Risk discrimination system and method, computer storage medium and electronic equipment

Abstract

The invention discloses a risk discrimination system and method, a computer storage medium and electronic equipment. The system comprises a risk management platform and a risk discrimination platform, the risk management platform is used for receiving risk marking data which is fed back by the risk discrimination platform and is used for carrying out risk marking on the transaction data, distributing the risk marking data as a risk discrimination task, initiating a risk discrimination request to the risk discrimination platform, and obtaining a discrimination auditing conclusion fed back by the risk discrimination platform; the risk discrimination platform is used for carrying out risk marking on the transaction data and determining risk marking data; in response to the risk discrimination request, determining the risk marking data and transaction data corresponding to the risk marking data as to-be-discriminated data; performing risk discrimination on the to-be-discriminated data according to a multi-task risk automatic discrimination model independently trained according to the risk type, determining a discrimination auditing conclusion of the to-be-discriminated data, and sending the discrimination auditing conclusion to a risk management platform; therefore, the accuracy of risk discrimination can be ensured.

Description

Technical Field

[0001] The present application relates to the field of information security technology, and more particularly to a risk identification system, and a method and device for the risk identification system. The present application also relates to a risk data output method and device, and a computer storage medium and electronic device. Background Art

[0002] Electronic Commerce (E-Commerce for short) refers to business activities that use information network technology as a means and exchange of goods as a purpose. It can also be understood as trading activities and related service activities that rely on the Internet and computer technologies and are conducted in an electronic transaction manner. It is another form of electronicization, networking, and informatization of all aspects of traditional business activities. Activities in the form of electronic transactions include online transactions between consumers and merchants, online electronic payments, and various business activities, transaction activities, financial activities, and related comprehensive service activities. It is a new business operation model. Nowadays, the way of relying on the Internet to realize transactions has become a routine activity in daily life.

[0003] As a bridge connecting buyers and sellers, e-commerce platforms (transaction software platforms) have greatly promoted the convenience and efficiency of transactions, but they have also brought about many risks and challenges.

[0004] First, network information security risks cannot be ignored. In e-commerce transactions, a large amount of sensitive information is transmitted, such as personal identity information, bank account details and transaction data. If the platform security measures are not in place, it is very vulnerable to security threats such as hacker attacks and data leaks, which not only damages user privacy, but may also cause financial losses and a crisis of trust.

[0005] Secondly, transaction fraud risk is another major hidden danger. The anonymity and remote transaction characteristics of e-commerce platforms provide opportunities for criminals to take advantage of, such as false product information, phishing websites, malicious refund requests and other fraudulent behaviors, which may cause economic losses to consumers and merchants and affect the reputation of the platform.

[0006] Furthermore, legal compliance risk is also a factor that must be considered. Different countries and regions have different laws and regulations on e-commerce, including consumer rights protection, data protection, tax regulations, etc. If the platform fails to comply with relevant laws and regulations, it may face the risk of legal proceedings, fines, or even business prohibition.

[0007] Therefore, paying attention to and discovering risks and judging risks are crucial to the healthy operation of e-commerce software platforms. Summary of the invention

[0008] The present application provides a risk identification system to solve the problems of poor identification accuracy and high cost caused by relying on manual risk review in the prior art.

[0009] The present application provides a risk identification system, characterized in that it includes: a risk management platform and a risk identification platform;

[0010] The risk management platform is used to receive risk-labeled data for risk-labeling transaction data fed back by the risk identification platform, and to distribute the risk-labeled data as a risk identification task, initiate a risk identification request to the risk identification platform, and obtain the identification review conclusion corresponding to the risk identification request fed back by the risk identification platform;

[0011] The risk identification platform is used to perform risk labeling on the transaction data and determine the risk labeling data of the transaction data; and is used to respond to the risk identification request initiated by the risk management platform for the risk identification task, determine the risk labeling data in the risk identification request and the transaction data corresponding to the risk labeling data as data to be identified; perform risk identification on the data to be identified using a multi-task risk automatic identification model trained separately according to risk types, determine the identification and review conclusions of the data to be identified, and send the identification and review conclusions to the risk management platform.

[0012] In some embodiments, the risk identification platform is used to obtain the transaction data through the first acquisition link and / or the second acquisition link according to the set automatic labeling trigger conditions; and to perform risk labeling on the transaction data according to the constructed risk judgment rule set to determine the risk labeling data corresponding to the transaction data.

[0013] In some embodiments, the risk identification platform is used to train a first classification sub-model according to a first risk type to determine a target first classification sub-model; to train a second classification sub-model according to a second risk type to determine a target second classification sub-model; and to train a third classification sub-model according to a third risk type to determine a target third classification sub-model;

[0014] The judgment and audit conclusion of the data to be judged is determined by inputting the data to be judged into a multi-task risk automatic judgment model fitted according to the first classification sub-model, the second classification sub-model and the third classification sub-model.

[0015] In some embodiments, the risk management platform further includes:

[0016] It is used to screen the target discrimination audit conclusions that meet the discrimination requirements based on the obtained discrimination audit conclusions, and push the target discrimination audit conclusions and the transaction data corresponding to the target discrimination audit conclusions as advancement data to the data processing stage to perform corresponding processing operations.

[0017] In some embodiments, the risk management platform is used to determine the data feature of the risk-tagged data extracted from the data to be identified as the first feature; to determine the selected feature of the transaction data extracted from the data to be identified as the second feature; to determine the statistical feature of the selected feature of the transaction data extracted from the data to be identified as the third feature; and to determine the risk profile data feature of the transaction data extracted from the data to be identified as the fourth feature;

[0018] One or more of the first feature, the second feature, the third feature and the fourth feature are input into the automatic discrimination model to perform risk type discrimination, and the risk type discrimination result is determined as the discrimination review conclusion of the data to be discriminated.

[0019] This application also provides a risk identification method, including:

[0020] In response to a risk identification request initiated for a risk identification task, determining the risk-labeled data in the risk identification request and the transaction data corresponding to the risk-labeled data as data to be identified;

[0021] According to a multi-task risk automatic discrimination model trained separately based on risk types, risk discrimination is performed on the data to be discriminated, and a discrimination review conclusion of the data to be discriminated is determined;

[0022] Send the judgment and review conclusion.

[0023] In some embodiments, it also includes:

[0024] The acquired transaction data is risk-labeled to determine risk-labeled data of the transaction data.

[0025] In some embodiments, the step of performing risk tagging on the acquired transaction data and determining risk tagging data of the transaction data includes:

[0026] According to the set automatic marking trigger condition, the transaction data is acquired through the first acquisition link and / or the second acquisition link;

[0027] The transaction data is risk-labeled according to the constructed risk judgment rule set, and the risk-labeled data corresponding to the transaction data is determined.

