Risk assessment processing method and device for service scene

Abstract

The embodiment of the invention provides a risk assessment processing method and device for a service scene, and the method comprises the steps: carrying out the risk assessment processing of the service scene in a process of carrying out the risk assessment processing of the service scene, the method comprises the following steps: firstly, obtaining asset association data and credit data of a service user according to a data authorization certificate of the service user synchronized by a service system, and carrying out asset association risk calculation based on the asset association data to obtain an asset association risk score of the service user; service risk assessment is carried out according to the asset association risk score, the asset association data and the credit data, service risk labels of the service users are obtained, risk matching processing is carried out based on the service risk labels, the asset association data and the credit data of the service users included in the user pairing, and a risk matching result of the user pairing is obtained. Therefore, the risk matching of the user pairing for the service user in the service scene is realized based on the credit data.

Description

Technical Field

[0001] This document relates to the field of data processing technology, and in particular to a risk assessment processing method and apparatus for a service scenario. Background Technology

[0002] With the continuous development of internet technology, various online service platforms have emerged, such as those for social networking and dating, providing users with diversified communication channels and methods. As the social credit system continues to advance, users' credit information in various life scenarios is becoming increasingly abundant. Against this backdrop, in relevant service scenarios, users not only pay attention to the other party's basic personal information but also gradually value their performance in terms of creditworthiness and asset status. Furthermore, as the user base of service platforms continues to expand, the complexity of service scenarios is gradually increasing, and the demand for risk identification and assessment in related services is becoming increasingly prominent. Summary of the Invention

[0003] This specification provides one or more embodiments of a risk assessment method for a service scenario, comprising: obtaining asset-related data and credit data of a service user based on the data authorization credentials of the service user synchronized by the service system; calculating asset-related risk based on the asset-related data to obtain an asset-related risk score for the service user; performing a service risk assessment based on the asset-related risk score, the asset-related data, and the credit data to obtain a service risk label for the service user; and performing risk matching processing based on the service risk labels, asset-related data, and credit data of each service user included in the user pairing to obtain a risk matching result for the user pairing.

[0004] This specification provides one or more embodiments of a risk assessment processing apparatus for a service scenario, comprising: a data acquisition module configured to acquire asset-related data and credit data of a service user based on data authorization credentials synchronized by a service system; a scoring calculation module configured to calculate asset-related risk based on the asset-related data to obtain an asset-related risk score for the service user; a risk assessment module configured to perform a service risk assessment based on the asset-related risk score, the asset-related data, and the credit data to obtain a service risk label for the service user; and a risk matching processing module configured to perform risk matching processing based on the service risk labels, asset-related data, and credit data of each service user included in a user pairing to obtain a risk matching result for the user pairing.

[0005] This specification provides one or more embodiments of a risk assessment processing device for a service scenario, comprising: a processor; and a memory configured to store computer-executable instructions, which, when executed, cause the processor to: obtain asset-related data and credit data of a service user based on data authorization credentials synchronized with a service system; calculate asset-related risk based on the asset-related data to obtain an asset-related risk score for the service user; conduct a service risk assessment based on the asset-related risk score, the asset-related data, and the credit data to obtain a service risk label for the service user; and perform risk matching processing based on the service risk labels, asset-related data, and credit data of each service user included in a user pairing to obtain a risk matching result for the user pairing.

[0006] This specification provides one or more embodiments of a computer-readable storage medium for storing computer-executable instructions. When executed, these instructions perform the following process: Obtaining the asset-related data and credit data of a service user based on the service user's data authorization credentials synchronized by the service system; calculating the asset-related risk score of the service user based on the asset-related data; conducting a service risk assessment based on the asset-related risk score, the asset-related data, and the credit data to obtain a service risk label for the service user; and performing risk matching processing based on the service risk labels, asset-related data, and credit data of each service user included in the user pairing to obtain the risk matching result of the user pairing. Attached Figure Description

[0007] To more clearly illustrate the technical solutions in one or more embodiments of this specification or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Figure 1 A schematic diagram of the implementation environment for a risk assessment and processing method for a service scenario provided in one or more embodiments of this specification; Figure 2 A flowchart illustrating a risk assessment and processing method for a service scenario provided in one or more embodiments of this specification; Figure 3 A flowchart illustrating a risk assessment and processing method for a service scenario applied to mate selection, provided by one or more embodiments of this specification; Figure 4 A schematic diagram of an embodiment of a risk assessment processing device for a service scenario provided by one or more embodiments of this specification; Figure 5 This is a schematic diagram of the structure of a risk assessment processing device for a service scenario provided in one or more embodiments of this specification. Detailed Implementation

[0008] To enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of this document.

[0009] The risk assessment and handling methods for service scenarios provided in one or more embodiments of this specification are applicable to the service system implementation environment. Figure 1 The implementation environment includes at least: Service System 101, Credit Information Platform 102; The service system 101 is used to provide the data authorization interface of the credit reporting platform 102 to enable service users to authorize data and perform risk assessment processing; the credit reporting platform 102 is used to respond to the interface call of the service system 101 to perform data authorization processing for service users; In addition, the implementation environment may also include a credit reporting node 103, a trusted data space 104, an isolated data space 105, and a user terminal 106 serving the user. Credit node 103 can apply to trusted data space 104 for isolated data space 105 to conduct risk assessment processing of service scenarios within the applied isolated data space 105; trusted data space 104 refers to an independent security area built with hardware, and isolated data space 105 refers to an isolated environment at the software level used to process sensitive data, which can open data usage permissions under the control of data permissions to ensure that data is not leaked or abused, such as a data sandbox; The user terminal 106 is used to cooperate with the service system 101 to perform risk assessment processing for the service scenario; the user terminal 106 can be a mobile phone, personal computer, tablet computer, e-book reader, device for information interaction based on VR (Virtual Reality) and AR (Augmented Reality), vehicle terminal, IoT device, wearable smart device, laptop computer and desktop computer, etc.

[0010] In this implementation environment, during the risk assessment process for service scenarios, service system 101 first obtains the asset association data and credit data of service users based on their data authorization credentials. Then, it calculates asset association risk based on the asset association data to obtain the asset association risk score of the service users. Further, it conducts service risk assessment based on the asset association risk score, asset association data, and credit data to obtain the service risk label of the service users. Based on the service risk labels, asset association data, and credit data of each service user included in the user pairing, it performs risk matching processing to obtain the risk matching result of the user pairing. In this way, risk matching of user pairing in the service scenario is achieved starting from credit data.

[0011] It should be noted that, considering that the asset-related data, credit data, asset-related risk scores, service risk labels, and other related data involved in this specification may, to some extent, belong to the privacy of service users, authorization from service users can be obtained before acquiring or processing such data to ensure that the data collection operation complies with relevant data management regulations. For example, authorization can be granted by generating a data authorization certificate and performing authorization processing on the data authorization certificate, or other methods can be used for authorization, which are not limited in this embodiment.

[0012] This specification provides one or more embodiments of a risk assessment and processing method for a service scenario, as follows: Reference Figure 2 The risk assessment and processing method for service scenarios provided in this embodiment specifically includes steps S202 to S208.

[0013] Step S202: Obtain the asset association data and credit data of the service user based on the data authorization certificate of the service user synchronized by the service system.

[0014] In this embodiment, the data authorization credential refers to a credential used by a service user to authorize data access to the service system. Specifically, it refers to a time-sensitive and unique data authorization credential generated by the service system or issued by the authorization management module after the service user authorizes data access within the service system. The data authorization credential may record the scope of authorization and / or the authorization period. Alternatively, the data authorization credential can also be obtained by the service user through the data authorization interface of the credit reporting platform provided by the service system. Optionally, the data authorization credential is obtained after the service user authorizes data access through the data authorization interface of the credit reporting platform provided by the service system.

[0015] In practice, to ensure compliance during data acquisition, the data authorization credentials of service users synchronized with the service system can be obtained first, and their validity verified. If verification is successful, the asset-related data and credit data of the service users can be obtained based on their data authorization credentials. Here, "service user" refers to a user accessing the service system, specifically a user accessing and / or processing services through the service system in a given service scenario. For example, in a dating scenario, a service user could be a dating-related user accessing and / or processing dating-related services through the dating system.

