difference is less than or equal to the second threshold, a change detection result can also be generated to reflect that the risk control data of the target type provided by the data provider is risk control data that has not been changed.

[0111] In the embodiments of this specification, after generating the detection results of changes in risk control data, it may also be necessary to train and update the risk control strategy based on the detection results of changes in risk control data, so as to make full use of the newly added types of risk control data and ensure the prediction accuracy and effectiveness of the risk control strategy.

[0112] Based on this, step 208: after generating the change detection result for the risk control data provided by the data provider based on the estimated information value and the historical estimated information value, may further include:

[0113] If the change detection result indicates that the risk control data of the target type is the risk control data of a newly added type, then it is determined whether the estimated information value of the risk control data of the target type is greater than the first threshold, and a second judgment result is obtained.

[0114] If the second judgment result indicates that the estimated information value of the risk control data of the target type is greater than the first threshold, then a first training sample containing the risk control data of the target type is generated based on the operation record information.

[0115] The risk control strategy is updated using the first training sample to obtain an updated risk control strategy; the updated risk control strategy is used to process risk control data of preset types, including the target type, to obtain risk decision results.

[0116] In the embodiments of this specification, when the risk control data of the target type is detected as newly added risk control data, if the estimated information value of the risk control data of the target type is greater than a first threshold, it generally indicates that the risk control data of the target type has a stronger predictive ability for the risk decision results generated by the risk control strategy. Therefore, the risk control data of the target type can be used as an input parameter for the risk control strategy to improve the accuracy and effectiveness of the risk control strategy. The first threshold can be set according to actual needs, for example, it can be 0.02, 0.1, etc., and is not specifically limited thereto.

[0117] Specifically, since the operation record information contains risk control data for the target type, the operation record information can be directly used to generate a first training sample containing risk control data for the target type. By using the subsequently verified risk decision results as the label data of the first training sample, the risk control strategy can be trained and updated using the first training sample carrying the label data to obtain the updated risk control strategy.

[0118] Subsequently, the updated risk control strategy trained can be deployed to the risk control engine, and a correspondence can be established between the updated risk control strategy and the key data of the target type's risk control data. This allows the updated risk control strategy to extract risk control data of preset types, including the target type, as input data during runtime, in order to generate more accurate risk decision results.

[0119] In practical applications, there may be multiple new types of risk control data whose estimated information value is greater than the first threshold. To improve the efficiency of strategy updates, the risk control data of each new type whose estimated information value is greater than the first threshold can be included in the first training sample. That is, the first training sample may contain multiple types of risk control data belonging to the new types (i.e., risk control data of the target types). Therefore, there may be multiple target types.

[0120] In this embodiment, since the risk control device can promptly and accurately detect and identify changes in risk control data without waiting for the data provider to actively inform it of such changes, it can adjust risk control strategies accordingly. This improves the effectiveness and reliability of risk control strategies and ensures secure business operations. Furthermore, the risk control device can utilize its local operational records to generate a first training sample containing newly added risk control data types, without needing to obtain this first training sample from the data provider. This saves device resources and improves the efficiency of obtaining the first training sample, further enhancing the real-time nature of risk control strategy adjustments and ensuring secure business operations.

[0121] In the embodiments of this specification, when the risk control data of the target type is detected to be modified, if the current risk control strategy needs to use the modified risk control data to generate risk decision results, it is necessary to identify whether the modified risk control data still has good predictive ability for the risk control strategy. If so, the risk control strategy does not need to be updated; otherwise, the risk control strategy needs to stop using the modified risk control data to generate risk decision results in the future, so as to ensure the operating efficiency and predictive accuracy of the risk control strategy.

[0122] Based on this, after generating the change detection result for the risk control data provided by the data provider according to the estimated information value and the historical estimated information value, it may further include:

[0123] If the change detection result indicates that the risk control data of the target type is modified, then it is determined whether the estimated information value of the risk control data of the target type is less than the third threshold, and the risk decision result is the decision result generated by the risk control strategy based on the risk control data of the target type, thus obtaining the fourth judgment result.

[0124] If the fourth judgment result indicates that the estimated information value of the risk control data of the target type is less than the third threshold, and the risk decision result is the decision result generated by the risk control strategy based on the risk control data of the target type, then a second training sample that does not contain the risk control data of the target type is generated based on the operation record information.