[0028] In some embodiments, acquiring transaction data through the first acquisition link and / or the second acquisition link according to the set automatic marking trigger condition includes:

[0029] According to the set selection conditions, a transaction data snapshot of a transaction participant that meets the risk identification is selected as the first acquisition link, and when the automatic marking trigger condition is met, the transaction data is acquired;

[0030] and / or,

[0031] According to the set selection conditions, the integrated data after integrating the transaction data of the transaction participants that meet the risk judgment is selected as the second acquisition link, and the transaction data is acquired when the automatic marking trigger condition is met.

[0032] In some embodiments, the risk identification of the data to be identified is performed based on the multi-task risk automatic identification model trained separately based on the risk type, and the identification and review conclusion of the data to be identified is determined, including:

[0033] Training the first classification sub-model according to the first risk type to determine a target first classification sub-model;

[0034] Training the second classification sub-model according to the second risk type to determine a target second classification sub-model;

[0035] The third classification sub-model is trained according to the third risk type to determine a target third classification sub-model;

[0036] Fitting the target first classification sub-model, the target second classification sub-model and the target third classification sub-model into a multi-task risk automatic discrimination model;

[0037] The data to be judged is input into the multi-task risk automatic judgment model to determine the judgment and audit conclusion of the data to be judged.

[0038] In some embodiments, the step of inputting the data to be identified into the multi-task risk automatic identification model to determine the identification and review conclusion of the data to be identified includes:

[0039] Determine the data feature of the risk-labeled data extracted from the data to be identified as the first feature;

[0040] Determine the selected feature of the transaction data in the extracted data to be identified as the second feature;

[0041] Determine the statistical features of the selected features of the transaction data in the extracted data to be identified as the third feature;

[0042] Determine the risk profile data feature of the transaction data extracted from the data to be identified as the fourth feature;

[0043] Inputting one or more of the first feature, the second feature, the third feature and the fourth feature into the multi-task risk automatic identification model to identify the risk type;

[0044] The risk type identification result is determined as the identification and review conclusion of the data to be identified.

[0045] In some embodiments, sending the determination and review conclusion includes:

[0046] According to the comparison between the judgment and audit conclusion and the set judgment threshold, the data to be judged that meets the advancement requirements is advanced to the end processing operation stage.

[0047] The present application also provides a risk data output method, including:

[0048] Distribute the received risk-labeled data as risk identification tasks;

[0049] Generate and send a risk identification request for target transaction data according to a request for viewing the risk identification task, wherein the risk identification request carries the target transaction data and target risk labeling data corresponding to the target transaction data;

[0050] A multi-task risk automatic discrimination model trained separately based on risk types is received, and a discrimination and audit conclusion corresponding to the target transaction data is output.

[0051] In some embodiments, the receiving of the multi-task risk automatic discrimination model trained separately based on the risk type and the output of the discrimination review conclusion corresponding to the target transaction data include:

[0052] receiving the target transaction data and the target risk labeling data corresponding to the target transaction data as input data of the multi-task risk automatic discrimination model, and outputting a discrimination and review conclusion on the risk type of the target transaction data;

[0053] The judgment and review conclusion is stored and displayed on the viewing page corresponding to the viewing request.

[0054] In some embodiments, it also includes:

[0055] The discrimination and audit conclusion display area displayed in the viewing page displays second prompt information corresponding to the discrimination and audit conclusion.

[0056] In some embodiments, it also includes:

[0057] According to the received risk labeling data, adding first prompt information corresponding to the risk labeling data;

[0058] The risk marking data and the first prompt information are output.

[0059] The present application also provides a computer storage medium for storing data generated by a network platform, and a program for processing the data generated by the network platform;

[0060] When the program is read and executed, it executes the logical steps executed in the above-mentioned risk identification system, or executes the steps of the above-mentioned risk identification method, or executes the steps of the above-mentioned risk data output method.

[0061] The present application also provides an electronic device, including:

[0062] processor;

[0063] A memory for storing a program for processing data generated by a network platform. When the program is read and executed by the processor, it executes the logical steps executed in the above-mentioned risk identification system, or executes the steps of the above-mentioned risk identification method, or executes the steps of the above-mentioned risk data output method.

[0064] Compared with the prior art, this application has the following advantages:

[0065] The risk identification system provided by the present application performs risk identification on the acquired transaction data through the risk identification platform, and sends the risk identification data after identification to the risk management platform. The risk management platform initiates a risk identification request to the risk identification platform based on the risk identification task, so that the risk points of the transaction data can be perceived through the risk identification data at the first time of invoking the risk identification task, and the risk identification platform determines the risk identification data carried in the risk identification request and the transaction data corresponding to the risk identification data as the data to be identified; according to the multi-task risk automatic identification model trained separately based on the risk type, the data to be identified is identified, the identification and review conclusion of the data to be identified is determined, and the identification and review conclusion is sent to the risk management platform, so as to ensure the accuracy of the risk identification of the transaction data and avoid omission and misjudgment of risk points. In addition, the risk identification conclusion that meets the requirements can be advanced to the end stage of risk identification, thereby improving the efficiency of risk identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 It is a structural schematic diagram of a risk identification system provided by this application;

[0067] Figure 2It is a timing diagram of an embodiment of a risk identification system performing identification tasks provided by the present application;

[0068] Figure 3 It is a schematic diagram of an embodiment of a multi-task risk automatic discrimination model training process in a risk discrimination system provided by the present application;

[0069] Figure 4 This is a schematic diagram of an automatic identification embodiment of a risk identification system provided by the present application;

[0070] Figure 5 It is a flow chart of a risk identification method provided by this application;

[0071] Figure 6 It is a structural schematic diagram of a risk identification device provided by the present application;

[0072] Figure 7 is a flow chart of a risk data output method provided by the present application;

[0073] Figure 8 It is a structural schematic diagram of a risk data output device provided by the present application;

[0074] Fig. 9 It is a structural schematic diagram of an electronic device provided by this application. DETAILED DESCRIPTION

[0075] Many specific details are described in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application, so the present application is not limited by the specific implementation disclosed below.

[0076] The terms used in this application are only for the purpose of describing specific embodiments and are not intended to limit this application. The descriptions used in this application and the appended claims, such as "a", "first", and "second", are not limitations on quantity or sequence, but are used to distinguish the same type of information from each other.