[0016] The asset-related data refers to data that has an asset-related relationship with the service user. Specifically, it refers to data reflecting the asset-related relationship between the service user and its associated users. For example, asset-related data can be kinship data reflecting the asset linkage between the service user and its immediate or close relatives. For instance, asset-related data may include information on the immediate relatives declared by the service user, joint real estate data between the service user and its immediate or close relatives, guarantee relationship data, asset transaction data, and / or joint account information. Optionally, asset-related data may include anonymized asset-related data after data de-identification.

[0017] The credit data refers to credit-related data, specifically the personal credit data of service users and / or credit-related data of service users. The personal credit data of service users can be credit indicators, such as credit scores and credit ratings, or credit-related behavioral data, such as loan data and repayment data, or performance data related to public sector performance records, such as property fee payment data and social security contribution base. The credit-related data of service users can be related data arising from credit behavior between service users and related users, such as credit guarantee related data or credit guarantee data between service users and related users. Optionally, the credit data includes de-identified credit data after data anonymization.

[0018] Here, when the asset-related data includes de-identified asset-related data after data anonymization and the credit data includes de-identified credit data after data anonymization, the above-mentioned acquisition of asset-related data and credit data based on the service user's data authorization certificate can also be replaced by: acquiring de-identified asset-related data and de-identified credit data based on the service user's data authorization certificate.

[0019] Step S204: Calculate the asset association risk based on the asset association data to obtain the asset association risk score of the service user.

[0020] In practice, after acquiring asset-related data, in order to transform the asset-related relationship between service users and associated users into quantifiable indicators, thereby extending the risk assessment dimension from the individual service user level to the level of the relationship between service users and associated users, asset-related risk calculation is performed based on the asset-related data to obtain the service user's asset-related risk score. In addition, asset-related risk calculation can also be performed based on anonymized asset-related data. In this case, asset-related risk calculation based on asset-related data can also be replaced by asset-related risk calculation based on anonymized asset-related data to obtain the service user's asset-related risk score.

[0021] In the specific implementation process, during the calculation of asset-related risks based on asset-related data, in order to improve the granularity of the asset-related risk calculation and provide data support for subsequent accurate risk matching, asset-related risk calculation can be performed based on the association scores and corresponding association weights of multiple asset-related dimensions. In one optional implementation method provided in this embodiment, the asset-related risk score of the service user is obtained by calculating asset-related risks based on asset-related data, including: Extract sub-data from multiple asset association dimensions contained in the asset association data, and calculate the association score for each asset association dimension based on the sub-data; The asset association risk score is calculated based on the association scores and corresponding association weights for each asset association dimension.

[0022] The asset association dimension refers to the dimension used to assess the asset association relationship between service users and associated users. For example, the asset association dimension may include the intensity of financial assistance, the guarantee risk dimension, and / or the degree of financial dependence dimension. Among them, the intensity of financial assistance is used to measure the continuity and intensity of financial assistance provided by the service user to the associated user, such as paying medical expenses for parents or regularly subsidizing the living expenses of relatives; the guarantee risk dimension is used to measure the joint repayment risk faced by the service user due to providing debt guarantees to external parties; the degree of financial dependence dimension is used to measure whether the service user depends on financial input from associated users to maintain daily expenses, such as receiving transfers from parents for a long time; in addition, the asset association dimension may also include other dimensions of asset association relationship, such as the frequency of fund transactions.

[0023] Specifically, in the process of calculating asset-related risk based on asset-related data, the first step is to extract sub-data from multiple asset-related dimensions contained in the asset-related data. After cleaning the extracted sub-data, the association score of each asset-related dimension is calculated based on the sub-data. With multiple association scores obtained and the association weight of each asset-related dimension determined, the asset-related risk score is calculated based on the association score and corresponding association weight of each asset-related dimension.

[0024] In the process of calculating the correlation scores of each asset correlation dimension based on sub-data, the correlation scores of the funding intensity dimension, the guarantee risk dimension, and / or the funding dependence dimension can be calculated. For example, the sum of the proportion of each funding amount transferred by the service user to the related user and the inflow of funds can be calculated, the sum of the product of the guarantee amount provided by the service user to the related user and the probability of guarantee delinquency can be calculated, and / or the ratio of the amount of funds transferred by the service user to the related user to the amount of funds spent by the service user can be calculated. In one optional implementation of this embodiment, the correlation score for each asset correlation dimension is calculated based on the sub-data, including: The funding intensity is calculated based on the transfer data and fund inflow data of service users transferring funds to associated users; the guarantee risk value is calculated based on the fund guarantee data and guarantee delinquency probability of service users; and / or, the funding dependence is calculated based on the fund inflow data and fund outflow data of service users receiving funds.

[0025] The term "related user" refers to a user who has a specific relationship with the service user. Specifically, related users can be users who have a specific social or economic relationship with the service user and may have a consequential impact on the service user's asset status and / or credit risk. For example, related users can include the service user's parents, spouse, children, and / or siblings. In addition, they can also include other direct relatives and / or close relatives who have an asset relationship with the service user. Furthermore, in the case where the service user is a user seeking a spouse, similarly, related users can also be users who have a relationship with the user seeking a spouse.

[0026] Specifically, in calculating the correlation score for the funding intensity dimension, the sum of the proportions of fund transfer sub-data and fund inflow sub-data from service users to related users can be calculated, and the sum of the calculated proportions can be used as the funding intensity; in calculating the correlation score for the guarantee risk dimension, the sum of the products of the service user's fund guarantee sub-data and the probability of guarantee delinquency can be calculated, and the sum of the calculated products can be used as the guarantee risk value; in calculating the correlation score for the funding dependence dimension, the ratio of fund inflow sub-data and fund outflow sub-data from service users can be calculated, and the calculated ratio can be used as the funding dependence.

[0027] For example, the funding intensity H can be calculated using the following formula: H = ∑ (transfer sub-data / fund inflow sub-data) Persistent weights The guarantee risk value S can be calculated using the following formula: S = ∑ (fund guarantee sub-data) (Probability of guarantee delinquency) The funding dependence ratio D can be calculated as follows: D = Fund inflow sub-data / Fund outflow sub-data; Based on this, the aforementioned asset-related risk score R can be calculated using the following formula: R=w1 H+w2 S+w3 D Where w1 represents the correlation weight corresponding to the funding intensity dimension, that is, the correlation weight corresponding to the funding intensity H; w2 represents the correlation weight corresponding to the guarantee risk dimension, that is, the correlation weight corresponding to the guarantee risk value S; and w3 represents the correlation weight corresponding to the funding dependence dimension, that is, the correlation weight corresponding to the funding dependence D.

[0028] It should be noted that the three sub-steps in the above-described implementation method for calculating the correlation scores of each asset correlation dimension based on sub-data can be selected for execution according to actual processing needs. For example, only the transfer sub-data and fund inflow sub-data of fund transfers from service users to associated users can be executed to calculate the funding intensity. In this case, the above-described implementation method for calculating asset correlation risk based on asset correlation data, which involves extracting sub-data of multiple asset correlation dimensions contained in the asset correlation data, calculating the correlation score of each asset correlation dimension based on the sub-data, and calculating the asset correlation risk score based on the correlation score and corresponding correlation weight of each asset correlation dimension, can also be replaced by: extracting sub-data of the funding intensity dimension contained in the asset correlation data, calculating the correlation score based on the sub-data, and calculating the asset correlation risk score based on the correlation score and corresponding correlation weight. For example, the calculation of funding intensity can be performed based solely on the transfer data and fund inflow data of service users transferring funds to associated users, and the calculation of guarantee risk value can be performed based on the service user's fund guarantee data and guarantee delinquency probability. In this case, the above-mentioned method of calculating asset-related risk based on asset-related data, which involves extracting sub-data of multiple asset-related dimensions contained in the asset-related data, calculating the correlation score of each asset-related dimension based on the sub-data, and calculating the asset-related risk score based on the correlation score of each asset-related dimension and the corresponding correlation weight, can be replaced by: extracting sub-data of multiple asset-related dimensions contained in the asset-related data, calculating the first correlation score of the funding intensity dimension and the second correlation score of the guarantee risk dimension based on the sub-data, and calculating the asset-related risk score based on the first correlation score, the second correlation score and the corresponding correlation weight. Furthermore, in the process of calculating the correlation scores of each asset correlation dimension based on sub-data, other implementation methods can be introduced according to actual processing needs, and can be combined with any one or more of the above-mentioned other implementation methods after adaptive modification, change or deletion. For example, calculating the correlation scores of each asset correlation dimension based on sub-data includes: calculating the financial burden intensity based on the service user's fund inflow sub-data and the number of related users who provide financial assistance; and calculating the guarantee risk value based on the service user's financial guarantee sub-data and the probability of guarantee delinquency. Here, the number of related users who provide financial assistance to the service user can be the number of the service user's immediate family members and / or close relatives who need to be supported and / or cared for.