[0125] The risk control strategy is updated using the second training sample to obtain an updated risk control strategy; the updated risk control strategy is used to process risk control data of preset types that do not include the target type to obtain risk decision results.

[0126] In this embodiment of the specification, if the fourth judgment result indicates that the estimated information value of the risk control data for the target type is less than the third threshold, and the risk decision result is the decision result generated by the risk control strategy based on the risk control data for the target type, then it can be said that the modified risk control data for the target type no longer has a good predictive ability for the risk control strategy. Therefore, the risk control strategy no longer needs to use the modified risk control data for the target type to generate risk decision results. Thus, the operation record information can be directly used to generate a second training sample that does not contain the risk control data for that target type. By using the subsequently verified risk decision results as the label data of the second training sample, the risk control strategy can be trained and updated using the second training sample carrying the label data to obtain the updated risk control strategy. The third threshold can be set according to actual needs, and in practical applications, the third threshold and the first threshold can be the same or different, without specific limitations.

[0127] Subsequently, the updated risk control strategy obtained from the training can be deployed to the risk control engine, and the correspondence between the historically established risk control strategy and the key data of the target type risk control data can be canceled. This allows the updated risk control strategy to no longer extract the target type risk control data as input data when it is running.

[0128] In practical applications, there may be multiple types of risk control data (i.e., risk control data of target categories) that have been used by risk control strategies and whose estimated information value is less than the third threshold. Based on this, the second training sample can be made to no longer contain risk control data of the above target categories, so as to improve the efficiency of strategy update.

[0129] In this embodiment, since the risk control device can promptly and accurately detect and identify changes in risk control data without waiting for the data provider to actively inform it of such changes, it can adjust risk control strategies accordingly. This improves the effectiveness and reliability of risk control strategies and ensures secure business operations. Furthermore, the risk control device can utilize its local operational records to generate a second training sample containing newly added risk control data types, without needing to obtain this second training sample from the data provider. This saves device resources and improves the efficiency of obtaining the second training sample, further enhancing the real-time nature of risk control strategy adjustments and contributing to secure business operations.

[0130] In practical applications, when anomalies occur at the data provider's location, even if the data provider has not modified the risk control data for the target category, the risk control data for that category will still exhibit characteristics of modification. For example, the absolute value of the difference between the historical estimated information value and the estimated information value corresponding to the risk control data for the target category may exceed a second threshold. In such cases, updating the risk control strategy often does not improve its accuracy.

[0131] Based on this, before updating the risk control strategy using the second training sample, the process may further include:

[0132] Send data anomaly alarm information for the risk control data of the target type to the data provider's device.

[0133] Receive alarm feedback information sent by the device of the data provider.

[0134] Correspondingly, the risk control strategy is updated using the second training sample, which may specifically include:

[0135] If the alarm feedback information indicates that the risk control data of the target type does not have the risk of data anomaly, then the risk control strategy is updated using the second training sample.

[0136] If the alarm feedback information indicates that the risk control data of the target type has a data anomaly risk, then updating the risk control strategy using the second training sample can also be prohibited.

[0137] In this embodiment, by enabling the risk control device to communicate with the data provider, the detection results of changes identified in the risk control data for the target type can be further clarified. Specifically, it can be determined whether the changes were actually made by the data provider or due to temporary data anomalies. After confirming that the risk control data for the target type has not experienced any data anomalies but has indeed been modified by the data provider, the risk control strategy is updated using the second training sample. This helps to ensure the operational stability and predictive accuracy of the risk control strategy.

[0138] Figure 3 The embodiments provided in this specification correspond to Figure 2 The swimlane flowchart illustrates the method for detecting changes in risk control data. (Example) Figure 3 As shown, the change detection process for risk control data can involve entities such as risk control equipment and data provider equipment.

[0139] During the change detection phase, the risk control device can call a preset interface on the data provider's device to request risk control data. The data provider's device can then send a JSON string carrying the risk control data to the risk control device. The risk control device can parse the JSON string to obtain risk control data in key-value pair format, and process the preset types of risk control data using risk control strategies to obtain risk decision results. Based on the risk decision results and the obtained risk control data, the device can then generate operational log information.

