A method, apparatus, and device for processing data

Abstract

The embodiments of this specification disclose a method, apparatus, and device for data processing. The method includes: obtaining first features required for constructing a trial qualification strategy for a preset risk, and obtaining basic features corresponding to second features. A trial qualification strategy is constructed based on the first features and the basic features. Complaint data submitted by a user for the preset risk is obtained. If there are features in the third features corresponding to the complaint data that are not included in the first features and / or the basic features, then the third features are retroactively processed based on the basic features. Based on the retroactive information corresponding to the third features, the first features, and the basic features, the complaint data is subject to trial qualification processing to obtain a first trial result. If the first trial result is better than the second trial result obtained by subjecting the complaint data to trial qualification processing based on the trial qualification strategy, then the trial qualification strategy is updated based on the retroactive information corresponding to the third features and the first features.

Description

Technical Field

[0001] This document relates to the field of computer technology, and in particular to a data processing method, device and equipment. Background Art

[0002] Risk complaints or risk reports refer to the act of users reporting or exposing the risk facts or suspected risk events discovered during the use of a certain service to the service platform through manual customer service or self-service channels. For users, risk complaints are one of the few scenarios where users actively provide security feedback. Therefore, the review and characterization of risk complaints plays a very important role in the risk control system and is an indispensable key link.

[0003] At present, the method of case qualitative backtracking is generally to use the current online trial qualitative strategy to conduct a trial qualitative review of historical data, that is, to use the online formal trial qualitative review strategy to judge the historical data, and evaluate the historical data to obtain the corresponding trial qualitative review results. However, the above method may have the phenomenon of feature crossing. Specifically, when the party complained against is complained against, there are multiple users complaining in a short period of time (such as 7 days or 30 days, etc.), but when the case is qualitatively traced back, there are no user complaints in a short period of time. This will make the previous and subsequent features inconsistent, which may lead to differences in the case qualitative results. For this reason, it is necessary to provide a technical solution that can ensure that the features used in qualitative backtracking are effective, so as to effectively avoid the occurrence of feature crossing phenomena, so as to overcome the inconsistency between the qualitative backtracking results and the expected results of the current trial qualitative review strategy. Summary of the invention

[0004] The purpose of the embodiments of this specification is to provide a technical solution that can ensure that the features used in qualitative backtracking are valid, thereby effectively avoiding the occurrence of feature crossing phenomena and overcoming the inconsistency between the qualitative backtracking results and the expected results of the current trial qualitative strategy.

[0005] In order to implement the above technical solution, the embodiments of this specification are implemented as follows:

[0006] A data processing method provided by an embodiment of this specification, the method includes: obtaining a first feature required for constructing a trial qualitative strategy for a preset risk, and obtaining a basic feature corresponding to a second feature, constructing a trial qualitative strategy for the preset risk based on the first feature and the basic feature, where the second feature is a feature with a complexity higher than a preset complexity threshold. Obtaining complaint data submitted by a user for the preset risk. If there are features in the third feature corresponding to the complaint data that are not included in the first feature and / or the basic feature, then perform a backtracking process on the third feature based on the basic feature to obtain backtracking information corresponding to the third feature, and perform a trial qualitative process on the complaint data based on the backtracking information corresponding to the third feature, the first feature, and the basic feature to obtain a corresponding first trial result. If the first trial result is better than a second trial result obtained by performing a trial qualitative process on the complaint data based on the trial qualitative strategy, then update the trial qualitative strategy based on the backtracking information corresponding to the third feature and the first feature.

[0007] A data processing device provided by an embodiment of this specification, the device includes: a strategy construction module, which obtains a first feature required for constructing a trial qualitative strategy for a preset risk, and obtains a basic feature corresponding to a second feature, and constructs a trial qualitative strategy for the preset risk based on the first feature and the basic feature, where the second feature is a feature with a complexity higher than a preset complexity threshold. A data acquisition module, which acquires complaint data submitted by a user for the preset risk. A trial module, if there are features in the third feature corresponding to the complaint data that are not included in the first feature and / or the basic feature, then perform a backtracking process on the third feature based on the basic feature to obtain backtracking information corresponding to the third feature, and perform a trial qualitative process on the complaint data based on the backtracking information corresponding to the third feature, the first feature, and the basic feature to obtain a corresponding first trial result. A strategy update module, if the first trial result is better than a second trial result obtained by performing a trial qualitative process on the complaint data based on the trial qualitative strategy, then update the trial qualitative strategy based on the backtracking information corresponding to the third feature and the first feature.

[0008] A data processing device provided in an embodiment of this specification, the data processing device includes: a processor; and a memory arranged to store computer-executable instructions, the executable instructions, when executed, cause the processor to: obtain a first feature required for constructing a trial qualitative strategy for a preset risk, and obtain a basic feature corresponding to a second feature, construct a trial qualitative strategy for the preset risk based on the first feature and the basic feature, the second feature being a feature with a complexity higher than a preset complexity threshold. Obtain complaint data submitted by a user for the preset risk. If there are features in the third feature corresponding to the complaint data that are not included in the first feature and / or the basic feature, then perform a backtracking process on the third feature based on the basic feature to obtain backtracking information corresponding to the third feature, and perform a trial qualitative process on the complaint data based on the backtracking information corresponding to the third feature, the first feature, and the basic feature to obtain a corresponding first trial result. If the first trial result is better than a second trial result obtained by performing a trial qualitative process on the complaint data based on the trial qualitative strategy, then update the trial qualitative strategy based on the backtracking information corresponding to the third feature and the first feature.

[0009] An embodiment of this specification also provides a storage medium, the storage medium is used to store computer-executable instructions, and the executable instructions, when executed by a processor, implement the following process: obtain a first feature required for constructing a trial qualitative strategy for a preset risk, and obtain a basic feature corresponding to a second feature, construct a trial qualitative strategy for the preset risk based on the first feature and the basic feature, the second feature being a feature with a complexity higher than a preset complexity threshold. Obtain complaint data submitted by a user for the preset risk. If there are features in the third feature corresponding to the complaint data that are not included in the first feature and / or the basic feature, then perform a backtracking process on the third feature based on the basic feature to obtain backtracking information corresponding to the third feature, and perform a trial qualitative process on the complaint data based on the backtracking information corresponding to the third feature, the first feature, and the basic feature to obtain a corresponding first trial result. If the first trial result is better than a second trial result obtained by performing a trial qualitative process on the complaint data based on the trial qualitative strategy, then update the trial qualitative strategy based on the backtracking information corresponding to the third feature and the first feature. Brief Description of the Drawings

[0010] To more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only some embodiments recorded in this specification. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0011] Figure 1 This is an embodiment of a data processing method in this specification;

[0012] Figure 2 This is another embodiment of a data processing method in this specification;

[0013] Figure 3 This is a schematic structural diagram of a data processing system in this specification;

[0014] Figure 4 This is an embodiment of a data processing device in this specification;

[0015] Figure 5 This is an embodiment of a data processing device in this specification. Detailed implementation manners

[0016] The embodiments of this specification provide a data processing method, device, and device.