[0029] Step S206: Based on the asset-related risk score, the asset-related data, and the credit data, conduct a service risk assessment to obtain the service risk label of the service user.

[0030] In practical implementation, to prevent service users from falling into risky relationships due to information asymmetry during the user matching process in the service system, and to improve the efficiency of transmitting risk information to service users, corresponding service risk labels can be generated based on each service user's asset-related risk score, asset-related data, and / or credit data. Accordingly, here, after obtaining the service user's asset-related risk score, asset-related data, and credit data, a service risk assessment is performed based on the asset-related risk score, asset-related data, and credit data to obtain the service user's service risk label.

[0031] The service risk label refers to a label used to characterize the risk characteristics of service users in a service scenario; for example, a service risk label can be a risk level label, an asset association label, or a service recommendation label; for example, a service risk label can be a mate selection risk label used to characterize the risk characteristics of mate selection users (service users) in a mate selection scenario. Among them, the risk level label refers to the label generated based on the risk level classification of service users to characterize the comprehensive risk level of service users. For example, the risk level can be divided into the first risk level, the second risk level, and the third risk level. The first risk level can indicate that the service user has a low asset liability burden and / or high credit stability, the second risk level can indicate that the service user has a moderate asset liability burden and / or moderate credit stability, and the third risk level can indicate that the service user has a high asset liability burden and / or low credit stability. Asset association tags can be tags used to characterize the asset association characteristics and / or potential asset association risks of service users. For example, asset association tags can be tags such as "with financial guarantees", "long-term financial support", and "no financial dependence". Service recommendation tags can be matching suggestion tags generated by combining asset association risk scores, asset association data and / or credit data, used to guide user matching and screening. Alternatively, service recommendation tags can also be recommendation tags used to recommend similar user groups. For example, service recommendation tags can be tags such as "avoid users with financial guarantees", "suggest clarifying guarantee liability", "suggest communicating family financial boundaries", or "same occupation" and "same region".

[0032] In the specific execution process, to improve processing accuracy and efficiency during the generation of service risk labels, a risk rating model can be introduced. For example, a pre-trained risk rating model specifically designed for risk rating processing can be used to combine the risk rating model with risk rating processing to obtain service risk labels. In the first optional implementation method provided in this embodiment, service risk assessment is performed based on asset-related risk scores, asset-related data, and credit data to obtain service risk labels for service users, including: The risk rating model is processed by inputting asset-related risk scores, asset-related data, and credit data to obtain risk level labels.

[0033] The risk rating model refers to a model that performs risk rating processing on relevant data of service users. The risk rating model can be an algorithmic model, such as a model based on a decision tree algorithm, a gradient boosting algorithm, a neural network algorithm, or other algorithms. The inputs to the risk rating model include asset-related risk scores, asset-related data, and / or credit data, and the output includes risk level labels. Optionally, the risk rating model includes a gradient boosting model for risk rating processing.

[0034] In this process, when calling the risk rating model for risk rating, the risk rating model can obtain complex correlations of multi-dimensional risk features through ensemble learning of multiple decision trees, thereby improving the accuracy of risk rating. In one optional implementation of this embodiment, the risk rating process includes: Feature extraction is performed on the input asset-related risk scores, asset-related data, and credit data, and structured features are constructed based on the extracted features; Structured features and baseline risk scores are input into multiple pre-trained decision trees to perform rule matching decisions to obtain risk scores, and risk scores are then mapped to risk levels to obtain risk level labels.

[0035] For example, in the risk rating model's risk rating process, the feature extraction module first standardizes the asset-related risk score, asset-related data, and credit data, converting them into quantitative features. It then performs feature filtering and extraction on multi-dimensional data. Next, the structuring module integrates and quantifies the extracted features according to a preset data structure, forming structured features. Then, the risk score fusion module uses the benchmark risk score as a fixed feature dimension and merges it with the structured features to form a complete feature set with a fused benchmark score. Next, in the multi-decision-tree rule matching module, each decision tree performs layer-by-layer rule matching on the input feature set based on its own trained splitting rules, outputting a local risk score for the corresponding dimension. Gradient boosting is used to weight and fuse the local risk scores of multiple decision trees, offsetting the bias and variance of a single decision tree, and outputting the risk score. Finally, the risk level mapping module compares the risk score with a threshold and maps it to the corresponding discrete risk level, obtaining and outputting the risk level label.

[0036] In the specific execution process, to improve the accuracy of risk feature identification and processing efficiency during the generation of asset-related tags, asset-related data can be input into a rule engine so that the rule engine can match based on preset rules to obtain asset-related tags. In the second optional implementation provided in this embodiment, service risk assessment is performed based on asset-related risk scores, asset-related data, and credit data to obtain service risk tags for service users, including: The asset association data is input into the rule engine to perform asset association detection and obtain asset association tags.

[0037] Specifically, during the asset association detection process, after receiving and preprocessing the asset association data, the rule engine can call the rule library to match the multi-dimensional sub-data in the asset association data with the corresponding rules in the rule library. The rule parser performs conditional judgment on each sub-data and triggers the generation of corresponding asset association tags when a rule is matched. It supports parallel matching of multiple rules and output of multiple tags. After that, the rule engine can perform tag deduplication and output the final asset association tags.

[0038] In the specific implementation process, during the generation of service risk labels for service users, service recommendation labels for service users can also be generated based on asset-related risk scores, asset-related data, and / or credit data. In the third optional implementation provided in this embodiment, service risk assessment is performed based on asset-related risk scores, asset-related data, and credit data to obtain service risk labels for service users, including: User pairing and recommendation detection is performed based on asset-related risk scores, asset-related data, and credit data to obtain pairing and recommendation tags.

[0039] In this process of generating matching recommendation tags for service users, a user matching recommendation detection model can be introduced to perform user matching recommendation detection. In one optional implementation of this embodiment, user matching recommendation detection is performed based on asset-related risk scores, asset-related data, and credit data, including: Input at least one of the following into the user pairing recommendation detection model: asset-related risk score, asset-related data, and credit data, to obtain pairing recommendation labels.

[0040] The user pairing recommendation detection model refers to the model used for user pairing recommendation detection. The user pairing recommendation detection model can be an algorithm model, such as a model implemented based on a neural network algorithm, or a model implemented based on other algorithms. The input of the user pairing recommendation detection model includes asset association risk score, asset association data and / or credit data, and the output includes pairing recommendation tags.

[0041] For example, in the case of a mate selection scenario, mate selection recommendation detection can be performed based on asset-related risk scores, asset-related data, and credit data to obtain mate selection recommendation labels. Specifically, in the process of mate selection recommendation detection, at least one of the asset-related risk scores, asset-related data, and credit data can be input into the mate selection recommendation detection model to obtain mate selection recommendation labels.

[0042] Specifically, in one optional implementation method provided in this embodiment, user pairing recommendation detection is implemented in the following way: The attribute matching engine retrieves similar user sets of service users from the service user database; The standardized features of service users and the standardized features of each user in the same category set are extracted using a feature extraction network. The feature fusion distribution is obtained by performing feature fusion calculation and feature distribution calculation on the standardized features of users of the same type through the feature distribution calculation module. The feature fusion module performs feature fusion calculations based on the standardized features of service users to obtain feature fusion values, and maps the feature fusion values ​​to the feature fusion distribution to obtain the recommendation percentiles of service users. The recommendation percentiles are then mapped to paired recommendation tags through the ranking mapping engine.