[0140] The risk control equipment can also calculate the estimated information value of each risk control data point based on operational record information, and obtain the historical estimated information value of the risk control data. It then determines whether the historical estimated information value of the target type of risk control data has been obtained. If not, it can generate a change detection result indicating that the target type of risk control data is a newly added risk control data point. If so, it can determine whether the absolute value of the difference between the historical estimated information value and the estimated information value of the target type of risk control data is greater than a second threshold. If so, it can generate a change detection result indicating that the target type of risk control data is modified; otherwise, it can generate a change detection result indicating that the target type of risk control data is not modified.

[0141] During the strategy update phase, if the risk control data for the target type is newly added risk control data, the risk control device can determine whether the estimated information value of the risk control data for the target type is greater than a first threshold. If so, it can generate a first training sample containing the risk control data for the target type based on the operation record information, and use the first training sample to update the risk control strategy to obtain the updated risk control strategy. The updated risk control strategy is used to process the risk control data of preset types, including the target type, to obtain risk decision results.

[0142] If the risk control data for the target category is modified, it can be determined whether the estimated information value of the risk control data for the target category is less than a third threshold, and whether the risk control data for the target category is the data required by the risk control strategy. If so, a data anomaly alarm message for the risk control data of the target category can be sent to the data provider's device. The data provider's device can respond to the data anomaly alarm message by generating and sending alarm feedback information to the risk control device. If the risk control device determines that the alarm feedback message indicates no data anomaly risk, it can generate a second training sample that does not contain the risk control data of the target category based on the operation record information, and update the risk control strategy using the second training sample to obtain an updated risk control strategy. The updated risk control strategy is used to process risk control data of preset categories that do not include the target category to obtain risk decision results.

[0143] Based on the same idea, embodiments of this specification also provide apparatus corresponding to the above methods. Figure 4 The embodiments provided in this specification correspond to Figure 2 A schematic diagram of a change detection device for risk control data. Figure 4 As shown, the device may include:

[0144] The first acquisition module 402 is used to acquire the operation record information of the risk control strategy; the operation record information carries various risk control data provided by the data provider, and risk decision results generated by the risk control strategy based on the risk control data of preset types.

[0145] The estimated information value calculation module 404 is used to calculate the estimated information value of each of the risk control data at the risk control strategy based on the operation record information.

[0146] The second acquisition module 406 is used to acquire the historical estimated information value of the risk control data.

[0147] The detection result generation module 408 is used to generate a detection result of the change of risk control data provided by the data provider based on the estimated information value and the historical estimated information value.

[0148] based on Figure 4 The embodiments of this specification also provide some specific implementations of the device, which will be described below.

[0149] Optional, Figure 4 The device may further include:

[0150] The risk control data receiving module is used to receive key-value pair format risk control data to be parsed sent by the data provider.

[0151] The data parsing module is used to parse the risk control data to be parsed, and obtain various types of risk control data provided by the data provider.

[0152] The determination module is used to determine, from the various types of risk control data, the preset types of risk control data required for the risk control strategy.

[0153] The risk control processing module is used to process the preset types of risk control data using the risk control strategy to obtain the risk decision results.

[0154] The operation record information generation module is used to generate operation record information of the risk control strategy based on the risk decision results and the various types of risk control data.

[0155] Optionally, the risk control data receiving module can be specifically used to: receive a JSON string carrying risk control data sent by the data provider.

[0156] The data parsing module can be specifically used to: parse the JSON string to obtain risk control data in key-value pair format; wherein the key data contained in the risk control data is used to reflect the type of the risk control data, and the value data contained in the risk control data is used to reflect the numerical value of the risk control data.

[0157] Optionally, the estimated information value calculation module 404 may specifically include:

[0158] The numerical grouping determination unit is used to determine the numerical grouping corresponding to the risk control data of the target type based on the risk control data of the target type contained in the operation record information.

[0159] The evidence weight calculation unit is used to calculate the evidence weight of the risk control data of the target type under each of the numerical groups based on the risk decision results contained in the operation record information and the risk control data of the target type.

[0160] The estimated information value generation unit is used to generate the estimated information value of the risk control data of the target type at the risk control strategy based on the evidence weight under each of the numerical groupings.