[0017] In order to enable those skilled in the art to better understand the technical solutions in this specification, the following will clearly and completely describe the technical solutions in the embodiments of this specification with reference to the accompanying drawings in the embodiments of this specification. Obviously, the described embodiments are only some embodiments of this specification, rather than all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this specification.

[0018] Embodiment 1

[0019] Such as Figure 1As shown in the figure, an embodiment of this specification provides a data processing method. The execution subject of this method can be a terminal device or a server, etc. The terminal device can be a mobile terminal device such as a mobile phone or a tablet computer, or a computer device such as a laptop or a desktop computer, or it can also be an IoT device (such as a smart watch, a vehicle-mounted device, etc.). The server can be an independent server or a server cluster composed of multiple servers. The server can be a background server for financial services or online shopping services, etc., or a background server for a certain application program, etc. In this embodiment, the server is used as the execution subject for detailed description. For the case where the execution subject is a terminal device, the situation of the following server can be referred to for processing, which will not be elaborated here. This method can specifically include the following steps:

[0020] In step S102, obtain the first features required to construct a trial and determination strategy for a preset risk, and obtain the basic features corresponding to the second features. Based on the first features and the basic features, construct a trial and determination strategy for the preset risk, where the second features are features with a complexity higher than a preset complexity threshold.

[0021] Among them, the preset risk can include multiple types. For example, the preset risk can be a fraud risk, an appropriation risk, etc., which can be specifically set according to the actual situation, and this specification embodiment does not limit this. The trial and determination strategy can be a strategy that after a user complains (or reports) a certain risk, the risk prevention and control system will combine the complaint information submitted by the user and other relevant information to determine whether there is a risk in the case complained by the user, what type of risk it is, and what means are used to achieve the risk, etc. The trial and determination strategy can include multiple types. For example, it can include the key features of the user complaint information and the account information of the party being complained about, etc., which can be specifically set according to the actual situation, and this specification embodiment does not limit this. The first features can be relatively commonly used features for constructing the trial and determination strategy. For example, if the preset risk is a fraud risk, the first features can include the key features of the fraud risk, etc., and can be obtained through multiple different channels, which can be specifically set according to the actual situation. The complexity threshold can be set according to the actual situation, such as 60% or 80%, etc.

[0022] In implementation, a risk complaint or risk report is an act in which a user, during the process of using a certain service, reports and exposes to the service platform the discovered risk facts or events with suspected risks through means such as manual customer service or self-service channels. For example, after a user discovers that their account or funds have been stolen, they report the above risks of theft or fraud. After the user reports the existing risks, the risk control system of the service will combine the complaint information submitted by the user (such as text descriptions, evidentiary images, etc.) and factual information (such as transaction information, account information, the relationship between the complainant and the respondent, etc.) to determine whether the event complained by the user has risks, what types of risks it is (such as theft risks, fraud risks, illegal financing risks, etc.), and what methods are used to achieve the risks (such as fraud by impersonating public security, procuratorial, and judicial organs, mobile phone loss and theft, etc.). Therefore, the trial and determination of risk complaints play a very important role in the risk control system. For users, risk complaints are one of the few scenarios where users actively provide security feedback. The trial and determination of cases are important means to provide users with security solutions and meet their reasonable security demands. For the risk control system, the trial and determination are the radar for perceiving risks, an important source for obtaining risk samples, and also an objective and scientific risk level quantification indicator. Therefore, the qualitative retrospective of risk complaints has important practical significance for the trial and determination and the risk control system and is an indispensable key link. Through the qualitative retrospective of historical cases, the changing trends of the same type of risk and risk methods can be measured more objectively and fairly, and then an accurate evaluation of the risk management effect in the past stage can be made. At the same time, it provides a scientific basis for the reasonable setting of the risk control management objectives in the next stage. Moreover, it can provide more accurate and comprehensive risk samples for model training, strategy construction, etc.

[0023] At present, the general way of case qualitative retrospective is to use the current online trial qualitative strategy to conduct qualitative trials on historical data, that is, to use the officially effective online trial qualitative strategy to judge historical data and evaluate historical data to obtain corresponding trial qualitative results. However, there will be a phenomenon of feature crossing in the above method. That is, in the trial qualitative strategy, in addition to using unstructured features such as text descriptions and evidentiary images in user complaints, a large number of structured features are also used (such as the records of the party being complained against, the transaction information of the party being complained against, the transaction historical data between the complainant and the party being complained against, etc.). If the current online trial qualitative strategy is directly used for qualitative retrospective of historical data, there will be a phenomenon of feature crossing. Specifically, for example, when the party being complained against is complained by multiple users within a short period (such as 7 days or 30 days, etc.), but there are no user complaints within a short period when conducting qualitative retrospective on the case, which will make the features before and after inconsistent, thus may lead to differences in the case qualitative results. Therefore, it is necessary to provide a technical solution that can ensure that the features used in qualitative retrospective are effective, so as to effectively avoid the occurrence of feature crossing phenomenon and overcome the inconsistency between the qualitative retrospective result and the expected result of the current trial qualitative strategy. The embodiments of this specification provide an implementable technical solution, which can specifically include the following content.

[0024] In order to avoid the phenomenon of feature crossing, it is necessary to sort out in advance the features that may be used in the trial qualitative strategy. It is difficult to completely predict in practice the features that may be used in the trial qualitative strategy to be constructed in the future. In this embodiment, the corresponding features can be collected in two ways to construct the trial qualitative strategy. First, the first features required for constructing the trial qualitative strategy for preset risks can be obtained through a variety of different ways. For example, the above first features can be obtained from a specified database, which can be used to store the first features for constructing a variety of different trial qualitative strategies. The first features therein can be obtained from a variety of different platforms or systems. Or, the historical complaint data for preset risks stored in advance can be obtained, the features of the historical complaint data can be extracted, and the features that can represent the commonly used features of the trial qualitative strategy can be selected from the extracted features, and the selected features can be used as the above first features. Or, the first features required for constructing the trial qualitative strategy can also be determined based on expert experience.

[0025] In addition, considering that there may be more complex features (i.e., the second feature) in the future, in order to construct the above-mentioned second feature, the basic features for constructing the second feature can be obtained in various different ways. For example, the basic features for constructing the second feature can be obtained from a specified database, which can be used to store the basic features corresponding to various different second features. The second feature can be obtained from various different platforms or systems, or, based on expert experience, the basic features (i.e., the basic features) that may be used to process complex features (i.e., the second feature) in the future can be sorted out. Most of the new features different from the first feature (which can be the second feature with a complexity higher than the preset complexity threshold) used in the subsequent constructed trial qualitative strategy can be processed from the above basic features.