[0043] For example, in the process of user pairing recommendation detection, the user pairing recommendation detection model may include an attribute matching engine, a feature extraction network, a feature distribution calculation module, a feature fusion module, and / or a ranking mapping engine. First, the attribute matching engine performs targeted retrieval of the service user database based on preset coarse-grained matching rules, filtering users who do not meet the basic attribute matching conditions and obtaining a set of similar users. Then, the feature extraction network extracts features from the multi-dimensional data of the service users and each similar user in the set, and standardizes the extracted features, outputting standardized features for the service users and standardized features for each similar user. Next, the feature distribution calculation module performs fusion calculation and feature distribution calculation on the standardized features of similar users to obtain a feature fusion distribution. Further, the feature fusion module performs feature fusion calculation on the standardized features of the service users to obtain a feature fusion value, and maps this feature fusion value to the feature fusion distribution of similar users to determine the recommendation percentile of the service users. Finally, the ranking mapping engine compares the recommendation percentile with the rule threshold, matches the corresponding recommendation level, and generates standardized pairing recommendation tags.

[0044] Here, in the process of user pairing recommendation detection model, in addition to the user pairing recommendation detection model extracting standardized features of service users and standardized features of each user in the same category set through the feature extraction network, the user pairing recommendation detection model can also extract features from at least one of asset-related risk scores, asset-related data, and credit data through the feature extraction network and perform feature standardization to obtain standardized features. Based on this, the above-mentioned extraction of standardized features of service users and standardized features of each user in the same category set through the feature extraction network can be replaced by: extracting features from at least one of asset-related risk scores, asset-related data, and credit data through the feature extraction network and performing feature standardization to obtain standardized features, and forming a new implementation method with other processing sub-steps provided in this embodiment.

[0045] In addition, in the process of generating matching recommendation tags for service users, besides the above-mentioned implementation method of detecting user matching recommendations based on asset-related risk scores, asset-related data and / or credit data to obtain matching recommendation tags, feature extraction, feature standardization and feature fusion calculation can also be performed from asset-related risk scores, asset-related data and / or credit data, so as to determine the matching recommendation tags by the recommendation percentile determined based on the feature fusion values. In the fourth optional implementation provided in this embodiment, service risk assessment is performed based on asset-related risk scores, asset-related data, and credit data to obtain service risk labels for service users, including: Retrieve similar user sets of service users from the service user database, and construct a feature fusion distribution based on the standardized characteristics of each similar user in the similar user set; Feature extraction and standardization are performed on at least one of the asset-related risk scores, asset-related data, and credit data, and feature fusion calculation is performed on the standardized features of the obtained service users to obtain feature fusion values; The recommended quantiles are obtained by mapping the feature fusion values ​​to the feature fusion distribution, and the paired recommended labels are determined based on the recommended quantiles.

[0046] Among them, the set of similar users refers to the collection of users in the service user database who are similar to the service user in specific attributes. For example, similar users can be users who are similar to the service user in terms of asset-related risk, asset-related data and / or credit data, or similar users can also be users who are similar to the service user in terms of region, education and / or other attributes.

[0047] Specifically, in the process of generating paired recommendation tags for service users, the process begins by retrieving a set of similar users from the service user database to obtain at least one similar user, and then constructing a feature fusion distribution based on the standardized features of each similar user. Next, features are extracted and standardized from the service user's asset-related risk score, asset-related data, and / or credit data to obtain standardized features. Feature fusion calculations are then performed on the standardized features to obtain feature fusion values. Finally, the feature fusion values ​​are mapped to the feature fusion distribution to obtain recommendation quantiles, and paired recommendation tags for service users are determined based on these recommendation quantiles.

[0048] It should be noted that the above-mentioned implementation method for assessing service risk based on asset-related risk scores, asset-related data, and credit data to obtain paired risk labels for service users can be implemented using any of the first, second, and third implementation methods as needed. For example, the first, second, or third optional implementation method can be selected. Furthermore, in the specific implementation process, any two or all three implementation methods can be combined, or adapted, modified, changed, or deleted as needed before combining any two or all three implementation methods. Based on this, the above-mentioned method for assessing service risk based on asset-related risk scores, asset-related data, and credit data to obtain paired risk labels for service users can be implemented using any of the first, second, and third optional implementation methods as needed. The method of using asset-related data and credit data to assess service risk and obtain paired risk labels for service users can be replaced by: inputting asset-related risk scores, asset-related data, and credit data into a risk rating model for risk rating processing to obtain risk level labels; inputting asset-related data into a rule engine for asset-related detection to obtain asset-related labels; retrieving similar user sets of service users from the service user database and constructing a feature fusion distribution based on the standardized characteristics of each similar user in the similar user set; extracting and standardizing features from at least one of the asset-related risk scores, asset-related data, and credit data, and performing feature fusion calculations on the obtained standardized features of service users to obtain feature fusion values; mapping the feature fusion values ​​to the feature fusion distribution to obtain recommendation quantiles, and determining paired recommendation labels based on the recommendation quantiles.

[0049] Step S208: Perform risk matching processing based on the service risk tags, asset association data and credit data of each service user included in the user pairing to obtain the risk matching result of the user pairing.

[0050] In practical applications, service users can match users through the service system. During the matching process, service users can initiate matching independently or receive matching recommendations from the service system. In this process, when responding to matching initiated by service users or recommending matching to target service users, the service system can perform risk matching based on the service risk tags of each service user.

[0051] In practice, during the risk matching process based on each service user, in order to improve the comprehensiveness and accuracy of the risk matching process, the service system can start from the service risk tags of each service user included in the user pairing, and perform risk matching based on the service risk tags, asset association data and credit data of each service user to obtain the risk matching result of the user pairing. Here, during the risk matching process, the service system may perform risk matching based solely on service risk tags or solely on at least one of service risk tags and asset association data and / or credit data. Based on this, the above implementation method can also be replaced by: performing risk matching based on the service risk tags of each service user included in the service pairing to obtain the risk matching result of the user pairing; or, it can also be replaced by: performing risk matching based on the service risk tags of each service user included in the user pairing and asset association data and / or credit data to obtain the risk matching result of the user pairing, and forming a new implementation method with other processing steps provided in this embodiment.

[0052] For example, in the case of a mate selection scenario, after obtaining the risk tags, asset association data and / or credit data of the mate selection users, risk matching processing can be performed based on the mate selection risk tags, asset association data and / or credit data of each mate selection user included in the mate selection pairing to obtain the risk matching result of the mate selection pairing.

[0053] In the specific implementation process, during the risk matching process, risk compatibility analysis and asset liability symmetry calculation can be performed on each service user to achieve two-way risk matching for each service user. In the first optional implementation method provided in this embodiment, risk matching is performed based on the service risk tags, asset association data, and credit data of each service user included in the user pairing to obtain the risk matching result of the user pairing, including: Based on the risk tag combinations constructed from the service risk tags of each service user, determine the corresponding risk handling strategy for each pair recommendation level and implement the strategy; Based on the asset association data of each service user, perform asset symmetry calculations for user pairing, and issue risk warnings based on the obtained asset symmetry results.

[0054] Optionally, the risk tagging combination may include risk level tags, asset association tags, and / or pairing recommendation tags.

[0055] Specifically, given that the user pairing includes all the service users, on the one hand, service risk tags for both service users can be extracted, risk tag combinations can be constructed based on the risk tags of each service user, and the risk handling strategy corresponding to the pairing recommendation level can be determined and processed according to the pairing recommendation level matched by the risk tag combination. On the other hand, asset association data for both service users can also be obtained, and asset symmetry calculations can be performed on the user pairing based on the asset association data of each service user to obtain asset symmetry results. Subsequently, risk warning rules can be matched based on the asset symmetry results, and corresponding risk warning content can be generated for risk warning.

[0056] For example, in a mate selection scenario, user A's risk level is labeled as medium risk, their asset association is labeled as "long-term family support," and their mate selection recommendation label is "suitable for mate selection users who accept family support." User B's risk level is labeled as low risk, their asset association is labeled as "normal asset association," and their mate selection recommendation label is "prioritize matching with low-risk users." The combined risk labels of user A and user B are: medium risk + long-term family support + suitable for mate selection users who accept family support + low risk + normal asset association + priority matching with low-risk users. The matching recommendation level is determined to be a cautious recommendation. The cautious recommendation level corresponds to a risk warning pre-emptive strategy. Based on the risk warning pre-emptive strategy, the following strategy is implemented: a risk warning pop-up window is displayed to inform the mate selection users of the differences in their risk labels.

[0057] For example, asset correlation data can be obtained for users A and B in the mate selection process. User A has a guarantee amount of 0%, a financial support intensity of a%, and 2 family members (supporting 2 elderly people); User B has a guarantee amount of 0%, a financial support intensity of b% (a>b), and 1 family member. Symmetrical calculations of the guarantee amount ratio, financial support intensity, and number of family members can be performed for mate selection. Based on these calculations, a comprehensive asset symmetry calculation can be performed, and risk warnings can be issued based on the obtained asset symmetry results.