[0161] Optionally, the second acquisition module 406 can be specifically used for:

[0162] The historical estimated information value generated during the historical change detection process for a specified type of historical risk control data is obtained, and the historical estimated information value corresponding to the specified type of risk control data is obtained; the historical estimated information value is generated based on the evidence weight of the specified type of historical risk control data under each preset numerical group.

[0163] Optionally, the detection result generation module 408 may include:

[0164] The first judgment result generation unit is used to determine whether the historical estimated information value corresponding to the risk control data of the target type has been obtained, and to obtain the first judgment result.

[0165] The first detection result generation unit is used to generate a detection result reflecting the change of the risk control data of the target type provided by the data provider to the risk control data of a newly added type if the first judgment result indicates that the historical estimated information value corresponding to the risk control data of the target type has not been obtained.

[0166] Optional, Figure 4 The device may further include:

[0167] The second judgment result generation module is used to determine whether the estimated information value of the risk control data of the target type is greater than a first threshold if the change detection result indicates that the risk control data of the target type is the risk control data of a newly added type, and to obtain a second judgment result.

[0168] The first training sample generation module is used to generate a first training sample containing the risk control data of the target type based on the operation record information if the second judgment result indicates that the estimated information value of the risk control data of the target type is greater than a first threshold.

[0169] The first update module is used to update the risk control strategy using the first training sample to obtain the updated risk control strategy; the updated risk control strategy is used to process risk control data of preset types, including the target type, to obtain risk decision results.

[0170] Optionally, the detection result generation module 408 may further include:

[0171] The difference calculation unit is used to calculate the difference between the historical estimated information value corresponding to the risk control data of the target type and the estimated information value of the risk control data of the target type if the first judgment result indicates that the historical estimated information value corresponding to the risk control data of the target type has been obtained.

[0172] The third judgment result generation unit is used to determine whether the absolute value of the difference is greater than the second threshold, and to obtain the third judgment result.

[0173] The second detection result generation unit is used to generate a change detection result reflecting that the risk control data of the target type provided by the data provider is modified risk control data if the third judgment result indicates that the absolute value of the difference is greater than the second threshold.

[0174] Optional, Figure 4 The device may further include:

[0175] The fourth judgment result generation module is used to determine whether the estimated information value of the risk control data of the target type is less than a third threshold if the change detection result indicates that the risk control data of the target type is modified risk control data, and the risk decision result is the decision result generated by the risk control strategy based on the risk control data of the target type, thus obtaining the fourth judgment result.

[0176] The second training sample generation module is used to generate a second training sample that does not contain the risk control data of the target type if the fourth judgment result indicates that the estimated information value of the risk control data of the target type is less than the third threshold, and the risk decision result is the decision result generated by the risk control strategy based on the risk control data of the target type.

[0177] The second update module is used to update the risk control strategy using the second training sample to obtain the updated risk control strategy; the updated risk control strategy is used to process risk control data of preset types that do not include the target type to obtain risk decision results.

[0178] Optional, Figure 4 The device may further include:

[0179] An alarm information sending module is used to send data anomaly alarm information for the risk control data of the target type to the device of the data provider.

[0180] The alarm feedback information receiving module is used to receive alarm feedback information sent by the data provider's device.

[0181] The second update module can be specifically used for:

[0182] If the alarm feedback information indicates that the risk control data of the target type does not have the risk of data anomaly, then the risk control strategy is updated using the second training sample.

[0183] Based on the same idea, this specification also provides devices corresponding to the above methods in its embodiments.

[0184] Figure 5 The embodiments provided in this specification correspond to Figure 2 A schematic diagram of a change detection device for risk control data. (See diagram below.) Figure 5 As shown, device 500 may include:

[0185] At least one processor 510; and,

[0186] Memory 530 communicatively connected to the at least one processor; wherein,

[0187] The memory 530 stores instructions 520 that can be executed by the at least one processor 510, the instructions being executed by the at least one processor 510 to enable the at least one processor 510 to:

[0188] Obtain the operation record information of the risk control strategy; the operation record information carries various risk control data provided by the data provider, and risk decision results generated by the risk control strategy based on the risk control data of preset types.

[0189] Based on the operation record information, calculate the estimated information value of each of the risk control data at the risk control strategy.

[0190] The historical forecast information value of the risk control data is obtained.