[0077] In the existing technology, although significant progress has been made in the risk detection of online financial transactions, it still faces a series of challenges and problems. First, the rapid development of technology does not match the pace of financial innovation, making it difficult for risk detection methods to fully cover all new risks. For example, with the application of emerging technologies such as blockchain and artificial intelligence, the methods and forms of online financial transactions are constantly evolving, and the existing risk detection system may not be able to adapt to these changes in a timely manner, resulting in detection blind spots. Secondly, data privacy and information security issues remain a major problem in risk detection. Online financial transactions involve a large amount of sensitive information, such as personal identity information, transaction records, etc., which are vulnerable to threats such as hacker attacks and internal leaks during transmission and storage. Although security measures such as data encryption and firewalls are widely used, there is still a risk of being cracked or bypassed, resulting in data leaks and other incidents. In addition, there are also challenges in the accuracy and timeliness of risk detection systems. Due to the complexity and diversity of online financial transactions, risk detection systems need to process a large amount of data and identify potential risk signals from them. However, due to limitations in algorithm design and data processing capabilities, the accuracy and timeliness of the system are often difficult to reach an ideal level. This may lead to some real risk events being missed or delayed, especially in the process of relying on manual review to determine the authenticity of risk probability based on potential risk signals, which will further lead to problems such as missed detection, wrong detection or misjudgment. Finally, the update and maintenance of the risk detection system also face difficulties. With the continuous advancement of technology and the continuous changes in the market, the risk detection system needs to be continuously updated and optimized to adapt to new risk types and transaction patterns. However, the update and maintenance of the system requires a lot of manpower, material and financial resources, and may face challenges in technical compatibility and stability.

[0078] In summary, there are still many problems and challenges in the risk detection of online financial transactions in the prior art. In order to further improve the accuracy and timeliness of risk detection, and enhance the intelligence and automation level of the system. In view of this, the present application provides a risk identification system, which can improve the accuracy of risk identification, avoid the occurrence of problems such as missed detection, wrong detection or misjudgment of risks through automated execution, and can also improve the speed of risk identification and improve work efficiency.

[0079] In view of this, it can be understood that the invention concept of this application comes from the need for risk prevention and control in existing transactions through online platforms, especially in the face of a large number of transactions where purchases are made first and payments are made later, where the prevention and control of transaction risks is particularly important. Before describing in detail a risk identification system provided by this application, the relevant technical terms involved in the embodiments provided by this application are first explained as follows:

[0080] Anti-fraud: refers to the prevention and combating of various fraudulent activities through a series of measures and methods in financial credit business. Fraud refers to the act of obtaining loans, defrauding insurance claims, embezzling credit and evading repayment responsibilities by deception, false statements or other illegal means during the process of applying for loans or purchasing insurance, approval, issuance and repayment.

[0081] Anti-fraud analysis (judgment): refers to the process of early warning, analysis, evaluation and disposal of fraudulent behavior, so as to timely discover and respond to fraudulent behavior, reduce fraud risks, and protect the interests of enterprises and customers.

[0082] Anti-fraud Analysis and Judgment Center: refers to a one-stop financial anti-fraud analysis and judgment platform that supports the entire life cycle of anti-fraud cases from case creation, case merging, case allocation, team queues, case analysis and judgment, to case disposal and closure.

[0083] Risk labeling: refers to the operation of identifying, classifying and marking potential risk factors during the risk management and control process; it facilitates a clearer understanding and analysis of the case.

[0084] Transaction data: refers to the data generated by financial transactions through the Internet, which may include buyers and sellers or third parties, including: data on completed payments and / or data on uncompleted payments.

[0085] Purchase first, pay later: A scenario based on the trading platform where users obtain items first and then pay. The advance payment can be made by the trading platform or a third party, or the agreement may stipulate that the items must be obtained first and then paid.

[0086] The risk in this embodiment may be abnormal behavior in the transaction process. For ease of understanding, some abnormal examples are given in this embodiment for illustration, such as fraud, cashing out, etc., but are not limited to these contents. The purpose is to ensure the security of transactions and maintain a good network environment.

[0087] like Figure 1 As shown, Figure 1 It is a structural diagram of a risk identification system provided by the present application, which includes: a risk management platform 101 and a risk identification platform 102.

[0088] The risk management platform 101 is used to receive risk labeling data for risk labeling of transaction data fed back by the risk identification platform, and distribute the risk labeling data as a risk identification task, initiate a risk identification request to the risk identification platform, and obtain the identification review conclusion corresponding to the risk identification request fed back by the risk identification platform.

[0089] The risk identification platform 102 is used to perform risk labeling on the transaction data and determine the risk labeling data of the transaction data; in response to the risk identification request initiated by the risk management platform for the risk identification task, the risk labeling data carried in the risk identification request and the transaction data corresponding to the risk labeling data are determined as data to be identified; risk identification is performed on the data to be identified based on a multi-task risk automatic identification model trained separately based on risk types, an identification review conclusion of the data to be identified is determined, and the identification review conclusion is sent to the risk management platform.

[0090] like Figure 2 As shown, Figure 2It is a timing diagram of a risk identification system provided by the present application for performing identification tasks. The risk identification platform 102 needs to perform risk tagging on the transaction data. In this embodiment, in order to avoid the omission, mislabeling, wrong labeling and other operations that occur in manual labeling, and also to improve the efficiency of the labeling operation, the automatic labeling logic of the risk can be adopted. This process requires obtaining the transaction data first. In this embodiment, there are at least two ways to obtain the transaction data, namely, obtaining through a first acquisition link and / or obtaining through a second acquisition link. That is, through a snapshot of the transaction data and / or through the data source of the transaction data, the risk identification platform is driven according to the acquired transaction data to perform risk tagging on the transaction data. In this embodiment, for the risk marking of the transaction data, the marking action can be automatically driven according to the set automatic marking trigger condition, for example: T+15mins is the automatic marking trigger condition, that is: the transaction generation date (T: risk review requirement generation date or transaction date) + 15 minutes to drive the automatic marking operation (wherein, the risk review requirement generation date can be the generation date of the risk review form generated for the transaction data, that is, not all transaction data are reviewed, but risk review forms will be generated selectively. The specific selection method can be implemented in combination with different scenarios. Of course, risk review can also be performed on all transaction data. Therefore, T can also represent the transaction day, that is, the transaction generation date or the transaction generation date. Of course, the meaning represented by T can be determined according to different needs, and is not limited to the example given in this embodiment). Of course, the automatic marking trigger condition can also be set according to the risk identification scenario requirements or the transaction data amount, etc., and is not limited to the above examples. In this embodiment, the first acquisition link is a transaction data snapshot, which can be based on the set selection conditions, and the transaction data snapshot of the transaction participants who meet the risk identification is selected. For example, the selection conditions can be T+1D (risk audit requirement generation date or transaction date + 1 day), T+15mins, etc. The transaction data can be understood as order data, and the transaction data snapshot can be understood as an order data snapshot. The order data includes data that has completed transaction operations and data that cancels transaction operations. The second acquisition link is the data source of the transaction data, which can be based on the set selection conditions, and the transaction data of the transaction participants who meet the risk identification is integrated. For example, the transaction data of the buyer and seller can be integrated as the data source of the transaction data. The update of the data source can also be based on the set update conditions, such as T+15mins; so as to ensure the real-time nature of the transaction data in the data source. In this embodiment, the transaction data snapshot and the data source can be stored in a database or a data warehouse, and the specific storage form and method are not limited. In this embodiment, the integration of the transaction data can be carried out on the risk management platform, and the integrated data can be stored in a database or a data warehouse, or a server.