[0058] In the specific implementation process, during the risk matching process, interpersonal risk superposition calculation can also be performed on each service user to achieve two-way risk matching for each service user; in the second optional implementation provided in this embodiment, risk matching is performed based on the service risk tags, asset association data and credit data of each service user included in the user pairing to obtain the risk matching result of the user pairing, including: Extract the asset-related data of each service user, including the amount of guaranteed funds and the amount of funds flowing in. Calculate the guarantee ratio for each user based on the amount of guaranteed funds and the amount of funds flowing in, and issue risk warnings based on the guarantee ratio.

[0059] Specifically, during the risk matching process, the asset-related data of each service user, including the amount of guaranteed funds and the amount of funds flowing in, is extracted. The guarantee ratio of the users is calculated based on the amount of guaranteed funds and the amount of funds flowing in. The risk level can be determined based on the obtained guarantee ratio, thereby triggering risk alerts.

[0060] In practical applications, during the risk matching process, credit habit matching can also be performed on each service user. For example, it can be determined whether the credit values ​​of each service user match, whether the performance stability of each service user is similar, and / or whether the consumption habits of each service user are similar, thereby achieving two-way risk matching for each service user. In the third optional implementation provided in this embodiment, risk matching is performed based on the service risk tags, asset association data, and credit data of each service user included in the user pairing to obtain the risk matching result for the user pairing, including: Extract credit habit features from asset-related data and credit data of each service user and align the features accordingly; The credit habit characteristics of each service user are numerically mapped to obtain the credit mapping characteristics of each service user. The feature similarity of the credit mapping characteristics of each service user is calculated, and the credit matching result is determined based on the feature similarity.

[0061] Among them, credit habit characteristics refer to the characteristics used to characterize the stability and / or regularity of the credit behavior patterns of service users; for example, credit habit characteristics may include service users' repayment habit characteristics, debt management characteristics and / or credit inquiry characteristics, such as repayment on-time rate, consistency of public bill performance, stability of inflow and outflow and / or debt inflow ratio.

[0062] Specifically, during the risk matching process, the credit habit features of asset-related data and credit data of each service user are extracted and feature alignment is performed. Then, the aligned credit habit features of each service user are numerically mapped according to a preset numerical mapping rule to obtain the credit mapping features of each service user. The feature similarity of the credit mapping features of each service user is calculated, and the credit matching result is determined based on the feature similarity.

[0063] In addition to the above-mentioned implementation methods for risk matching, risk matching sub-results can also be obtained through risk matching processing by each matching sub-model, and the risk matching sub-results can be fused through a decision sub-model to obtain the risk matching result; in the fourth optional implementation method provided in this embodiment, risk matching processing is performed based on the service risk tags, asset association data and credit data of each service user included in the user pairing to obtain the risk matching result of the user pairing, including: The decision sub-model extracts risk matching sub-data corresponding to each matching sub-model from the service risk tags, asset association data and credit data of each service user; The extracted matching sub-data is input into the corresponding risk matching sub-model for risk matching processing to obtain risk matching sub-results; The risk matching text is obtained by fusing the risk matching sub-results through a decision sub-model.

[0064] Among them, the decision sub-model refers to the model used to coordinate the input allocation and output fusion of multiple risk matching sub-models in the risk matching process of user pairing; The risk matching sub-model can include a risk-compatible sub-model, an asset-symmetric sub-model, a guarantee ratio sub-model, and / or a credit matching sub-model. The risk-compatible sub-model is a model used to assess risk indicators arising from the external guarantees and / or joint liabilities of both service users. The asset-symmetric sub-model is a model used to analyze whether the burden of economic support from related users is balanced between the two service users; for example, it can be used to analyze the asset responsibilities of both service users in supporting parents and relatives. The guarantee ratio sub-model is a model used to calculate and assess the reasonableness of the external guarantee burden of both service users relative to their income levels. The credit matching sub-model is a model used to assess the similarity or synergy in the credit behavior patterns and asset habits of both service users.

[0065] Specifically, in the process of risk matching to obtain risk matching results, the decision sub-model classifies and analyzes the service risk tags, asset association data, and / or credit data of both service users, and extracts the corresponding risk matching sub-data for each matching sub-model according to the data mapping rules. Then, the extracted matching sub-data is input into the corresponding risk matching sub-model, and each risk matching sub-model performs corresponding risk matching processing to obtain risk matching sub-results. Subsequently, the decision sub-model can prioritize the risk matching sub-results according to preset weights, extract core evaluation indicators, and perform fusion processing to obtain fused core information. Based on this, it calls a preset copywriting template library to match copywriting to generate risk matching copywriting, and outputs the risk matching copywriting as the risk matching result. Specifically, in the process of inputting matching sub-data into the corresponding risk matching sub-model, for example, the risk level labels of both service users can be input into the risk compatibility sub-model to obtain risk compatibility sub-results; the asset symmetry related sub-data in the asset association data of both service users can be input into the asset symmetry sub-model to obtain asset symmetry sub-results; the amount of fund guarantee and / or the amount of fund inflow in the asset association data of both service users can be input into the guarantee ratio sub-model to obtain guarantee ratio sub-results; and the credit habit characteristics of the asset association data and credit data of both service users can be input into the credit matching sub-model to obtain credit matching results.

[0066] It should be noted that the above-mentioned method of obtaining risk matching results for user pairing based on service risk tags, asset association data, and credit data of each service user included in the user pairing process can be implemented using any one of the first, second, and third implementation methods according to actual needs. For example, the first optional implementation method, the second optional implementation method, or the third optional implementation method can be selected. In addition, during the specific implementation process, any two or all three implementation methods can be combined according to the actual needs of the execution process, or any two or all three implementation methods can be combined after adaptive modification, change, or deletion according to the actual needs of the execution process. For example, the above-mentioned method of performing risk matching processing based on the service risk tags, asset association data, and credit data of each service user included in user pairing to obtain the risk matching result of user pairing can be replaced by: determining the pairing recommendation level corresponding to the risk tag combination constructed based on the service risk tags of each service user, determining the risk disposal strategy corresponding to the pairing recommendation level, and performing strategy processing; extracting the amount of guaranteed funds and the amount of funds flowing in from the asset association data of each service user, calculating the guarantee ratio of user pairing based on the amount of guaranteed funds and the amount of funds flowing in, and performing risk warning processing based on the guarantee ratio; extracting the credit habit features of the asset association data and credit data of each service user and performing feature alignment; numerically mapping the credit habit features of each service user to obtain the credit mapping features of each service user, calculating the feature similarity of the credit mapping features of each service user, and determining the credit matching result based on the feature similarity.

[0067] In this embodiment, during the risk assessment process in the service scenario, the risk assessment process provided above can be executed by the service system or by a credit reporting node; that is, the executing entity of the risk assessment process in the service scenario provided above can be the service system or, alternatively, a credit reporting node. When the executing entity is a credit reporting node, before conducting risk assessment processing in the service scenario, the service system can also submit a call request to the credit reporting platform to initiate an interface call to the risk assessment interface of the credit reporting platform. After the credit reporting platform verifies the legality of the data authorization certificate of the service user, and if the legality verification is successful, the credit reporting node in the trusted data space responds to the interface call and applies for isolated data space in the trusted data space. After the isolated data space is created in the trusted data space, the credit reporting node can conduct risk assessment processing in the service scenario within the applied isolated data space.

[0068] In this context, a credit reporting node refers to a system node in which a credit reporting platform or credit reporting agency participates in the relevant risk assessment and processing of the service system. Specifically, in this embodiment, a credit reporting node can be a server-side or a credit reporting service terminal provided by the credit reporting platform to service users for credit reporting services or credit reporting applications. Furthermore, the credit reporting service terminal accesses a trusted data space and performs specific processing of credit reporting services or credit reporting applications within the environment provided by the trusted data space. Specifically, the risk assessment and processing of service scenarios can be carried out within the isolated data space provided by the trusted data space.