[0191] Based on the estimated information value and the historical estimated information value, a change detection result is generated for the risk control data provided by the data provider.

[0192] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on its differences from other embodiments. In particular, for... Figure 5 As the device shown is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0193] In the 1990s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the methodology). However, with technological advancements, many methodological improvements today can be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that a methodological improvement cannot be implemented using hardware physical modules. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program a digital system themselves to "integrate" it onto a PLD, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software. Similar to the software compiler used in program development, the original code before compilation must be written in a specific programming language, called a Hardware Description Language (HDL). There are many HDLs, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, the most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also understand that by simply performing some logic programming on the method flow using one of these hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0194] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, ASICs, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0195] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0196] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0197] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0198] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0199] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0200] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0201] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0202] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0203] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0204] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0205] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0206] This application can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0207] The above description is merely an embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of this application should be included within the scope of the claims of this application.

Claims

1. A method for detecting changes in risk control data, comprising: Obtain operational logs of risk control strategies; The operation record information carries various risk control data provided by the data provider, as well as risk decision results generated by the risk control strategy based on the preset types of risk control data; Based on the operation record information, the estimated information value of each of the risk control data at the risk control strategy is calculated, wherein the estimated information value of each of the risk control data at the risk control strategy is calculated based on the risk control data contained in multiple operation record information and the risk decision results; The historical forecast information value of the risk control data; Based on the estimated information value and the historical estimated information value, a change detection result is generated for the risk control data provided by the data provider. The change detection result for each risk control data provided by the data provider is generated based on the comparison result of comparing the estimated information value and the historical estimated information value. The change detection result is used to reflect the changes in the risk control data provided by the data provider.

2. The method as described in claim 1, further comprising, before obtaining the operation record information of the risk control strategy: Receive key-value pair format risk control data to be parsed from the data provider; The risk control data to be parsed is parsed to obtain various types of risk control data provided by the data provider; From the various types of risk control data, determine the preset types of risk control data required for the risk control strategy; The risk control strategy is used to process the risk control data of the preset types to obtain the risk decision results; Based on the risk decision results and the various types of risk control data, the operation record information of the risk control strategy is generated.

3. The method as described in claim 2, wherein the key-value pair format risk control data to be parsed sent by the data provider specifically includes: Receive a JSON string containing risk control data sent by the data provider; The process of parsing the risk control data to be parsed yields various types of risk control data provided by the data provider, specifically including: The JSON string is parsed to obtain risk control data in key-value pair format; wherein the key data of the risk control data is used to reflect the type of the risk control data, and the value data of the risk control data is used to reflect the numerical value of the risk control data.

4. The method as described in claim 1, wherein calculating the estimated information value of each of the risk control data at the risk control strategy based on the operation record information specifically includes: Based on the risk control data of the target type contained in the operation record information, determine the numerical grouping corresponding to the risk control data of the target type; Based on the risk decision results contained in the operation record information and the risk control data of the target type, calculate the evidence weight of the risk control data of the target type under each of the numerical groups; Based on the evidence weights under each of the numerical groupings, the estimated information value of the risk control data for the target category at the risk control strategy is generated.

5. The method as described in claim 4, wherein obtaining the historical estimated information value of the risk control data specifically includes: The historical estimated information value generated during the historical change detection process for a specified type of historical risk control data is obtained, and the historical estimated information value corresponding to the specified type of risk control data is obtained; the historical estimated information value is generated based on the evidence weight of the specified type of historical risk control data under each preset numerical group.

6. The method as described in claim 4 or 5, wherein generating a change detection result for the risk control data provided by the data provider based on the estimated information value and the historical estimated information value specifically includes: Determine whether the historical estimated information value corresponding to the risk control data of the target type has been obtained, and obtain a first determination result; If the first judgment result indicates that the historical estimated information value corresponding to the risk control data of the target type has not been obtained, then a change detection result is generated to reflect the change of the risk control data of the target type provided by the data provider to the risk control data of a newly added type.

7. The method of claim 6, further comprising, after generating a change detection result for the risk control data provided by the data provider based on the estimated information value and the historical estimated information value: If the change detection result indicates that the risk control data of the target type is the risk control data of a newly added type, then it is determined whether the estimated information value of the risk control data of the target type is greater than the first threshold, and a second judgment result is obtained. If the second judgment result indicates that the estimated information value of the risk control data of the target type is greater than the first threshold, then a first training sample containing the risk control data of the target type is generated based on the operation record information. The risk control strategy is updated using the first training sample to obtain the updated risk control strategy; The updated risk control strategy is used to process risk control data of preset categories, including the target category, to obtain risk decision results.