[0091] According to the above, the risk identification platform needs to mark the transaction data for risks. In this embodiment, automatic marking is used. Therefore, the automatic marking logic can be started according to the set automatic marking trigger conditions. The automatic marking trigger conditions can be set in combination with the specific risk identification scenario requirements, or they can be set according to the conventional requirements of normal risk identification. In this embodiment, T+15mins is used as an example to describe the automatic marking trigger condition, but it is not limited to this. When the automatic marking trigger condition is triggered, the transaction data can be obtained through the first acquisition link and / or the second acquisition link, and the automatic marking logic is executed on the transaction data. In this embodiment, the transaction data can be risk marked according to the constructed risk judgment rule set, and the risk marked data corresponding to the transaction data can be determined. Among them, the risk judgment rule set can also be understood as a marking strategy package, which can be determined by data that has historically generated risks or data with potential risks. For example: for buyers, the risk judgment rule can be to identify whether the buyer's shopping habits have changed suddenly, and query the highest transaction mutation score of the buyer in the last 15 days (15 days) through the transaction mutation risk level algorithm model and judge whether it belongs to the risk level. If it belongs to the risk level, output "buyer shopping habits mutation = true", that is, according to the shopping habits related data features identified in the buyer's transaction data, use the corresponding algorithm model to determine whether there is a risk, and if so, perform risk marking. The risk marking data may include: risk description information and risk value, etc. The risk value may include: yes, no, undetermined and other judgment values, and these values ​​have modifiable permissions. It is also possible to determine whether there is a risk by identifying the buyer's delivery address, contact information, transaction IP, etc., that is, the risk judgment rule set for the buyer can include any one or more information in the transaction data such as buyer transaction behavior, buyer transaction attributes, buyer portrait, etc. to construct the risk judgment rule set. Of course, different risk strategies can be adopted for different transaction items corresponding to the transaction data, for example: set targeted risk strategies for the value and attributes of the transaction items. Therefore, the risk judgment rule set in this embodiment is only described by the above example, and is not limited to the above method. For sellers, the risk judgment rules can be used to identify whether the transaction data related to the seller includes high-risk categories, whether it is on the negative list, whether it has been listed on the blacklist, etc. The specific details will not be listed one by one, and they can also be detected through the algorithm model. The risk judgment rule set can be constructed based on relevant data such as transaction scenarios, transaction items, transaction methods, etc. The risk identification platform marks the risk points of the acquired transaction data through the risk judgment rule set and determines the risk-marked data corresponding to the transaction data.The risk marking data is risk marking for the transaction data, so there may be an association between the two, for example: the corresponding transaction data can be found through the risk marking data, or the risk marking data of the transaction data can be obtained through the transaction data.

[0092] In order to facilitate the intuitive capture of risk points and reduce processing operations in demand scenarios such as table lookup, statistics, and analysis, in this embodiment, the risk marking data may also include: a first prompt information (icon). In this embodiment, the first prompt information may be displayed as a "recommended" marking information in the risk judgment value area, and the marking information may be used as recommended data for subsequent judgment.

[0093] like Figure 2 As shown, when the risk identification platform completes the labeling, the risk labeling data is sent to the risk management platform 101, and the risk management platform distributes the risk identification task to the risk labeling data so as to perform the risk identification task later. In this embodiment, the execution of the risk identification task can be initiated by the risk management platform to the risk identification platform through a risk identification request, and the risk identification request carries the acquired risk labeling data and the transaction data corresponding to the risk labeling data. The risk identification request can be generated by the risk management platform based on an operation request such as viewing the risk identification task. When the risk identification request is generated, the risk management platform can obtain the risk labeling data and the transaction data corresponding to the risk labeling data in real time, and send them together to the risk identification platform. The risk identification platform determines the received risk-labeled data and the transaction data corresponding to the risk-labeled data as data to be identified and executes the risk automatic identification logic, that is, through a multi-task risk automatic identification model trained separately based on risk type, the data to be identified is identified, the identification review conclusion of the data to be identified is determined, and the identification review conclusion is sent to the risk management platform. In this embodiment, the multi-task risk automatic identification model can be a sub-model trained separately for different risk types, and a multi-task risk automatic identification model is obtained after fitting multiple sub-models.

[0094] like Figure 3 As shown, Figure 3It is a schematic diagram of an embodiment of the training process of a multi-task risk automatic discrimination model in a risk discrimination system provided by the present application. In this embodiment, the risk type targeted by the model training can be determined in combination with a specific risk discrimination scenario, which may include a first risk type, a second risk type, a third risk type, etc., and the three risk types are trained separately. The model can be a binary classification model, for example: a decision tree model. Specifically, the training can be to train the first classification sub-model according to the first risk type to determine the target first classification sub-model, train the second classification sub-model according to the second risk type to determine the target second classification sub-model; train the third classification sub-model according to the third risk type to determine the target third classification sub-model; fit the target first classification sub-model, the target second classification sub-model and the target third classification sub-model into a multi-task risk automatic discrimination model. For different risk types, the corresponding weights of the sub-models can be set according to the risk discrimination scenarios and requirements, and the training data of the sub-models can use historical data or existing data sets. The first risk type, the second risk type and the third risk type in this embodiment can be the types of whether there is risk, whether there is money laundering risk, whether there is fraud risk, and of course other risk types can also be included, such as: false transaction risk, malicious buyer / seller risk, credit risk, etc. There can be a variety of different risk types for different risk identification scenarios. The determination of specific risk types is not limited to the above examples. In the transaction scenario, the risk type can be understood as a potential transaction abnormality data. The transaction abnormality data can be abnormal data that exists when the transaction participants have completed the relevant transaction agreement, or abnormal data that exists in the transaction process, or abnormal data that appears in a certain participant. In this embodiment, risk identification can be performed for abnormal data that exists in the process of first purchase and then pay transaction. The transaction data can be transaction data for a certain transaction type, and of course it can also be all transaction data, which is determined according to the risk identification requirements and is not specifically limited in this embodiment. The transaction participants include buyers and sellers, and can also include third parties. There is no specific limitation on the transaction participants. All those who have an association relationship with the transaction behavior belong to the transaction participants.