[0069] Trusted data space refers to an infrastructure system that integrates technological means and rule mechanisms and connects with various processing agencies involved in credit reporting. Specifically, it uses credit data as the core resource to achieve cross-domain resource interaction, and relies on technologies such as privacy computing and blockchain to build a trusted processing environment that ensures the usability but not visibility of credit data. At the same time, it also supports applications in related scenarios such as access control and data storage, providing the possibility for efficient data circulation and data value release in the credit reporting field.

[0070] An isolated data space refers to a temporarily secure computing environment, logically or physically isolated, created within a trusted data space framework for performing specific data processing tasks. The isolated data space can be allocated on demand by the trusted data space's resource scheduling mechanism, exists only during the execution of the data processing task, and is automatically destroyed upon completion. For example, an isolated data space can be created within the trusted data space during risk assessment processing in a service scenario, after a credit node responds to a service system's risk assessment interface call, and possesses independent resource allocation and security isolation mechanisms. Another example is an isolated data space created within the trusted data space during risk assessment processing in a mate selection scenario, after a credit node responds to a service system's risk assessment interface call.

[0071] Optionally, after the service system calls the risk assessment interface of the credit reporting platform, the credit reporting platform performs the risk assessment process for the service scenario at the credit reporting node in the trusted data space; specifically, the credit reporting node responds to the interface call by applying for isolated data space from the trusted data space, and performs the risk assessment process for the service scenario within the applied isolated data space.

[0072] It should be noted that the risk assessment process for the service scenarios provided in this implementation can be continuous. In this case, during the initial user pairing process based on the service system, the process of obtaining asset-related data and credit data based on the service user's data authorization credentials, calculating asset-related risk based on the asset-related data to obtain the service user's asset-related risk score, and conducting service risk assessment based on the asset-related risk score, asset-related data, and credit data to obtain the service user's service risk label can be performed. On this basis, risk matching processing can be performed based on the service risk labels, asset-related data, and credit data of each service user included in the service pairing to obtain the risk matching result of the user pairing. Furthermore, the risk assessment process for the service scenarios provided in this embodiment can also be discontinuous and executed in stages. For example, when a service user logs into the service system for the first time, they only determine the service risk label and do not initiate a matching request for user matching. In this case, based on the service risk label of the service user, if a matching request is subsequently obtained from the service user, the target service user of the service user can be retrieved based on the matching request submitted by the service user. Risk matching processing is then performed based on the service risk labels, asset association data, and credit data of the service user and the target service user included in the user pairing to obtain the risk matching result of the user pairing. Based on this, the above step S208 can also be replaced by: if a matching request submitted by a service user is detected, risk matching processing is performed based on the service risk labels, asset association data, and credit data of each service user included in the user pairing to obtain the risk matching result of the user pairing, and this, together with other processing steps provided in this embodiment, forms a new implementation method.

[0073] It should be added that each optional implementation method and each feasible execution method in steps S202 to S208 provided in this embodiment can be executed independently as needed, or they can be combined and referenced with each other. At the same time, each specific execution step in each optional implementation method or each feasible execution method can also be executed independently or combined as needed. Any feature in each execution step can also be deleted, or any feature in one execution step can be added to another execution step or replace any feature in another execution step. The execution conditions of "if" or "under what circumstances" involved in each step or operation can be directly deleted. This embodiment does not specifically limit the subsequent operations after the execution conditions.

[0074] In summary, the risk assessment method for service scenarios provided in this embodiment, during the risk assessment process, firstly, to ensure compliance during data acquisition, asset-related data and credit data are obtained based on the service user's data authorization credentials. Then, to transform the asset-related relationship between the service user and associated users into quantifiable indicators, thereby extending the risk assessment dimension from the individual service user level to the level of the relationship between the service user and associated users, asset-related risk is calculated based on the asset-related data to obtain the service user's asset-related risk score. Furthermore, to prevent service users from falling into risky relationships due to information asymmetry during user matching within the service system, and to improve the efficiency of risk information transmission to service users, service risk assessment is performed based on the asset-related risk score, asset-related data, and / or credit data to obtain the service user's service risk label. Subsequently, if a service user performs user matching through the service system, to improve the comprehensiveness and accuracy of risk matching, risk matching is performed based on the service labels of each service user included in the user matching, along with the service risk labels, asset-related data, and credit data, to obtain the risk matching result for the user matching. This improves the comprehensiveness and accuracy of risk assessment in service scenarios.

[0075] The following example uses the risk assessment and processing method for a service scenario provided in this embodiment in the context of mate selection. Figure 3 The risk assessment and handling method for the service scenarios provided in this embodiment will be further explained below. Figure 3 The risk assessment and handling method for service scenarios applied to mate selection includes the following steps.

[0076] Step S302: Obtain asset association data and credit data based on the data authorization certificate of the user seeking a spouse.

[0077] Optionally, the data authorization certificate is obtained after the user authorizes the data through the data authorization interface of the credit reporting platform provided by the dating system; asset-related data includes de-identified asset-related data after data anonymization, and credit data includes de-identified credit data after data anonymization.

[0078] Step S304: Extract sub-data of multiple asset association dimensions contained in the asset association data, and calculate the association score of each asset association dimension based on the sub-data.

[0079] Among them, the correlation score of each asset correlation dimension is calculated based on the sub-data, including: calculating the financial support intensity based on the transfer sub-data of funds transferred from the dating user to the associated user and the fund inflow sub-data; calculating the guarantee risk value based on the financial guarantee sub-data of the dating user and the probability of guarantee overdue; and calculating the financial dependence based on the fund inflow sub-data of funds received by the dating user and the expenditure sub-data.

[0080] Step S306: Calculate the asset association risk score based on the association scores and corresponding association weights of each asset association dimension.

[0081] Step S308: Input the asset-related risk score, asset-related data, and credit data into the risk rating model for risk rating processing to obtain the risk level label.

[0082] Step S310: Input the asset association data into the rule engine to perform asset association detection and obtain asset association tags.

[0083] Step S312: Based on asset-related risk scores, asset-related data, and credit data, perform mate selection recommendation detection to obtain mate selection recommendation labels.

[0084] Step S314: Based on the risk tag combination corresponding to the mate selection risk tag of each mate selection user, determine the risk handling strategy corresponding to the mate selection recommendation level and perform strategy processing.

[0085] Step S316: Perform asset symmetry calculation for mate matching based on the asset association data of each mate-selecting user, and issue risk warnings based on the obtained asset symmetry results.

[0086] Step S318: Extract the asset-related data of each mate selection user, including the amount of financial guarantees and the amount of funds flowing in.

[0087] Step S320: Calculate the guarantee ratio for mate matching based on the amount of financial guarantee and the amount of capital inflow, and handle risk warnings based on the guarantee ratio.

[0088] It should be noted that any one or more steps in steps S302 to S320 can be combined with any one or more steps in steps S202 to S208 to form a new implementation method according to the needs of implementation and deployment. In addition, any one or more technical features in steps S302 to S320 can be selected and combined with any one or more technical features provided in steps S202 to S208 to form a new implementation method according to the actual deployment needs. Alternatively, any one or more technical features in steps S302 to S320 can be replaced with any one or more technical features provided in steps S202 to S208 to form a new implementation method according to the actual deployment needs. These will not be elaborated on here.

[0089] This manual provides an example of a risk assessment and processing device for a service scenario, as follows: In the above embodiments, a risk assessment and processing method for a service scenario is provided, and correspondingly, a risk assessment and processing device for a service scenario is also provided, which will be described below with reference to the accompanying drawings.

[0090] Reference Figure 4 This illustration shows a schematic diagram of an embodiment of a risk assessment processing device for a service scenario provided in this embodiment.

[0091] Since the apparatus embodiments correspond to the method embodiments, the descriptions are relatively simple. For relevant parts, please refer to the corresponding descriptions of the method embodiments provided above. The apparatus embodiments described below are merely illustrative.

[0092] This embodiment provides a risk assessment and processing device for a service scenario, the device comprising: The data acquisition module 402 is configured to acquire the asset association data and credit data of the service user based on the data authorization credentials of the service user synchronized by the service system; The scoring calculation module 404 is configured to calculate the asset association risk based on the asset association data to obtain the asset association risk score of the service user. The risk assessment module 406 is configured to perform a service risk assessment based on the asset-related risk score, the asset-related data, and the credit data, and obtain the service risk label of the service user. The risk matching processing module 408 is configured to perform risk matching processing based on the service risk tags, asset association data and credit data of each service user included in the user pairing, and obtain the risk matching result of the user pairing.