8. The method of claim 6, wherein generating a change detection result for the risk control data provided by the data provider based on the estimated information value and the historical estimated information value further includes: If the first judgment result indicates that the historical estimated information value corresponding to the risk control data of the target type has been obtained, then the difference between the historical estimated information value corresponding to the risk control data of the target type and the estimated information value of the risk control data of the target type is calculated. Determine whether the absolute value of the difference is greater than the second threshold to obtain a third determination result; If the third judgment result indicates that the absolute value of the difference is greater than the second threshold, then a change detection result is generated to reflect that the risk control data of the target type provided by the data provider is risk control data that has been modified.

9. The method of claim 8, further comprising, after generating a change detection result for the risk control data provided by the data provider based on the estimated information value and the historical estimated information value: If the change detection result indicates that the risk control data of the target type is modified, then it is determined whether the estimated information value of the risk control data of the target type is less than the third threshold, and the risk decision result is the decision result generated by the risk control strategy based on the risk control data of the target type, thus obtaining the fourth judgment result; If the fourth judgment result indicates that the estimated information value of the risk control data of the target type is less than the third threshold, and the risk decision result is the decision result generated by the risk control strategy based on the risk control data of the target type, then a second training sample that does not contain the risk control data of the target type is generated based on the operation record information. The risk control strategy is updated using the second training sample to obtain the updated risk control strategy; The updated risk control strategy is used to process risk control data of preset categories that do not include the target category in order to obtain risk decision results.

10. The method of claim 9, further comprising, before updating the risk control strategy using the second training sample: Send data anomaly alarm information for the risk control data of the target type to the data provider's device; Receive alarm feedback information sent by the data provider's device; The step of updating the risk control strategy using the second training sample specifically includes: If the alarm feedback information indicates that the risk control data of the target type does not have the risk of data anomaly, then the risk control strategy is updated using the second training sample.

11. A change detection device for risk control data, comprising: The first acquisition module is used to acquire the operation record information of the risk control strategy; The operation record information carries various risk control data provided by the data provider, as well as risk decision results generated by the risk control strategy based on the preset types of risk control data; The estimated information value calculation module is used to calculate the estimated information value of each of the risk control data at the risk control strategy based on the operation record information, wherein the estimated information value of each of the risk control data at the risk control strategy is calculated based on the risk control data contained in multiple operation record information and the risk decision results; The second acquisition module is used to acquire the historical estimated information value of the risk control data; The detection result generation module is used to generate a change detection result for the risk control data provided by the data provider based on the estimated information value and the historical estimated information value. The change detection result for each risk control data provided by the data provider is generated based on the comparison result of comparing the estimated information value and the historical estimated information value. The change detection result is used to reflect the change of the risk control data provided by the data provider.

12. The apparatus of claim 11, further comprising: The risk control data receiving module is used to receive key-value pair format risk control data to be parsed sent by the data provider. The data parsing module is used to parse the risk control data to be parsed, and obtain various types of risk control data provided by the data provider; The determination module is used to determine, from the various types of risk control data, the preset types of risk control data required for the risk control strategy; The risk control processing module is used to process the preset types of risk control data using the risk control strategy to obtain the risk decision results; The operation record information generation module is used to generate operation record information of the risk control strategy based on the risk decision results and the various types of risk control data.

13. The apparatus of claim 12, wherein the risk control data receiving module is specifically used for: Receive a JSON string containing risk control data sent by the data provider; The data parsing module is specifically used for: parsing the JSON string to obtain risk control data in each key-value pair format; wherein The risk control data includes key data to reflect the type of risk control data, and value data to reflect the numerical value of risk control data.

14. The apparatus of claim 11, wherein the estimated information value calculation module comprises: The numerical grouping determination unit is used to determine the numerical grouping corresponding to the risk control data of the target type based on the risk control data of the target type contained in the operation record information. The evidence weight calculation unit is used to calculate the evidence weight of the risk control data of the target type under each of the numerical groups based on the risk decision results contained in the operation record information and the risk control data of the target type. The estimated information value generation unit is used to generate the estimated information value of the risk control data of the target type at the risk control strategy based on the evidence weight under each of the numerical groupings.