[0026] Corresponding calculations can be performed based on the first feature and the basic features. Through the calculations, a trial qualitative strategy for the preset risk can be obtained. Specifically, according to historical data, the sorting of the first feature and the basic features in terms of time can be determined, so as to obtain the time series corresponding to the first feature and the basic features. A trial qualitative strategy for the preset risk can be constructed based on this time series. In practical applications, in addition to constructing the initial trial qualitative strategy in the above manner, the initial trial qualitative strategy can also be constructed in various different ways, which can be specifically set according to the actual situation, and the embodiments of this specification do not make any limitations in this regard.

[0027] In step S104, complaint data submitted by the user for the preset risk is obtained.

[0028] Among them, the user can be any user. In this embodiment, the user can be a user who encounters the preset risk during the execution of a certain business, which can be specifically set according to the actual situation. The complaint data can include various different data. For example, the text information describing the preset risk by the user, the information with the preset risk provided by the party being complained about to the complainant (which can be text information or screenshots, etc.), the images for evidence, and can also include information such as the account information of the complainant, the account information of the party being complained about, and the transaction information between the complainant and the party being complained about. It can be specifically set according to the actual situation, and the embodiments of this specification do not make any limitations in this regard.

[0029] In practice, when a user discovers the existence of a preset risk during the execution of a certain business, the user can search for the complaint mechanism provided by the business and can file a complaint with the business platform or the corresponding risk prevention and control system through this complaint mechanism. After the user fills in the information of each item of the complaint (i.e., the complaint data), the above complaint information can be submitted to the risk prevention and control system. When it is necessary to conduct a trial and qualitative analysis of the complaint data, the complaint data submitted by the user for the preset risk can be obtained.

[0030] In step S106, if there are features in the third feature corresponding to the above complaint data that are not included in the first feature and / or the basic feature, then backtracking processing is performed on the third feature based on the basic feature to obtain the backtracking information corresponding to the third feature, and the above complaint data is processed for trial and qualitative determination based on the backtracking information corresponding to the third feature, the first feature, and the basic feature to obtain the corresponding first trial result.

[0031] In implementation, after obtaining the complaint data submitted by the user through the above method, feature extraction can be performed on the complaint data to obtain the features included in the complaint data (i.e., the third feature). Then, the third feature can be analyzed. Through analysis, it can be obtained whether the third feature is included in the first feature and / or the basic feature. If the third feature is included in the first feature and / or the basic feature, then the complaint data can continue to be processed for trial and qualitative determination based on the above constructed trial and qualitative determination model. If there are features in the third feature that are not included in the first feature and / or the basic feature, it indicates that the complaint data contains newly added features. At this time, calculations can be performed based on the basic feature to perform backtracking processing on the third feature to obtain the backtracking information corresponding to the third feature. Subsequently, the above complaint data can be processed for trial and qualitative determination based on the backtracking information corresponding to the third feature and in combination with the first feature and the basic feature to obtain the corresponding first trial result.

[0032] In step S108, if the first trial result is better than the second trial result obtained by processing the above complaint data for trial and qualitative determination based on the trial and qualitative determination strategy, then the trial and qualitative determination strategy is updated based on the backtracking information corresponding to the third feature and the first feature.

[0033] In implementation, in order to determine whether the above first trial result is a better result, the above trial and qualitative determination strategy can be used to calculate the above complaint data, so as to process the above complaint data for trial and qualitative determination to obtain the corresponding second trial result. Then, the first trial result and the second trial result can be compared and analyzed to determine whether the first trial result is better than the second trial result. If the first trial result is better than the second trial result, it indicates that the features corresponding to the user's complaint data can play a positive role in the qualitative determination of the complaint trial. At this time, the above trial and qualitative determination strategy can be updated, that is, the trial and qualitative determination strategy can be updated based on the backtracking information corresponding to the third feature and the first feature to obtain the updated trial and qualitative determination strategy. The updated trial and qualitative determination strategy can be deployed to the corresponding business, so that during the execution of the business by the user, the complaint data provided by the user is processed for trial and qualitative determination.

[0034] The embodiments of this specification provide a method for processing data, by obtaining a first feature required for constructing a trial qualitative strategy for a preset risk, and obtaining a basic feature corresponding to a second feature, and constructing a trial qualitative strategy for the preset risk based on the first feature and the basic feature, wherein the second feature is a feature with a complexity higher than a preset complexity threshold, and then, complaint data submitted by a user for the preset risk can be obtained, and if a third feature corresponding to the complaint data contains a feature that is not included in the first feature and / or the basic feature, the third feature is back-traced based on the basic feature to obtain back-tracing information corresponding to the third feature, and the complaint data is reviewed and qualitatively processed based on the back-tracing information corresponding to the third feature, the first feature and the basic feature to obtain a corresponding first review result, and if the first review result is better than the first review result obtained by reviewing and qualitatively processing the complaint data based on the trial qualitative strategy For the second trial result, the trial qualitative strategy is updated based on the backtracking information corresponding to the third feature and the first feature. In this way, through the combing of core features (i.e., the first feature and the basic feature), the backtracking of new features or changed features, and the backtracking calculation of qualitative strategies, it can be ensured that the features used in the case qualitative backtracking are true and valid, in line with the time sequence, and can effectively avoid the occurrence of feature crossings. It can overcome the situation where the qualitative backtracking results of the risk event trial are inconsistent with the expected results of the current trial qualitative strategy, and through objective, true and accurate event qualitative backtracking methods, it can provide more accurate, complete and objective qualitative data assets, modeling training samples, etc. for subsequent risk prevention and control systems. It can also provide more objective and scientific risk levels and risk event quantitative indicators for each risk prevention and control system, so that the trial qualitative analysis can fully play its due role in risk control and become a better security infrastructure.

[0035] Embodiment 2

[0036] like Figure 2 As shown, an embodiment of this specification provides a method for processing data, and the execution subject of the method may be a terminal device or a server, etc., wherein the terminal device may be a mobile terminal device such as a mobile phone, a tablet computer, or a computer device such as a laptop or a desktop computer, or an IoT device (specifically a smart watch, a car device, etc.), etc., wherein the server may be an independent server, or a server cluster composed of multiple servers, etc., and the server may be a background server for a financial service or an online shopping service, or a background server for an application, etc. In this embodiment, a detailed description is given by taking the execution subject as a server as an example. For the case where the execution subject is a terminal device, please refer to the following server case processing, which will not be repeated here. The method may specifically include the following steps:

[0037] In step S202, historical complaint data for preset risks is obtained.