[0095] like Figure 4 As shown, Figure 4It is a schematic diagram of an automatic discrimination embodiment in a risk discrimination system provided by the present application. The data to be discriminated will be input into a trained multi-task risk automatic discrimination model to determine the output result, i.e., the discrimination review conclusion. Specifically, the data feature of the risk-labeled data extracted from the data to be discriminated is determined as the first feature; the selected feature of the transaction data extracted from the data to be discriminated is determined as the second feature; the statistical feature of the selected feature of the transaction data extracted from the data to be discriminated is determined as the third feature; the risk portrait data feature of the transaction data extracted from the data to be discriminated is determined as the fourth feature; one or more of the first feature, the second feature, the third feature and the fourth feature are input into the multi-task automatic discrimination model for risk type discrimination; the risk type discrimination result is determined as the discrimination review conclusion of the data to be discriminated. Among them, the data features of the risk-labeled data are, for example: easy-to-cash-out goods, sudden changes in shopping habits, etc. The selected features may refer to the selected features of the transaction data currently undergoing risk identification, such as small-amount split orders, abnormal payment duration, etc. The statistical features of the selected features of the transaction data may refer to the risk ratio of the transaction data in the past 1m (1 month), the number of hits in risk identification, etc. The risk profile data features of the transaction data may be historical risk profile data of the transaction data, such as fraud score, stability score, credit score, etc. The extracted features are input into the multi-task risk automatic discrimination model to determine the input result of the risk type corresponding to the transaction data and the risk discrimination review conclusion. For example: the multi-task risk automatic discrimination model outputs has_risk=0, that is, it is recommended as risk-free, and the corresponding risk discrimination review conclusion is "risk-free"; the multi-task risk automatic discrimination model outputs has_risk=1&risk_detail=CASH OUT, that is, it is recommended as risky, and the risk type is cash-out risk, and the corresponding risk discrimination review conclusion is "confirm cash-out"; the multi-task risk automatic discrimination model outputs has_risk=1&risk_detail=NOTSURE, that is, it is recommended as risky, and the risk type is uncertain; the multi-task risk automatic discrimination model outputs has_risk=-1, that is, it is uncertain whether there is a risk, because the risk type is uncertain and whether there is a risk is also uncertain, so there is no risk discrimination review conclusion.

[0096] In this embodiment, the multi-task risk discrimination model may output a score for a risk type or a probability value for a risk type. By comparing the value of the risk type discrimination review conclusion with a set discrimination threshold, the data to be discriminated that meets the advancement requirements may be advanced to the stage of terminating the processing operation. For example, when the value of the discrimination review conclusion is greater than or equal to the discrimination threshold, the transaction data corresponding to the discrimination review conclusion may be determined as data with a higher discrimination accuracy, and the transaction data may be advanced to the stage of terminating the processing operation to avoid the occurrence of risks and ensure the transaction security of the transaction participants.

[0097] In such Figure 2 As shown, the risk identification platform will send the output identification and audit conclusion to the risk management platform, and the risk management platform can output and display the identification and audit conclusion, for example, it can be output in the viewing page corresponding to the viewing request. In this embodiment, the identification and audit conclusion display area displayed in the viewing page can also display the second prompt information corresponding to the identification and audit conclusion. The second prompt information can be the audit prompt information, for example: the current transaction data is identified as risky, and the risk category is the second risk type (such as fraud risk). In this embodiment, when the identification and audit conclusion is the second risk type (such as fraud risk) and / or the third risk type (such as money laundering risk), the first prompt information (icon) can be displayed in the identification and audit conclusion display area, such as: "Recommended" marking information.

[0098] The above is a description of an embodiment of a risk identification system provided by the present application. In this embodiment, the acquired transaction data is risk-labeled by the risk identification platform, and the risk-labeled data is sent to the risk management platform. The risk management platform initiates a risk identification request to the risk identification platform based on the risk identification task, so that the risk points of the transaction data can be perceived through the risk-labeled data at the first time of invoking the risk identification task, and the risk identification platform determines the risk-labeled data carried in the risk identification request and the transaction data corresponding to the risk-labeled data as data to be identified; according to the multi-task risk automatic identification model trained separately based on the risk type, the data to be identified is risk identified, the identification and review conclusion of the data to be identified is determined, and the identification and review conclusion is sent to the risk management platform, so as to ensure the accuracy of the risk identification of the transaction data and avoid missing or misjudging risk points. In addition, the risk identification conclusion that meets the requirements can be advanced to the end stage of risk identification, thereby improving the efficiency of risk identification.

[0099] It can be understood that the risk identification platform and risk management platform involved in the risk identification system provided in this application can be located on the same execution side to complete corresponding data processing and interaction functions, or can be distributed on different end sides to realize corresponding data processing and interaction, for example: local end and server end.

[0100] Based on the above, the present application also provides a risk identification method, which is described with the risk identification platform as the execution end. Figure 5 As shown, Figure 5 This is a flow chart of a risk identification method provided by the present application, which may include:

[0101] Step S501: In response to a risk identification request initiated for a risk identification task, the risk marking data carried in the risk identification request and the transaction data corresponding to the risk marking data are determined as data to be identified; wherein the risk identification request may be invoked when executing a risk identification task, and the execution of the risk identification task may be related processing operations such as viewing of the risk identification task. In this embodiment, it is also necessary to perform risk marking on the acquired transaction data to determine the risk marking data of the transaction data. The specific implementation process may include: acquiring the transaction data through the first acquisition link and / or the second acquisition link according to the set automatic marking trigger condition; performing risk marking on the transaction data according to the constructed risk judgment rule set, and determining the risk marking data corresponding to the transaction data. In this embodiment, the risk marking data may include: risk description information and risk value, etc. The risk value may include: judgment values ​​such as yes, no, and cannot be determined, and these values ​​have modifiable permissions. It is also possible to determine whether there is a risk by identifying the buyer's delivery address, contact information, transaction IP, etc., that is, the risk judgment rule set for the buyer can include any one or more information in the transaction data such as buyer transaction behavior, buyer transaction attributes, buyer portrait, etc. to construct the risk judgment rule set. Of course, different risk strategies can be adopted for different transaction items corresponding to the transaction data, for example: setting targeted risk strategies for the value and attributes of the transaction items. Therefore, the risk judgment rule set in this embodiment is only illustrated by the above example and is not limited to the above method. For the seller, the risk judgment rule can be to identify whether the transaction data related to the seller includes high-risk categories, whether it is a negative list, whether it has been listed as a blacklist, etc., which will not be listed one by one, and can also be detected by an algorithm model. The risk judgment rule set can be constructed based on relevant data such as transaction scenarios, transaction items, transaction methods, etc. The risk identification platform marks the risk points of the acquired transaction data through the risk judgment rule set to determine the risk marking data corresponding to the transaction data. The risk labeling data is risk labeling for the transaction data, so there can be an association between the two, for example: the corresponding transaction data can be found through the risk labeling data, or the risk labeling data of the transaction data can be obtained through the transaction data. In order to facilitate the intuitive capture of risk points and reduce the processing operations in demand scenarios such as table lookup, statistics, and analysis, the risk labeling data can also include: first prompt information (icon). In this embodiment, the first prompt information can be displayed in the risk judgment value area as "recommended" marking information, which can be used as recommended data for subsequent judgment.