15. The apparatus of claim 14, wherein the second acquisition module is specifically used for: The historical estimated information value generated during the historical change detection process for a specified type of historical risk control data is obtained, and the historical estimated information value corresponding to the specified type of risk control data is obtained; the historical estimated information value is generated based on the evidence weight of the specified type of historical risk control data under each preset numerical group.

16. The apparatus of claim 14 or 15, wherein the detection result generation module comprises: The first judgment result generation unit is used to determine whether the historical estimated information value corresponding to the risk control data of the target type has been obtained, and to obtain the first judgment result; The first detection result generation unit is used to generate a detection result reflecting the change of the risk control data of the target type provided by the data provider to the risk control data of a newly added type if the first judgment result indicates that the historical estimated information value corresponding to the risk control data of the target type has not been obtained.

17. The apparatus of claim 16, further comprising: The second judgment result generation module is used to determine whether the estimated information value of the risk control data of the target type is greater than a first threshold if the change detection result indicates that the risk control data of the target type is the risk control data of a newly added type, and to obtain a second judgment result. The first training sample generation module is used to generate a first training sample containing the risk control data of the target type based on the operation record information if the second judgment result indicates that the estimated information value of the risk control data of the target type is greater than a first threshold. The first update module is used to update the risk control strategy using the first training sample to obtain the updated risk control strategy. The updated risk control strategy is used to process risk control data of preset categories, including the target category, to obtain risk decision results.

18. The apparatus of claim 16, wherein the detection result generation module further comprises: The difference calculation unit is used to calculate the difference between the historical estimated information value corresponding to the risk control data of the target type and the estimated information value of the risk control data of the target type if the first judgment result indicates that the historical estimated information value corresponding to the risk control data of the target type has been obtained. The third judgment result generation unit is used to determine whether the absolute value of the difference is greater than the second threshold, and to obtain the third judgment result. The second detection result generation unit is used to generate a change detection result reflecting that the risk control data of the target type provided by the data provider is modified risk control data if the third judgment result indicates that the absolute value of the difference is greater than the second threshold.

19. The apparatus of claim 18, further comprising: The fourth judgment result generation module is used to determine whether the estimated information value of the risk control data of the target type is less than a third threshold if the change detection result indicates that the risk control data of the target type is modified risk control data, and the risk decision result is the decision result generated by the risk control strategy based on the risk control data of the target type, thus obtaining the fourth judgment result; The second training sample generation module is used to generate a second training sample that does not contain the risk control data of the target type if the fourth judgment result indicates that the estimated information value of the risk control data of the target type is less than the third threshold, and the risk decision result is the decision result generated by the risk control strategy based on the risk control data of the target type. The second update module is used to update the risk control strategy using the second training sample to obtain the updated risk control strategy. The updated risk control strategy is used to process risk control data of preset categories that do not include the target category in order to obtain risk decision results.

20. The apparatus of claim 19, further comprising: An alarm information sending module is used to send data anomaly alarm information of the risk control data for the target type to the device of the data provider; An alarm feedback information receiving module is used to receive alarm feedback information sent by the data provider's device; The second update module is specifically used for: If the alarm feedback information indicates that the risk control data of the target type does not have the risk of data anomaly, then the risk control strategy is updated using the second training sample.

21. A change detection device for risk control data, comprising: At least one processor; as well as, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to: Obtain the operation record information of the risk control strategy; the operation record information carries various risk control data provided by the data provider, and risk decision results generated by the risk control strategy based on the risk control data of preset types; Based on the operation record information, the estimated information value of each of the risk control data at the risk control strategy is calculated, wherein the estimated information value of each of the risk control data at the risk control strategy is calculated based on the risk control data contained in multiple operation record information and the risk decision results; The historical forecast information value of the risk control data; Based on the estimated information value and the historical estimated information value, a change detection result is generated for the risk control data provided by the data provider. The change detection result for each risk control data provided by the data provider is generated based on the comparison result of comparing the estimated information value and the historical estimated information value. The change detection result is used to reflect the changes in the risk control data provided by the data provider.

Images