[0093] For ease of description, the above devices are described by dividing them into various modules or units based on their functions. Of course, when implementing one or more of these specifications, the functions of each module or unit can be implemented in the same or different software and / or hardware, or a module that performs the same function can be implemented by a combination of multiple sub-modules or sub-units, etc. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0094] This manual provides an example of a risk assessment and processing device for a service scenario, as follows: Corresponding to the risk assessment and processing method for a service scenario described above, based on the same technical concept, one or more embodiments of this specification also provide a risk assessment and processing device for a service scenario. This device is used to execute the risk assessment and processing method for a service scenario provided above. Figure 5 This is a schematic diagram of the structure of a risk assessment processing device for a service scenario provided in one or more embodiments of this specification.

[0095] This embodiment provides a risk assessment and processing device for a service scenario, including: like Figure 5As shown, device 500 mainly consists of a communication interface 502, a user interface 504, a processor 506, and a data storage 508. These components are interconnected and communicate with each other via a system bus, network, or other connection mechanism 510. The communication interface 502 enables device 500 to communicate with other devices, access networks, and transmission networks via analog or digital modulation. For example, the communication interface 502 may include a chipset and antenna for wireless communication with a radio access network or access point. Furthermore, the communication interface 502 can be a wired interface such as Ethernet, Token Ring, or a USB port, or a wireless interface such as Wi-Fi, Bluetooth, Global Positioning System (GPS), or a wide-area wireless interface (e.g., WiMAX or LTE). Of course, the communication interface 502 can also support other forms of physical layer interfaces and standard or proprietary communication protocols. The communication interface 502 may also include multiple physical communication interfaces, such as Wi-Fi, Bluetooth, and wide-area wireless interfaces. The user interface 504 includes receiving user input and providing output to the user. Therefore, user interface 504 may include input components such as a keypad, keyboard, touch-sensitive or presence-sensitive panel, computer mouse, trackball, joystick, microphone, still camera, and video camera, and output components such as a display screen (which may be combined with a touch-sensitive panel), CRT, LCD, LED, display using DLP technology, printer, and other similar devices known or developed in the future. User interface 504 may also generate auditory output via speakers, speaker jacks, audio output ports, audio output devices, headphones, and other similar devices known or developed in the future. In some embodiments, user interface 504 may include software, circuitry, or other forms of logic capable of transmitting and receiving data to and from external user input / output devices. Additionally or alternatively, device 500 may support remote access from other devices via communication interface 502 or another physical interface (not shown). User interface 504 may be configured to receive user input, the position and movement of which may be indicated by indicators or cursors described herein. User interface 504 may also be configured as a display device for rendering or displaying text fragments.

[0096] Processor 506 may include one or more general-purpose processors and / or special-purpose processors. Data storage 508 may include one or more volatile and / or non-volatile storage components and may be integrated wholly or partially with processor 506. Data storage 508 may include removable and non-removable components.

[0097] Processor 506 is capable of executing program instructions 518 (e.g., compiled or uncompiled program logic and / or machine code) stored in data storage 508 to perform the various functions described herein. Data storage 508 may contain a non-transitory computer-readable medium on which program instructions are stored, which, when executed by device 500, enable device 500 to perform any methods, processes, or functions disclosed in this specification and / or the accompanying drawings. Execution of program instructions 518 by processor 506 may result in processor 506 using data 512. For example, program instructions 518 may include an operating system 522 (e.g., an operating system kernel, device drivers, and / or other modules) installed on device 500 and one or more application programs 520 (e.g., a browser, social application, or game application). Similarly, data 512 may include operating system data 516 and application data 514. Operating system data 516 is primarily accessible to operating system 522, while application data 514 is primarily accessible to one or more application programs 520. Application data 514 may reside in a file system visible or hidden from the user of device 500. Application 520 can communicate with operating system 522 through one or more application programming interfaces (APIs). These APIs facilitate application 520 in reading and / or writing application data 514, transmitting or receiving information via communication interface 502, and receiving or displaying information on user interface 504. In some terms, application 520 may be simply referred to as "app". Furthermore, application 520 can be downloaded to device 500 through one or more online app stores or app markets. However, applications can also be installed on device 500 in other ways, such as through a web browser or a physical interface on device 500 (e.g., a USB port).

[0098] In one specific embodiment, the risk assessment processing device for a service scenario includes a memory and one or more programs, wherein one or more programs are stored in the memory, and one or more programs may include one or more modules, and each module may include a series of computer-executable instructions for the risk assessment processing device for the service scenario, and is configured to be executed by one or more processors. The one or more programs include computer-executable instructions for performing the following: Based on the data authorization credentials of the service user synchronized by the service system, obtain the asset association data and credit data of the service user; Based on the asset association data, an asset association risk calculation is performed to obtain the asset association risk score of the service user. Service risk assessment is performed based on asset-related risk scores, asset-related data, and credit data to obtain service risk labels for service users. Risk matching is performed based on the service risk tags, asset association data, and credit data of each service user included in the user pairing to obtain the risk matching result of the user pairing.

[0099] This specification provides an embodiment of a computer-readable storage medium as follows: Corresponding to the risk assessment and processing method for the service scenario described above, based on the same technical concept, one or more embodiments of this specification also provide a computer-readable storage medium.

[0100] The computer-readable storage medium provided in this embodiment is used to store computer-executable instructions, which, when executed, implement the following process: Based on the data authorization credentials of the service user synchronized by the service system, obtain the asset association data and credit data of the service user; Based on the asset association data, an asset association risk calculation is performed to obtain the asset association risk score of the service user. Service risk assessment is performed based on asset-related risk scores, asset-related data, and credit data to obtain service risk labels for service users. Risk matching is performed based on the service risk tags, asset association data, and credit data of each service user included in the user pairing to obtain the risk matching result of the user pairing.

[0101] It should be noted that the embodiments of a computer-readable storage medium described in this specification and the embodiments of a risk assessment processing method for a service scenario described in this specification are based on the same inventive concept. Therefore, the specific implementation of this embodiment can be referred to the implementation of the corresponding method described above, and the repeated parts will not be described again.

[0102] This specification provides an example of a computer program product as follows: Corresponding to the risk assessment and processing method for the service scenario described above, based on the same technical concept, one or more embodiments of this specification also provide a computer program product.

[0103] A computer program product includes a computer program / instructions that, when executed by a processor, perform the following steps: Based on the data authorization credentials of the service user synchronized by the service system, obtain the asset association data and credit data of the service user; Based on the asset association data, an asset association risk calculation is performed to obtain the asset association risk score of the service user. Service risk assessment is performed based on asset-related risk scores, asset-related data, and credit data to obtain service risk labels for service users. Risk matching is performed based on the service risk tags, asset association data, and credit data of each service user included in the user pairing to obtain the risk matching result of the user pairing.

[0104] It should be noted that the embodiments of a computer program product described in this specification and the embodiments of a risk assessment processing method for a service scenario described in this specification are based on the same inventive concept. Therefore, the specific implementation of this embodiment can be referred to the implementation of the corresponding method described above, and the repeated parts will not be described again.

[0105] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments. For example, the device embodiment, equipment embodiment and computer-readable storage medium embodiment are all similar to the method embodiment, so the description is relatively simple. When reading the relevant content of the device embodiment, equipment embodiment and computer-readable storage medium embodiment, please refer to the description of the method embodiment.

[0106] While one or more embodiments of this specification provide method steps as described in the embodiments or flowcharts, it is understood that the order of steps listed in the embodiments or flowcharts is merely one possible execution order among many steps, and does not represent the only execution order. Therefore, when the claims involve method steps, any changes or adjustments to the order of such steps, or the parallelism between steps, are also within the scope of protection of the claims. This specification uses specific terms to describe embodiments of this specification. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that "an embodiment," "one embodiment," or "an alternative embodiment" mentioned twice or more in different locations in this specification do not necessarily refer to the same embodiment. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0107] The foregoing has described 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 that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0108] In the 1930s, 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 improvements to the methodology 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 an improvement to the methodology 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 and "integrate" a digital system onto a PLD themselves, 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 also 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.

[0109] 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: ARC 525D, Atmel AT91SAM, Microchip PIC18F25K20, 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, application-specific integrated circuits, 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.

[0110] 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.