[0102] In this embodiment, according to the set automatic marking trigger condition, the transaction data can be obtained through the first acquisition link and / or the second acquisition link in the following manner:

[0103] According to the set selection conditions, a transaction data snapshot of a transaction participant that meets the risk identification is selected as the first acquisition link, and when the automatic marking trigger condition is met, the transaction data is acquired;

[0104] and / or,

[0105] According to the set selection conditions, the integrated data after integrating the transaction data of the transaction participants that meet the risk judgment is selected as the second acquisition link, and the transaction data is acquired when the automatic marking trigger condition is met.

[0106] Step S502: Perform risk discrimination on the data to be discriminated according to the multi-task risk automatic discrimination model trained separately based on the risk type, and determine the discrimination review conclusion of the data to be discriminated; the specific implementation process may include: training the first classification sub-model according to the first risk type to determine the target first classification sub-model; training the second classification sub-model according to the second risk type to determine the target second classification sub-model; training the third classification sub-model according to the third risk type to determine the target third classification sub-model; fitting the target first classification sub-model, the target second classification sub-model and the target third classification sub-model into a multi-task risk automatic discrimination model. The data to be judged is input into the multi-task risk automatic judgment model to determine the judgment and review conclusion of the data to be judged. The specific implementation process may include: determining the data feature of the risk-labeled data extracted from the data to be judged as the first feature; determining the selected feature of the transaction data extracted from the data to be judged as the second feature; determining the statistical feature of the selected feature of the transaction data extracted from the data to be judged as the third feature; determining the risk portrait data feature of the transaction data extracted from the data to be judged as the fourth feature; inputting one or more of the first feature, the second feature, the third feature and the fourth feature into the multi-task risk automatic judgment model to judge the risk type; and determining the risk type judgment result as the judgment and review conclusion of the data to be judged.

[0107] Step S503: Sending the discrimination audit conclusion, specifically, the risk discrimination platform will send the output discrimination audit conclusion to the risk management platform, and the risk management platform can output and display the discrimination audit conclusion, for example, it can be output in the viewing page corresponding to the viewing request. In this embodiment, the discrimination audit conclusion display area displayed in the viewing page can also display the second prompt information corresponding to the discrimination audit conclusion, and the second prompt information can be the audit prompt information, for example: the current transaction data is identified as having risks, and the risk category is the second risk type (such as fraud risk). In this embodiment, when the discrimination audit conclusion is the second risk type (such as fraud risk) and / or the third risk type (such as money laundering risk), the first prompt information (icon) can be displayed in the discrimination audit conclusion display area, such as: "Recommendation" mark information. According to the comparison between the discrimination audit conclusion and the set discrimination threshold, the data to be discriminated that meets the advancement requirements is advanced to the end processing operation stage. That is, sending the discrimination audit conclusion can include directly advancing the data that meets the recommendation requirements to the end processing operation stage, that is, confirming the risk facts and performing subsequent related processing operations.

[0108] Regarding the description of the above-mentioned risk identification method embodiment, reference can be made to the description of the risk identification platform in the above-mentioned risk identification system. It can be understood that the risk identification method provided in this application is mainly based on the risk identification platform to realize the processing of automatic risk labeling and automatic risk identification. Automatic risk labeling can facilitate the perception of risk information existing in transaction data at the first time when executing risk identification tasks, thereby avoiding the omission of potential risks and facilitating the processing efficiency of subsequent other processing operations related to risk identification tasks. Automatic risk identification obtains the risk identification conclusion about transaction data in the risk identification task through the output results of the multi-task risk automatic identification model, which can reduce the probability of misjudgment, improve the accuracy of risk identification and the efficiency of related processing, and avoid losses to transaction participants due to time-consuming risk identification when transaction data is generated.

[0109] In other words, the core of the risk identification method provided by this application lies in the processing of automated risk labeling and automated risk identification. The identification and review conclusions obtained after the automated risk identification process do not pay attention to how the risk labeling data and the identification and review conclusions are output and displayed, or how the transaction data is sent.

[0110] It can be understood that the risk identification method provided in the present application is an example described with the risk identification platform as the execution side, and the risk identification platform is only the name of the execution side in the execution process of a method or system, and the corresponding execution process is not limited to the risk identification platform, but can also be other executors that can implement the risk identification method process.

[0111] The above is a description of an embodiment of a risk identification method provided by the present application. Corresponding to the embodiment of a risk identification method provided above, the present application also discloses an embodiment of a risk identification device. Figure 6 Since the device embodiment is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The device embodiment described below is only illustrative.

[0112] like Figure 6 As shown, Figure 6 It is a structural diagram of a risk identification device provided in the present application, which includes: a first determination unit 601, a second determination unit 602 and a sending unit 603.

[0113] The first determining unit 601 is configured to respond to a risk identification request initiated for a risk identification task and determine the risk-labeled data carried in the risk identification request and the transaction data corresponding to the risk-labeled data as data to be identified;

[0114] The second determination unit 602 is used to perform risk identification on the data to be identified based on a multi-task risk automatic identification model trained separately based on risk types, and determine the identification and review conclusion of the data to be identified;

[0115] The sending unit 603 is used to send the judgment and review conclusion.

[0116] It also includes: a marking unit, which is used to perform risk marking on the acquired transaction data and determine the risk marking data of the transaction data.

[0117] The marking unit may specifically include: an acquisition subunit and a determination subunit, the acquisition subunit being used to acquire the transaction data through a first acquisition link and / or a second acquisition link according to a set automatic marking trigger condition; the determination subunit being used to perform risk marking on the transaction data according to a constructed risk judgment rule set, and determine the risk marked data corresponding to the transaction data.

[0118] The acquisition subunit may specifically include: according to the set selection conditions, selecting the transaction data snapshot of the transaction participants who meet the risk identification as the first acquisition link, and acquiring the transaction data when the automatic labeling trigger condition is met; and / or, according to the set selection conditions, selecting the integrated data after integrating the transaction data of the transaction participants who meet the risk identification as the second acquisition link, and acquiring the transaction data when the automatic labeling trigger condition is met.

[0119] The specific implementation process of the second determination unit 602 may include: a first determination subunit, a second determination subunit, a third determination subunit, a fitting subunit, and a conclusion determination subunit. The first determination subunit is used to train the first classification submodel according to the first risk type to determine the target first classification submodel; the second determination subunit is used to train the second classification submodel according to the second risk type to determine the target second classification submodel; the third determination subunit is used to train the third classification submodel according to the third risk type to determine the target third classification submodel; the fitting subunit is used to fit the target first classification submodel, the target second classification submodel and the target third classification submodel into a multi-task risk automatic discrimination model; the conclusion determination subunit is used to input the data to be discriminated into the multi-task risk automatic discrimination model to determine the discrimination and audit conclusion of the data to be discriminated.