[0038] Among them, the preset risks include one or more of theft risk, fraud risk, and illegal financial activities. The historical complaint data can include various types. For example, the historical complaint data can include descriptive information about the preset risks, evidentiary images, account information of the party being complained against, transaction information between the user (i.e., the complainant) and the party being complained against, etc. Specifically, it can be set according to the actual situation, and the embodiments of this specification do not limit this.

[0039] In implementation, the historical complaint data for the preset risks can be obtained through various different methods. For example, as Figure 3 shown, the historical complaint data for the preset risks can be obtained from a specified database. This database can be used to store the historical complaint data provided by multiple different users for the preset risks. The historical complaint data therein can be obtained from various different platforms or systems. Or, the pre-stored historical complaint data can be obtained, or some of the historical complaint data obtained above can be selected as the historical complaint data for the preset risks.

[0040] In step S204, based on the historical complaint data for the preset risks, the first features required for constructing the trial and determination strategy for the preset risks are determined.

[0041] Among them, the trial and determination strategy for the preset risks can include one or more of the content description information of the complaint data, evidentiary images, complaint records of the party being complained against, transaction information of the party being complained against, and transaction information between the complainant and the party being complained against.

[0042] In implementation, the historical complaint data can be analyzed to sort out the commonly used features (i.e., the first features) used in the currently to-be-constructed trial and determination strategy. In actual applications, the number of the first features can be limited by a certain number. For example, the number of the first features can not exceed 1000, etc. Specifically, it can be set according to the actual situation, and the embodiments of this specification do not limit this. Then, through the analysis of the historical complaint data, the above-mentioned first features can be extracted from the historical complaint data.

[0043] In step S206, the basic features corresponding to the second features are obtained, and based on the first features and the basic features, the trial and determination strategy for the preset risks is constructed. The second features are features with a complexity higher than the preset complexity threshold.

[0044] For the specific processing process of the above step S206, reference can be made to the relevant content in the first embodiment above, and details will not be repeated here.

[0045] In step S208, the first features and the basic features are respectively calculated to obtain the feature values corresponding to the first features and the feature values corresponding to the basic features.

[0046] In step S210, store the eigenvalue corresponding to the first feature and the eigenvalue corresponding to the basic feature.

[0047] Regarding the above steps S208 and S210, based on the first feature and the basic feature obtained through the processing of the above steps S202 to S06, calculations are performed during the process of adjudicating and classifying the complaint data to obtain the eigenvalue corresponding to the first feature and the eigenvalue corresponding to the basic feature, and the above eigenvalues are stored at a specified location (usually it can be a data table). That is, after each event completes a complaint, during the adjudication and classification, calculations will be performed on all the first features and basic features to obtain the eigenvalue corresponding to the first feature and the eigenvalue corresponding to the basic feature, and the above eigenvalues are stored. Compared with the transaction event volume and the operation event volume, the complaint event volume is usually not large, and the storage period can be limited to a certain duration, such as 6 months, 1 year, or 3 years, etc.

[0048] In step S212, obtain the complaint data submitted by the user for a preset risk.

[0049] For the specific processing process of the above step S212, reference can be made to the relevant content in the first embodiment above, and details will not be elaborated here.

[0050] In step S214, if there are features in the third feature corresponding to the complaint data that are not included in the first feature and / or the basic feature, then perform a backtracking process on the third feature based on the basic feature to obtain the backtracking information corresponding to the third feature.

[0051] For the specific processing process of the above step S214, reference can be made to the relevant content in the first embodiment above, and details will not be elaborated here.

[0052] In step S216, perform an adjudication and classification process on the above complaint data based on the backtracking information corresponding to the third feature, the eigenvalue corresponding to the first feature, and the eigenvalue corresponding to the basic feature to obtain the corresponding first adjudication result.

[0053] For the specific processing process of the above step S216, reference can be made to the relevant content in the first embodiment above, and details will not be elaborated here.

[0054] In step S218, analyze the first adjudication result and the second adjudication result through the swap-in and swap-out analysis method to obtain the corresponding analysis result.

[0055] Among them, the Swap In&Swap Out analysis method is a very important method in risk control strategy analysis. When a new strategy replaces an old strategy or a new model replaces an old model, the problem of user group replacement usually occurs. Whether it is model development or strategy analysis, it is necessary to master the Swap In&Swap Out analysis method proficiently to maximize the business value of the new strategy (or new model). Among them, Swap In refers to the user group that the new strategy admits while the old strategy rejects, and Swap Out refers to the user group that the new strategy rejects while the old strategy admits. Generally, it is expected to swap in a group of good accounts, swap out a group of bad accounts, or swap in more good accounts, so as to replace the old accounts with good accounts and reduce the overall bad debt rate.

[0056] In implementation, after the above processing, it is necessary to conduct a comparative analysis of the first trial result and the second trial result. Specifically, the Swap In&Swap Out analysis method can be used to evaluate events with inconsistent trial characterizations before and after. Whether the first trial result is better than the second trial result. If the first trial result is better than the second trial result, it can be determined that the trial characterization strategy corresponding to the first trial result is superior to the current trial characterization strategy (the currently actually effective trial characterization strategy online). Otherwise, it is determined that the trial characterization strategy corresponding to the first trial result has not been improved.

[0057] In step S220, if the above analysis result indicates that the first trial result is better than the second trial result, the trial characterization strategy is updated based on the retrospective information corresponding to the third feature and the first feature.

[0058] In implementation, after the above processing, if the first trial result meets the expected result, the retrospective information corresponding to the third feature and the first feature can be used to overwrite the first feature and the basic feature, that is, the trial characterization strategy can be updated based on the retrospective information corresponding to the third feature and the first feature to obtain the updated trial characterization strategy.

[0059] In step S222, the historical complaint data for the preset risk and the complaint data submitted by the user for the preset risk are provided as training samples to the risk prevention and control model for the preset risk to train the risk prevention and control model through the training samples; or, the first feature and the third feature are provided as training samples to the risk prevention and control model for the preset risk to train the risk prevention and control model through the training samples.

[0060] In implementation, as Figure 3 shown, the historical complaint data for the preset risk and the complaint data submitted by the user for the preset risk can be provided as training samples to the risk prevention and control model for the preset risk ( Figure 3in the business server) to train the risk prevention and control model with the training samples; or, the first feature and the third feature can be provided as training samples to the risk prevention and control model for a preset risk to train the risk prevention and control model with the training samples, so as to provide more accurate qualitative data assets and modeling samples for the risk prevention and control model, and provide more accurate risk levels and risk case quantification indicators.