[0111] For ease of description, the above apparatus is described by dividing it into various functional units. Of course, when implementing the embodiments of this specification, the functions of each unit can be implemented in one or more software and / or hardware.

[0112] Those skilled in the art will understand that one or more embodiments of this specification can be provided as a method, system, or computer program product. Therefore, one or more embodiments of this specification may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this specification may take the form of a computer program product embodied on one or more computer-readable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0113] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. 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.

[0114] 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.

[0115] 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.

[0116] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0117] 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.

[0118] Computer-readable media include 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, program modules, or other data. Examples of computer-readable 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, 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.

[0119] 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 at least one…" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0120] One or more embodiments of this specification 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 particular task or implement a particular abstract data type. One or more embodiments of this specification 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.

[0121] The above description is merely an embodiment of this document and is not intended to limit the scope of this document. Various modifications and variations can be made to this document by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this document should be included within the scope of the claims of this document.

Claims

1. A risk assessment and processing method for a service scenario, characterized in that, The method includes: Based on the data authorization credentials of the service user synchronized by the service system, obtain the asset association data and credit data of the service user; Based on the asset association data, an asset association risk calculation is performed to obtain the asset association risk score of the service user. Service risk assessment is performed based on asset-related risk scores, asset-related data, and credit data to obtain service risk labels for service users. Risk matching is performed based on the service risk tags, asset association data, and credit data of each service user included in the user pairing to obtain the risk matching result of the user pairing.

2. The risk assessment and processing method for service scenarios according to claim 1, characterized in that, The process of calculating asset-related risk based on the asset-related data to obtain the asset-related risk score for the service user includes: Extract sub-data from multiple asset association dimensions contained in the asset association data, and calculate the association score for each asset association dimension based on the sub-data; The asset association risk score is calculated based on the association scores and corresponding association weights for each asset association dimension.

3. The risk assessment and processing method for service scenarios according to claim 2, characterized in that, The calculation of the association score for each asset association dimension based on the sub-data includes: The funding intensity is calculated based on the transfer data and fund inflow data of service users transferring funds to associated users; the guarantee risk value is calculated based on the fund guarantee data and guarantee delinquency probability of service users; and / or, the funding dependence is calculated based on the fund inflow data and fund outflow data of service users receiving funds.

4. The risk assessment and processing method for service scenarios according to claim 1, characterized in that, The process of assessing service risk based on asset-related risk scores, asset-related data, and credit data to obtain service risk tags for service users includes: The asset-related risk score, asset-related data, and credit data are input into the risk rating model for risk rating processing to obtain risk level labels; The asset association data is input into the rule engine to perform asset association detection and obtain asset association tags.

5. The risk assessment and processing method for service scenarios according to claim 4, characterized in that, The risk rating model includes a gradient boosting model for performing the risk rating process; The risk rating process includes: Feature extraction is performed on the input asset-related risk scores, asset-related data, and credit data, and structured features are constructed based on the extracted features; The structured features and baseline risk scores are input into multiple pre-trained decision trees to perform rule matching decisions to obtain risk scores, and the risk scores are then mapped to risk levels to obtain risk level labels.

6. The risk assessment and processing method for service scenarios according to claim 4, characterized in that, The step of assessing service risk based on asset-related risk scores, asset-related data, and credit data to obtain service risk labels for service users also includes: Based on the asset-related risk score, asset-related data, and credit data, user pairing recommendation detection is performed to obtain pairing recommendation tags; The step of performing user pairing recommendation detection based on the asset-related risk score, asset-related data, and credit data includes: inputting at least one of the asset-related risk score, asset-related data, and credit data into the user pairing recommendation detection model to perform user pairing recommendation detection and obtain the pairing recommendation label.

7. The risk assessment and processing method for service scenarios according to claim 1, characterized in that, The step of assessing service risk based on asset-related risk scores, asset-related data, and credit data to obtain service risk labels for service users also includes: Retrieve a set of users of the same type as the service user in the service user database, and construct a feature fusion distribution based on the standardized characteristics of each user of the same type in the set; Feature extraction and standardization are performed on at least one of the asset-related risk scores, asset-related data, and credit data, and feature fusion calculation is performed on the obtained standardized features of the service users to obtain feature fusion values; The feature fusion values ​​are mapped to the feature fusion distribution to obtain the recommendation quantiles, and the paired recommendation tags are determined based on the recommendation quantiles.

8. The risk assessment and processing method for service scenarios according to claim 1, characterized in that, The risk matching process based on the service risk tags, asset association data, and credit data of each service user included in the user pairing, to obtain the risk matching result of the user pairing, includes: Based on the risk tag combination constructed from the service risk tags of each service user, the corresponding pairing recommendation level is determined, and the risk handling strategy corresponding to the pairing recommendation level is processed. Based on the asset association data of each service user, the asset symmetry calculation of the user pairing is performed, and risk warnings are issued based on the obtained asset symmetry results.

9. The risk assessment and processing method for service scenarios according to claim 8, characterized in that, The risk matching process based on the service risk tags, asset association data, and credit data of each service user included in the user pairing, to obtain the risk matching result of the user pairing, further includes: Extract the asset association data of each service user, including the amount of guaranteed funds and the amount of funds flowing in, calculate the guarantee ratio of the user pairing based on the amount of guaranteed funds and the amount of funds flowing in, and perform risk warning processing based on the guarantee ratio.

10. The risk assessment and processing method for service scenarios according to claim 8, characterized in that, The risk matching process based on the service risk tags, asset association data, and credit data of each service user included in the user pairing, to obtain the risk matching result of the user pairing, further includes: Extract credit habit features from the asset-related data and credit data of each service user and perform feature alignment; The credit habit characteristics of each service user are numerically mapped to obtain the credit mapping characteristics of each service user. The feature similarity of the credit mapping characteristics of each service user is calculated, and the credit matching result is determined based on the feature similarity.

11. The risk assessment and processing method for service scenarios according to claim 1, characterized in that, The risk matching process based on the service risk tags, asset association data, and credit data of each service user included in the user pairing, to obtain the risk matching result of the user pairing, includes: The decision sub-model extracts the risk matching sub-data corresponding to each matching sub-model from the service risk tags, asset association data and credit data of each service user; The extracted matching sub-data is input into the corresponding risk matching sub-model for risk matching processing to obtain risk matching sub-results; The risk matching text is obtained by fusing the risk matching sub-results through the decision sub-model.

12. The risk assessment and processing method for service scenarios according to claim 1, characterized in that, The data authorization credential is obtained by the service user after authorizing data through the data authorization interface of the credit reporting platform provided by the service system. The asset association data includes de-identified asset association data after data anonymization, and the credit data includes de-identified credit data after data anonymization.

13. The risk assessment and processing method for service scenarios according to claim 12, characterized in that, The risk assessment and processing method for the service scenario is executed after the service system calls the risk assessment interface of the credit reporting platform, and is executed by the credit reporting platform at the credit reporting node in the trusted data space. The credit reporting node responds to the interface call by requesting an isolated data space from the trusted data space, and executes the risk assessment and processing method for the service scenario within the requested isolated data space.

14. A risk assessment and processing device for a service scenario, characterized in that, The device includes: The data acquisition module is configured to acquire the asset association data and credit data of the service user based on the data authorization credentials of the service user synchronized by the service system; The scoring calculation module is configured to calculate the asset association risk based on the asset association data to obtain the asset association risk score of the service user. The risk assessment module is configured to perform service risk assessment based on the asset-related risk score, the asset-related data, and the credit data, and obtain the service risk label of the service user. The risk matching processing module is configured to perform risk matching processing based on the service risk tags, asset association data and credit data of each service user included in the user pairing, and obtain the risk matching result of the user pairing.

15. A risk assessment and processing device for a service scenario, characterized in that, The device includes: A processor; and a memory configured to store computer-executable instructions, which, when executed, cause the processor to: Based on the data authorization credentials of the service user synchronized by the service system, obtain the asset association data and credit data of the service user; Based on the asset association data, an asset association risk calculation is performed to obtain the asset association risk score of the service user. Service risk assessment is performed based on asset-related risk scores, asset-related data, and credit data to obtain service risk labels for service users. Risk matching is performed based on the service risk tags, asset association data, and credit data of each service user included in the user pairing to obtain the risk matching result of the user pairing.

16. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store computer-executable instructions that, when executed, implement the steps of the method of claim 1.

Images