[0120] The specific implementation process of the conclusion determination subunit may include: a first feature determination subunit, a second feature determination subunit, a third feature determination subunit, a fourth feature determination subunit, a discrimination subunit and a determination subunit. The first feature determination subunit is used to determine the data feature of the risk-tagged data extracted from the data to be discriminated as the first feature; the second feature determination subunit is used to determine the selected feature of the transaction data extracted from the data to be discriminated as the second feature; the third feature determination subunit is used to determine the statistical feature of the selected feature of the transaction data extracted from the data to be discriminated as the third feature; the fourth feature determination subunit is used to determine the risk profile data feature of the transaction data extracted from the data to be discriminated as the fourth feature; the discrimination subunit is used to input one or more features of the first feature, the second feature, the third feature and the fourth feature into the multi-task risk automatic discrimination model for risk type discrimination; the determination subunit is used to determine the risk type discrimination result as the discrimination and audit conclusion of the data to be discriminated.

[0121] The specific implementation process of the sending unit 603 may include: a promotion subunit, which is used to compare the judgment and review conclusion with the set judgment threshold value, and promote the data to be judged that meets the promotion requirements to the end processing operation stage.

[0122] The above is a description of an embodiment of a risk identification device provided in the present application. The specific implementation process of the device can refer to the description of the above risk identification system and method, which will not be described in detail here.

[0123] Based on the above content, this application also provides a risk data output method, such as Figure 7 As shown, Figure 7This is a flowchart of a risk data output method provided by the present application. The output method is described by taking the risk management platform in the risk identification system as an example of the execution side. The core of the output method is how to provide transaction data to the risk identification platform, and how to output the acquired risk labeling data and identification and audit conclusions. It does not focus on how the risk labeling data is completed or how the identification and audit conclusions are completed. Regardless of the method in which the risk labeling data and the identification and audit conclusions are completed, they can be realized by the risk data output method. The risk data output method may include:

[0124] Step S701: Distribute the received risk-labeled data as a risk identification task; the distribution in this embodiment may distribute the risk identification task to the task execution side, without limitation.

[0125] Step S702: Generate and send a risk identification request for target transaction data according to a request for viewing the risk identification task, wherein the risk identification request carries the target transaction data and target risk labeling data corresponding to the target transaction data.

[0126] Step S703: receiving a multi-task risk automatic discrimination model trained separately based on risk types, and outputting a discrimination review conclusion corresponding to the target transaction data; the specific implementation process may include:

[0127] Step S703-1: receiving the target transaction data and the target risk labeling data corresponding to the target transaction data as input data of the multi-task risk automatic discrimination model, and outputting a discrimination and review conclusion on the risk type of the target transaction data;

[0128] Step S703-2: The judgment and review conclusion is stored and displayed on the viewing page corresponding to the viewing request.

[0129] It may also include:

[0130] According to the received risk labeling data, adding first prompt information corresponding to the risk labeling data;

[0131] The risk marking data and the first prompt information are output.

[0132] It may also include:

[0133] The discrimination and audit conclusion display area displayed in the viewing page displays second prompt information corresponding to the discrimination and audit conclusion.

[0134] The above-mentioned risk data output method can be combined with the description of the above-mentioned risk identification system and risk identification method embodiments, which will not be described in detail here.

[0135] The above is a specific description of an embodiment of a risk data output method provided by the present application. Corresponding to the embodiment of a risk data output method provided above, the present application also discloses an embodiment of a risk data output device, please refer to Figure 8 Since the device embodiment is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The device embodiment described below is only illustrative.

[0136] like Figure 8 As shown, Figure 8 It is a structural diagram of a risk data output device provided by the present application. The device embodiment includes: a distribution unit 801, a generation unit 802 and a receiving unit 803.

[0137] The distribution unit 801 is used to distribute the received risk labeling data as a risk identification task;

[0138] The generating unit 802 is used to generate and send a risk identification request for target transaction data according to a request for viewing the risk identification task, wherein the risk identification request carries the target transaction data and target risk labeling data corresponding to the target transaction data;

[0139] The receiving unit 803 is used to receive the discrimination and review conclusion corresponding to the target transaction data output by the multi-task risk automatic discrimination model trained separately based on the risk type, and specifically includes: a receiving subunit and a display subunit; the receiving subunit is used to receive the discrimination and review conclusion on the risk type of the target transaction data output by using the target transaction data and the target risk marking data corresponding to the target transaction data as input data of the multi-task risk automatic discrimination model; the display subunit is used to store the discrimination and review conclusion and display it on the viewing page corresponding to the viewing request.

[0140] The system further comprises: an adding unit and a label output unit, wherein the adding unit is used to add first prompt information corresponding to the risk labeling data according to the received risk labeling data; and the label output unit is used to output the risk labeling data and the first prompt information. The system further comprises: a display unit, which is used to display second prompt information corresponding to the judgment and review conclusion in the judgment and review conclusion display area displayed on the viewing page.

[0141] The above is a description of a risk data output method provided in the present application. For the specific implementation process of the risk data output method, please refer to the above-mentioned risk identification system, risk identification method and risk data output method related content, which will not be repeated here.

[0142] Based on the above, the present application also provides a computer storage medium for storing data generated by a network platform and a program for processing the data generated by the network platform;

[0143] When the program is read and executed, it executes the logical steps executed in the above-mentioned risk identification system, or executes the steps of the above-mentioned risk identification method, or executes the steps of the above-mentioned risk data output method.

[0144] Based on the above, the present application also provides an electronic device, such as Fig. 9 As shown, Fig. 9 : is a schematic diagram of the structure of an electronic device provided by the present application, and the electronic device may include:

[0145] Processor 901;

[0146] Memory 902 is used to store a program for processing data generated by a network platform. When the program is read and executed by the processor, it executes the logical steps executed in the above-mentioned risk identification system, or executes the steps of the above-mentioned risk identification method, or executes the steps of the above-mentioned risk data output method.

[0147] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0148] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0149] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0150] 1. Computer-readable media includes permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include non-transitory media such as modulated data signals and carrier waves.

[0151] 2. Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems or computer program products. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0152] Although the present application is disclosed as above in the form of a preferred embodiment, it is not intended to limit the present application. Any technical personnel in this field may make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims of the present application.

Claims

1. A risk identification system, characterized in that: include: Risk management platform and risk identification platform; The risk management platform is used to receive risk-labeled data for risk-labeling transaction data fed back by the risk identification platform, and to distribute the risk-labeled data as a risk identification task, initiate a risk identification request to the risk identification platform, and obtain the identification review conclusion corresponding to the risk identification request fed back by the risk identification platform; The risk identification platform is used to perform risk tagging on the transaction data and determine the risk tagging data of the transaction data; and for responding to the risk identification request initiated by the risk management platform for the risk identification task, determining the risk-labeled data in the risk identification request and the transaction data corresponding to the risk-labeled data as data to be identified; A multi-task risk automatic discrimination model trained separately according to risk types is used to perform risk discrimination on the data to be discriminated, determine the discrimination and review conclusion of the data to be discriminated, and send the discrimination and review conclusion to the risk management platform.