[0061] An embodiment of this specification provides a method for processing data. By obtaining the first feature required for constructing a trial qualitative strategy for a preset risk, and obtaining the basic features corresponding to the second feature, a trial qualitative strategy for the preset risk is constructed based on the first feature and the basic features. The second feature is a feature with a complexity higher than a preset complexity threshold. Then, complaint data submitted by a user for the preset risk can be obtained. If there are features in the third feature corresponding to the complaint data that are not included in the first feature and / or the basic features, the third feature is retroactively processed based on the basic features to obtain the retroactive information corresponding to the third feature. The complaint data is processed for trial qualification based on the retroactive information corresponding to the third feature, the first feature, and the basic features to obtain a corresponding first trial result. If the first trial result is better than the second trial result obtained by processing the complaint data for trial qualification based on the trial qualitative strategy, the trial qualitative strategy is updated based on the retroactive information corresponding to the third feature and the first feature. In this way, through the sorting of core features (i.e., the first feature and the basic features), the retroactive processing of new or changed features, and the retroactive calculation of the qualitative strategy, it can be ensured that the features used during the retroactive case qualification are real, valid, and in chronological order, and the phenomenon of feature crossing can be effectively avoided. It can overcome the situation where the retroactive result of the risk event trial qualification is inconsistent with the expected result of the current trial qualitative strategy, and through an objective, real, and accurate event retroactive qualification method, it can provide more accurate, comprehensive, and objective qualitative data assets, modeling training samples, etc. for subsequent risk prevention and control systems, and can also provide more objective and scientific risk levels and risk event quantification indicators for each risk prevention and control system, so that the trial qualification can fully play its due role in risk control and become a better security infrastructure.

[0062] Embodiment III

[0063] The above is the data processing method provided by the embodiments of this specification. Based on the same idea, the embodiments of this specification also provide a data processing device, as Figure 4 shown.

[0064] The data processing device includes: a strategy construction module 401, a data acquisition module 402, a trial module 403, and a strategy update module 404, where:

[0065] The policy construction module 401 obtains the first features required for constructing a trial qualification policy for a preset risk, and obtains the basic features corresponding to the second features. Based on the first features and the basic features, it constructs a trial qualification policy for the preset risk, where the second features are features with a complexity higher than a preset complexity threshold.

[0066] The data acquisition module 402 acquires the complaint data submitted by the user for the preset risk.

[0067] The trial module 403, if there are features in the third features corresponding to the complaint data that are not included in the first features and / or the basic features, performs a backtracking process on the third features based on the basic features to obtain the backtracking information corresponding to the third features, and performs a trial qualification process on the complaint data based on the backtracking information corresponding to the third features, the first features, and the basic features to obtain a corresponding first trial result.

[0068] The policy update module 404, if the first trial result is better than the second trial result obtained by performing a trial qualification process on the complaint data based on the trial qualification policy, updates the trial qualification policy based on the backtracking information corresponding to the third features and the first features.

[0069] In the embodiments of this specification, the policy construction module 401 includes:

[0070] The historical complaint acquisition unit acquires historical complaint data for a preset risk.

[0071] The feature acquisition unit determines the first features required for constructing a trial qualification policy for the preset risk based on the historical complaint data for the preset risk.

[0072] In the embodiments of this specification, the apparatus further includes:

[0073] The comparative analysis module analyzes the first trial result and the second trial result through an in-out analysis method to obtain a corresponding analysis result.

[0074] The policy update module 404, if the analysis result indicates that the first trial result is better than the second trial result, updates the trial qualification policy based on the backtracking information corresponding to the third features and the first features.

[0075] In the embodiments of this specification, the apparatus further includes:

[0076] The eigenvalue determination module calculates the first features and the basic features respectively to obtain the eigenvalue corresponding to the first features and the eigenvalue corresponding to the basic features.

[0077] A storage module that stores the eigenvalue corresponding to the first feature and the eigenvalue corresponding to the basic feature;

[0078] The trial module 403 performs a trial and qualitative process on the complaint data based on the backtracking information corresponding to the third feature, the eigenvalue corresponding to the first feature, and the eigenvalue corresponding to the basic feature, and obtains a corresponding first trial result.

[0079] In the embodiments of this specification, the device further includes:

[0080] A first model training module that provides historical complaint data for a preset risk and complaint data submitted by a user for the preset risk as training samples to a risk prevention and control model for the preset risk, so as to perform model training on the risk prevention and control model through the training samples; or,

[0081] A second model training module that provides the first feature and the third feature as training samples to a risk prevention and control model for the preset risk, so as to perform model training on the risk prevention and control model through the training samples.

[0082] In the embodiments of this specification, the preset risk includes one or more of the risks of theft, fraud, and illegal financial activities.

[0083] In the embodiments of this specification, the trial and qualitative strategy for the preset risk includes one or more of the content description information of the complaint data, the evidentiary image, the complaint record of the party being complained against, the transaction information of the party being complained against, and the transaction information between the complainant and the party being complained against.

[0084] The embodiments of this specification provide a data processing device. By obtaining the first features required for constructing a trial qualitative strategy for a preset risk, and obtaining the basic features corresponding to the second features, a trial qualitative strategy for the preset risk is constructed based on the first features and the basic features. The second features are features with a complexity higher than a preset complexity threshold. Then, complaint data submitted by a user for the preset risk can be obtained. If there are features in the third features corresponding to the complaint data that are not included in the first features and / or the basic features, the third features are retroactively processed based on the basic features to obtain the retroactive information corresponding to the third features. The complaint data is subject to trial qualitative processing based on the retroactive information corresponding to the third features, the first features, and the basic features to obtain a corresponding first trial result. If the first trial result is better than the second trial result obtained by performing trial qualitative processing on the complaint data based on the trial qualitative strategy, the trial qualitative strategy is updated based on the retroactive information corresponding to the third features and the first features. In this way, through the sorting of core features (i.e., the first features and the basic features), the retroactive calculation of new features or changed features, and the retroactive calculation of the qualitative strategy, it can be ensured that the features used during the case qualitative retroactive calculation are true and effective, conform to the chronological order, and can effectively avoid the occurrence of feature crossing. It can overcome the situation where the retroactive result of the risk event trial qualitative is inconsistent with the expected result of the current trial qualitative strategy. And through an objective, true, and accurate event qualitative retroactive method, it can provide more accurate, comprehensive, and objective qualitative data assets, modeling training samples, etc. for subsequent risk prevention and control systems, and can also provide more objective and scientific risk levels and risk event quantification indicators for each risk prevention and control system, so that the trial qualitative can fully play its due role in risk control and become a better security infrastructure.

[0085] Embodiment 4

[0086] The above is the data processing device provided by the embodiments of this specification. Based on the same idea, the embodiments of this specification also provide a data processing device, as Figure 5 shown.

[0087] The data processing device may be a terminal device or a server provided in the above embodiments, etc.