2. The risk identification system according to claim 1, characterized in that: The risk identification platform is used to obtain the transaction data through the first acquisition link and / or the second acquisition link according to the set automatic labeling trigger conditions; and to risk label the transaction data according to the constructed risk judgment rule set to determine the risk labeled data corresponding to the transaction data.

3. The risk identification system according to claim 1, characterized in that: The risk identification platform is used to train the first classification sub-model according to the first risk type to determine the target first classification sub-model; Used to train the second classification sub-model according to the second risk type and determine the target second classification sub-model; and for training the third classification sub-model according to the third risk type to determine a target third classification sub-model; The judgment and audit conclusion of the data to be judged is determined by inputting the data to be judged into a multi-task risk automatic judgment model fitted according to the first classification sub-model, the second classification sub-model and the third classification sub-model.

4. The risk identification system according to claim 1, characterized in that: The risk management platform also includes: It is used to screen the target discrimination audit conclusions that meet the discrimination requirements based on the obtained discrimination audit conclusions, and push the target discrimination audit conclusions and the transaction data corresponding to the target discrimination audit conclusions as advancement data to the data processing stage to perform corresponding processing operations.

5. The risk identification system according to claim 1, characterized in that: The risk management platform is used to determine the data feature of the risk-labeled data extracted from the data to be identified as the first feature; for determining the selected features of the transaction data in the extracted data to be identified as the second feature; for determining the statistical features of the selected features of the transaction data in the extracted data to be identified as the third feature; and used to determine the risk profile data feature of the transaction data extracted from the data to be identified as a fourth feature; One or more of the first feature, the second feature, the third feature and the fourth feature are input into the automatic discrimination model to perform risk type discrimination, and the risk type discrimination result is determined as the discrimination review conclusion of the data to be discriminated.

6. A risk identification method, characterized in that: include: In response to a risk identification request initiated for a risk identification task, determining the risk-labeled data in the risk identification request and the transaction data corresponding to the risk-labeled data as data to be identified; According to a multi-task risk automatic discrimination model trained separately based on risk types, risk discrimination is performed on the data to be discriminated, and a discrimination review conclusion of the data to be discriminated is determined; Send the judgment and review conclusion.

7. The risk identification method according to claim 6, characterized in that: Also includes: The acquired transaction data is risk-labeled to determine risk-labeled data of the transaction data.

8. The risk identification method according to claim 7, characterized in that: The step of performing risk tagging on the acquired transaction data and determining risk tagging data of the transaction data includes: According to the set automatic marking trigger condition, the transaction data is acquired through the first acquisition link and / or the second acquisition link; The transaction data is risk-labeled according to the constructed risk judgment rule set, and the risk-labeled data corresponding to the transaction data is determined.

9. The risk identification method according to claim 8, characterized in that: The acquiring of transaction data through the first acquisition link and / or the second acquisition link according to the set automatic marking trigger condition includes: According to the set selection conditions, a transaction data snapshot of a transaction participant that meets the risk identification is selected as the first acquisition link, and when the automatic marking trigger condition is met, the transaction data is acquired; and / or, According to the set selection conditions, the integrated data after integrating the transaction data of the transaction participants that meet the risk judgment is selected as the second acquisition link, and the transaction data is acquired when the automatic marking trigger condition is met.

10. The risk identification method according to claim 6, characterized in that: The method of performing risk identification on the data to be identified based on the multi-task risk automatic identification model trained separately based on the risk type and determining the identification and review conclusion of the data to be identified includes: Training the first classification sub-model according to the first risk type to determine a target first classification sub-model; Training the second classification sub-model according to the second risk type to determine a target second classification sub-model; The third classification sub-model is trained according to the third risk type to determine a target third classification sub-model; Fitting the target first classification sub-model, the target second classification sub-model and the target third classification sub-model into a multi-task risk automatic discrimination model; The data to be judged is input into the multi-task risk automatic judgment model to determine the judgment and audit conclusion of the data to be judged.

11. The risk identification method according to claim 10, characterized in that: The step of inputting the data to be identified into the multi-task risk automatic identification model to determine the identification and review conclusion of the data to be identified includes: Determine the data feature of the risk-labeled data extracted from the data to be identified as the first feature; Determine the selected feature of the transaction data in the extracted data to be identified as the second feature; Determine the statistical features of the selected features of the transaction data in the extracted data to be identified as the third feature; Determine the risk profile data feature of the transaction data extracted from the data to be identified as the fourth feature; Inputting one or more of the first feature, the second feature, the third feature and the fourth feature into the multi-task risk automatic identification model to identify the risk type; The risk type identification result is determined as the identification and review conclusion of the data to be identified.

12. The risk identification method according to claim 6, characterized in that: The sending of the determination and review conclusion comprises: According to the comparison between the judgment and audit conclusion and the set judgment threshold, the data to be judged that meets the advancement requirements is advanced to the end processing operation stage.

13. A risk data output method, characterized in that: include: Distribute the received risk-labeled data as risk identification tasks; Generate and send a risk identification request for target transaction data according to a request for viewing the risk identification task, wherein the risk identification request carries the target transaction data and target risk labeling data corresponding to the target transaction data; A multi-task risk automatic discrimination model trained separately based on risk types is received, and a discrimination and audit conclusion corresponding to the target transaction data is output.

14. The risk data output method according to claim 13, characterized in that: The receiving of the multi-task risk automatic discrimination model trained separately based on the risk type, and the output of the discrimination and review conclusion corresponding to the target transaction data, includes: receiving the target transaction data and the target risk labeling data corresponding to the target transaction data as input data of the multi-task risk automatic discrimination model, and outputting a discrimination and review conclusion on the risk type of the target transaction data; The judgment and review conclusion is stored and displayed on the viewing page corresponding to the viewing request.

15. A computer storage medium for storing data generated by a network platform and a program for processing the data generated by the network platform; When the program is read and executed, it executes the logical steps executed in any risk identification system as claimed in claim 1 to claim 5, or executes the steps of any risk identification method as claimed in claim 6 to claim 12, or executes the steps of the risk data output method as claimed in claim 13 or 14.

16. An electronic device, comprising: processor; A memory for storing a program for processing data generated by a network platform, wherein when the program is read and executed by the processor, the program executes the logical steps executed in any one of the risk identification systems of claims 1 to 5, or executes the steps of any one of the risk identification methods of claims 6 to 12, or executes the steps of the risk data output method of claim 13 or claim 14.

Images