[0088] The data processing device can vary significantly due to different configurations or performances, and may include one or more processors 501 and a memory 502. One or more applications or data can be stored in the memory 502. Among them, the memory 502 can be transient storage or persistent storage. The applications stored in the memory 502 can include one or more modules (not shown in the figure), and each module can include a series of computer-executable instructions in the data processing device. Further, the processor 501 can be set to communicate with the memory 502 and execute a series of computer-executable instructions in the memory 502 on the data processing device. The data processing device can also include one or more power supplies 503, one or more wired or wireless network interfaces 504, one or more input / output interfaces 505, and one or more keyboards 506.

[0089] Specifically, in this embodiment, the data processing device includes a memory and one or more programs. One or more of the programs are stored in the memory, and one or more of the programs can include one or more modules. Each module can include a series of computer-executable instructions in the data processing device and is configured to be executed by one or more processors. The one or more programs include computer-executable instructions for performing the following:

[0090] Obtain the first features required to construct a trial qualitative strategy for a preset risk, and obtain the basic features corresponding to the second features. Based on the first features and the basic features, construct a trial qualitative strategy for the preset risk, where the second features are features with a complexity higher than a preset complexity threshold;

[0091] Obtain the complaint data submitted by the user for the preset risk;

[0092] If there are features in the third features corresponding to the complaint data that are not included in the first features and / or the basic features, then perform a backtracking process on the third features based on the basic features to obtain the backtracking information corresponding to the third features. Based on the backtracking information corresponding to the third features, the first features, and the basic features, perform a trial qualitative process on the complaint data to obtain a corresponding first trial result;

[0093] If the first trial result is better than the second trial result obtained by performing a trial qualitative process on the complaint data based on the trial qualitative strategy, then update the trial qualitative strategy based on the backtracking information corresponding to the third features and the first features.

[0094] In the embodiments of this specification, obtaining the first features required for constructing the trial qualification strategy for the preset risk includes:

[0095] Obtaining historical complaint data for the preset risk;

[0096] Determining the first features required for constructing the trial qualification strategy for the preset risk based on the historical complaint data for the preset risk.

[0097] In the embodiments of this specification, it further includes:

[0098] Analyzing the first trial result and the second trial result through the in-out analysis method to obtain corresponding analysis results;

[0099] If the first trial result is better than the second trial result obtained by qualitatively processing the complaint data based on the trial qualification strategy, then updating the trial qualification strategy based on the backtracking information corresponding to the third feature and the first feature includes:

[0100] If the analysis result indicates that the first trial result is better than the second trial result, then updating the trial qualification strategy based on the backtracking information corresponding to the third feature and the first feature.

[0101] In the embodiments of this specification, it further includes:

[0102] Calculating the first feature and the basic feature respectively to obtain the feature value corresponding to the first feature and the feature value corresponding to the basic feature;

[0103] Storing the feature value corresponding to the first feature and the feature value corresponding to the basic feature;

[0104] Qualitatively processing the complaint data based on the backtracking information corresponding to the third feature, the first feature and the basic feature to obtain a corresponding first trial result includes:

[0105] Qualitatively processing the complaint data based on the backtracking information corresponding to the third feature, the feature value corresponding to the first feature and the feature value corresponding to the basic feature to obtain a corresponding first trial result.

[0106] In the embodiments of this specification, it further includes:

[0107] Providing the historical complaint data for the preset risk and the complaint data submitted by the user for the preset risk as training samples to the risk prevention and control model for the preset risk to train the risk prevention and control model through the training samples; or,

[0108] Provide the first feature and the third feature as training samples to a risk prevention and control model for the preset risk, so as to train the risk prevention and control model with the training samples.

[0109] In the embodiments of the present specification, the preset risk includes one or more of theft risk, fraud risk, and illegal financial activities.

[0110] In the embodiments of the present specification, the trial and determination strategy for the preset risk includes one or more of the content description information of the complaint data, the evidentiary image, the complaint record of the party being complained, the transaction information of the party being complained, and the transaction information between the complainant and the party being complained.

[0111] The embodiments of the present specification provide a data processing device. By obtaining the first feature required to construct the trial and determination strategy for the preset risk, and obtaining the basic feature corresponding to the second feature, constructing the trial and determination strategy for the preset risk based on the first feature and the basic feature, where the second feature is a feature with a complexity higher than the preset complexity threshold. Then, it is possible to obtain the complaint data submitted by the user for the preset risk. If there are features in the third feature corresponding to the complaint data that are not included in the first feature and / or the basic feature, then perform a backtracking process on the third feature based on the basic feature to obtain the backtracking information corresponding to the third feature, and perform a trial and determination process on the complaint data based on the backtracking information corresponding to the third feature, the first feature, and the basic feature to obtain the corresponding first trial result. If the first trial result is better than the second trial result obtained by performing a trial and determination process on the complaint data based on the trial and determination strategy, then update the trial and determination strategy based on the backtracking information corresponding to the third feature and the first feature. In this way, through the sorting of the core features (i.e., the first feature and the basic feature), the backtracking of new features or changed features, and the backtracking calculation of the determination strategy, it can be ensured that the features used in the case determination backtracking are true and valid, conform to the chronological order, and can effectively avoid the occurrence of feature crossing. It can overcome the situation where the backtracking result of the risk event trial and determination is inconsistent with the expected result of the current trial and determination strategy, and can provide more accurate, comprehensive, and objective qualitative data assets, modeling training samples, etc. for subsequent risk prevention and control systems through an objective, true, and accurate event determination backtracking method, and can also provide more objective and scientific risk levels and risk event quantification indicators for each risk prevention and control system, so as to enable the trial and determination to fully play its due role in risk control and become a better security infrastructure.

[0112] Embodiment Five

[0113] Further, based on the above Figures 1 to 3For the method shown, one or more embodiments of this specification also provide a storage medium for storing computer-executable instruction information. In a specific embodiment, the storage medium can be a USB flash drive, an optical disc, a hard disk, etc. When the computer-executable instruction information stored in the storage medium is executed by a processor, the following processes can be realized:

[0114] Obtain the first features required for constructing a trial qualitative strategy for a preset risk, and obtain the basic features corresponding to the second features. Based on the first features and the basic features, construct a trial qualitative strategy for the preset risk, where the second features are features with a complexity higher than a preset complexity threshold;

[0115] Obtain the complaint data submitted by the user for the preset risk;

[0116] If there are features in the third features corresponding to the complaint data that are not included in the first features and / or the basic features, then perform a backtracking process on the third features based on the basic features to obtain the backtracking information corresponding to the third features. Based on the backtracking information corresponding to the third features, the first features, and the basic features, perform a trial qualitative process on the complaint data to obtain a corresponding first trial result;

[0117] If the first trial result is better than the second trial result obtained by performing a trial qualitative process on the complaint data based on the trial qualitative strategy, then update the trial qualitative strategy based on the backtracking information corresponding to the third features and the first features.

[0118] In the embodiments of this specification, the obtaining of the first features required for constructing a trial qualitative strategy for the preset risk includes:

[0119] Obtain historical complaint data for the preset risk;

[0120] Based on the historical complaint data for the preset risk, determine the first features required for constructing a trial qualitative strategy for the preset risk.

[0121] In the embodiments of this specification, it further includes:

[0122] Analyze the first trial result and the second trial result through a swap-in and swap-out analysis method to obtain corresponding analysis results;

[0123] The step of if the first trial result is better than the second trial result obtained by performing a trial qualitative process on the complaint data based on the trial qualitative strategy, then update the trial qualitative strategy based on the backtracking information corresponding to the third features and the first features includes:

[0124] If the analysis result indicates that the first trial result is superior to the second trial result, update the trial qualification strategy based on the backtracking information corresponding to the third feature and the first feature.

[0125] In the embodiments of this specification, it further includes:

[0126] Calculate the eigenvalue corresponding to the first feature and the eigenvalue corresponding to the basic feature respectively for the first feature and the basic feature;

[0127] Store the eigenvalue corresponding to the first feature and the eigenvalue corresponding to the basic feature;

[0128] The trial qualification process of the complaint data based on the backtracking information corresponding to the third feature, the first feature, and the basic feature to obtain the corresponding first trial result includes:

[0129] Perform trial qualification processing on the complaint data based on the backtracking information corresponding to the third feature, the eigenvalue corresponding to the first feature, and the eigenvalue corresponding to the basic feature to obtain the corresponding first trial result.

[0130] In the embodiments of this specification, it further includes:

[0131] Provide the historical complaint data for the preset risk and the complaint data submitted by the user for the preset risk as training samples to the risk prevention and control model for the preset risk to train the risk prevention and control model through the training samples; or,

[0132] Provide the first feature and the third feature as training samples to the risk prevention and control model for the preset risk to train the risk prevention and control model through the training samples.

[0133] In the embodiments of this specification, the preset risk includes one or more of theft risk, fraud risk, and illegal financial activities.

[0134] In the embodiments of this specification, the trial qualification strategy for the preset risk includes one or more of the content description information of the complaint data, the evidentiary image, the complaint record of the party being complained against, the transaction information of the party being complained against, and the transaction information between the complainant and the party being complained against.

[0135] An embodiment of this specification provides a storage medium. By obtaining the first features required for constructing a trial qualitative strategy for a preset risk, and obtaining the basic features corresponding to the second features, a trial qualitative strategy for the preset risk is constructed based on the first features and the basic features. The second features are features with a complexity higher than a preset complexity threshold. Then, complaint data submitted by a user for the preset risk can be obtained. If there are features in the third features corresponding to the complaint data that are not included in the first features and / or the basic features, the third features are retroactively processed based on the basic features to obtain the retroactive information corresponding to the third features. The complaint data is processed for trial qualification based on the retroactive information corresponding to the third features, the first features, and the basic features, and a corresponding first trial result is obtained. If the first trial result is better than the second trial result obtained by processing the complaint data for trial qualification based on the trial qualitative strategy, the trial qualitative strategy is updated based on the retroactive information corresponding to the third features and the first features. In this way, through the sorting of core features (i.e., the first features and the basic features), the retroactive processing of new or changed features, and the retroactive calculation of the qualitative strategy, it can be ensured that the features used in the case qualitative retroactive process are true and valid, conform to the chronological order, and can effectively avoid the occurrence of feature crossing. It can overcome the situation where the qualitative retroactive result of a risk event is inconsistent with the expected result of the current trial qualitative strategy, and can provide more accurate, comprehensive, and objective qualitative data assets, modeling training samples, etc. for subsequent risk prevention and control systems through an objective, true, and accurate event qualitative retroactive method. It can also provide more objective and scientific risk levels and risk event quantification indicators for each risk prevention and control system, so that trial qualification can fully play its due role in risk control and become a better security infrastructure.

[0136] The above describes specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the figures do not necessarily require the particular order or sequential order shown to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0137] In the 1990s, it was quite obvious to distinguish whether an improvement to a technology was an improvement in hardware (e.g., improvement to circuit structures such as diodes, transistors, switches, etc.) or an improvement in software (improvement to method flows). However, with the development of technology, many improvements to method flows today can be regarded as direct improvements to hardware circuit structures. Almost all designers obtain the corresponding hardware circuit structures by programming the improved method flows into the hardware circuits. Therefore, it cannot be said that an improvement to a method flow cannot be implemented with a hardware entity module. For example, a Programmable Logic Device (PLD) (e.g., a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user's programming of the device. The designer can program by himself / herself to "integrate" a digital system onto a piece of PLD without having to ask a chip manufacturer to design and fabricate a dedicated integrated circuit chip. Moreover, nowadays, instead of manually fabricating integrated circuit chips, this programming is mostly implemented using "logic compiler" software, which is similar to the software compiler used in program development and writing. The original code before compilation also has to be written in a specific programming language, which is called a Hardware Description Language (HDL). There is not only one type of HDL, but many types, 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, RHDL (Ruby Hardware Description Language), etc. The most commonly used ones currently are VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also be aware that by simply performing a little logical programming on the method flow with the above-mentioned several hardware description languages and programming it into an integrated circuit, it is easy to obtain the hardware circuit that implements the logical method flow.

[0138] The controller can be implemented in any suitable manner. For example, the controller can take the form of, for example, a microprocessor or a processor and a computer-readable medium storing computer-readable program code (such as software or firmware) executable by the (micro)processor, logic gates, switches, an application specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller. Examples of the controller include, but are not limited to, the following microcontrollers: ARC625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320. The memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art also know that, in addition to implementing the controller in the form of pure computer-readable program code, it is entirely possible to logically program the method steps to enable the controller to be implemented in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers, etc. to achieve the same function. Therefore, such a controller can be considered a hardware component, and the devices included therein for implementing various functions can also be regarded as the structures within the hardware component. Or even, the devices for implementing various functions can be regarded as either software modules for implementing the method or structures within the hardware component.

[0139] The systems, devices, modules, or units illustrated in the above embodiments can be specifically implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0140] For the convenience of description, when describing the above devices, they are described separately as various units according to their functions. Of course, when implementing one or more embodiments of this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0141] Those skilled in the art should understand that the embodiments of this specification can be provided as a method, a system, or a computer program product. Therefore, one or more embodiments of this specification can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, one or more embodiments of this specification can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program code.

[0142] Embodiments of this specification are described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of this specification. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices produce means for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0143] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to operate in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including instruction means that implement the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0144] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operational steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0145] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0146] The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM), and / or non-volatile memory such as read-only memory (ROM) or flash memory (flash RAM). The memory is an example of computer-readable media.

[0147] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules 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 technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0148] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0149] It should be understood by those skilled in the art that the embodiments of this specification may be provided as methods, systems or computer program products. Therefore, one or more embodiments of this specification may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, one or more embodiments of this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0150] One or more embodiments of the present specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. One or more embodiments of the present specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0151] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other, and the differences between each embodiment and other embodiments are emphasized. In particular, for the system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and reference can be made to the relevant parts of the method embodiments for the relevant content.

[0152] The above description is only for the embodiments of this specification and is not intended to limit this application. For those skilled in the art, various changes and modifications can be made to this specification. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this specification shall be included within the scope of the claims of this specification.

Claims

1. A method for processing data, the method comprising: Obtaining a first feature required for constructing a trial qualitative strategy for a preset risk, and obtaining a basic feature corresponding to a second feature, and constructing a trial qualitative strategy for the preset risk based on the first feature and the basic feature, where the second feature is a feature with a complexity higher than a preset complexity threshold; Obtaining complaint data submitted by a user for the preset risk; If there are features in a third feature corresponding to the complaint data that are not included in the first feature and / or the basic feature, then performing a backtracking process on the third feature based on the basic feature to obtain backtracking information corresponding to the third feature, and performing a trial qualitative process on the complaint data based on the backtracking information corresponding to the third feature, the first feature, and the basic feature to obtain a corresponding first trial result; If the first trial result is better than a second trial result obtained by performing a trial qualitative process on the complaint data based on the trial qualitative strategy, then updating the trial qualitative strategy based on the backtracking information corresponding to the third feature and the first feature.

2. The method according to claim 1, wherein the obtaining the first feature required for constructing a trial qualitative strategy for the preset risk comprises: Obtaining historical complaint data for a preset risk; Determining a first feature required for constructing a trial qualitative strategy for the preset risk based on the historical complaint data for the preset risk.

3. The method according to claim 1, the method further comprising: Analyzing the first trial result and the second trial result by means of swap-in and swap-out analysis to obtain corresponding analysis results; The if the first trial result is better than a second trial result obtained by performing a trial qualitative process on the complaint data based on the trial qualitative strategy, then updating the trial qualitative strategy based on the backtracking information corresponding to the third feature and the first feature, comprises: If the analysis result indicates that the first trial result is better than the second trial result, then updating the trial qualitative strategy based on the backtracking information corresponding to the third feature and the first feature.

4. The method according to claim 1, the method further comprising: Calculating the first feature and the basic feature respectively to obtain a feature value corresponding to the first feature and a feature value corresponding to the basic feature; Storing the feature value corresponding to the first feature and the feature value corresponding to the basic feature; The performing a trial qualitative process on the complaint data based on the backtracking information corresponding to the third feature, the first feature, and the basic feature to obtain a corresponding first trial result, comprises: Performing a trial qualitative process on the complaint data based on the backtracking information corresponding to the third feature, the feature value corresponding to the first feature, and the feature value corresponding to the basic feature to obtain a corresponding first trial result.

5. The method according to claim 2, the method further comprising: Provide historical complaint data against a preset risk and complaint data submitted by a user against the preset risk as training samples to a risk prevention and control model for the preset risk, so as to train the risk prevention and control model through the training samples; Or, Provide the first feature and the third feature as training samples to a risk prevention and control model for the preset risk, so as to train the risk prevention and control model through the training samples.

6. The method according to any one of claims 1-5, wherein the preset risk includes one or more of embezzlement risk, fraud risk, and illegal financial activities.

7. The method according to claim 6, wherein the trial and determination strategy for the preset risk includes one or more of the content description information of the complaint data, the evidentiary image, the complaint record of the party being complained against, the transaction information of the party being complained against, and the transaction information between the complainant and the party being complained against.

8. A data processing device, the device comprising: A strategy construction module, which obtains the first feature required for constructing a trial and determination strategy for a preset risk, and obtains the basic features corresponding to the second feature, and constructs a trial and determination strategy for the preset risk based on the first feature and the basic features, where the second feature is a feature with a complexity higher than a preset complexity threshold; A data acquisition module, which acquires complaint data submitted by a user against the preset risk; A trial module, if there are features in the third feature corresponding to the complaint data that are not included in the first feature and / or the basic features, then perform backtracking processing on the third feature based on the basic features to obtain the backtracking information corresponding to the third feature, and perform trial and determination processing on the complaint data based on the backtracking information corresponding to the third feature, the first feature, and the basic features to obtain a corresponding first trial result; A strategy update module, if the first trial result is better than the second trial result obtained by performing trial and determination processing on the complaint data based on the trial and determination strategy, then update the trial and determination strategy based on the backtracking information corresponding to the third feature and the first feature.

9. A data processing device, the data processing device comprising: A processor; And A memory arranged to store computer-executable instructions, the executable instructions when executed causing the processor to: Obtain the first feature required for constructing a trial and determination strategy for a preset risk, and obtain the basic features corresponding to the second feature, and construct a trial and determination strategy for the preset risk based on the first feature and the basic features, where the second feature is a feature with a complexity higher than a preset complexity threshold; Obtain complaint data submitted by a user against the preset risk; If there are features in the third feature corresponding to the complaint data that are not included in the first feature and / or the basic feature, backtracking processing is performed on the third feature based on the basic feature to obtain the backtracking information corresponding to the third feature, and the complaint data is processed for trial and qualitative determination based on the backtracking information corresponding to the third feature, the first feature, and the basic feature to obtain a corresponding first trial result; If the first trial result is better than the second trial result obtained by processing the complaint data for trial and qualitative determination based on the trial and qualitative determination strategy, the trial and qualitative determination strategy is updated based on the backtracking information corresponding to the third feature and the first feature.

10. A storage medium for storing computer-executable instructions, the executable instructions, when executed by a processor, implement the following process: Obtain the first feature required to construct a trial and qualitative determination strategy for a preset risk, and obtain the basic feature corresponding to the second feature, and construct a trial and qualitative determination strategy for the preset risk based on the first feature and the basic feature, where the second feature is a feature with a complexity higher than a preset complexity threshold; Obtain complaint data submitted by a user for the preset risk; If there are features in the third feature corresponding to the complaint data that are not included in the first feature and / or the basic feature, backtracking processing is performed on the third feature based on the basic feature to obtain the backtracking information corresponding to the third feature, and the complaint data is processed for trial and qualitative determination based on the backtracking information corresponding to the third feature, the first feature, and the basic feature to obtain a corresponding first trial result; If the first trial result is better than the second trial result obtained by processing the complaint data for trial and qualitative determination based on the trial and qualitative determination strategy, the trial and qualitative determination strategy is updated based on the backtracking information corresponding to the third feature and the first feature.

Images