WO2020114110A1 - 商户风险防控方法及装置 - Google Patents

商户风险防控方法及装置 Download PDF

Info

Publication number
WO2020114110A1
WO2020114110A1 PCT/CN2019/112126 CN2019112126W WO2020114110A1 WO 2020114110 A1 WO2020114110 A1 WO 2020114110A1 CN 2019112126 W CN2019112126 W CN 2019112126W WO 2020114110 A1 WO2020114110 A1 WO 2020114110A1
Authority
WO
WIPO (PCT)
Prior art keywords
target
merchant
control
category
indicator
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2019/112126
Other languages
English (en)
French (fr)
Inventor
郑霖
陈帅
程羽
陈弢
聂茜倩
朱江
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alibaba Group Holding Ltd
Original Assignee
Alibaba Group Holding Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Alibaba Group Holding Ltd filed Critical Alibaba Group Holding Ltd
Publication of WO2020114110A1 publication Critical patent/WO2020114110A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/21Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/24Classification techniques
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063Operations research, analysis or management
    • G06Q10/0635Risk analysis of enterprise or organisation activities
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063Operations research, analysis or management
    • G06Q10/0639Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393Score-carding, benchmarking or key performance indicator [KPI] analysis
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/018Certifying business or products
    • G06Q30/0185Product, service or business identity fraud

Definitions

  • This application relates to the field of computer technology, in particular to a method and device for merchant risk prevention and control.
  • the embodiments of the present application provide a method and device for merchant risk prevention and control, aiming to perform abnormal detection on indicators related to merchant risk prevention and control, and whenever an abnormal index of the indicator data is detected, based on the abnormal situation of the index Merchant risk prevention and control, to realize early identification of risks before merchant risk confirmation, and improve the timeliness of merchant risk prevention and control.
  • the embodiments of the present application provide a method for merchant risk prevention and control, including:
  • the target indicator data of the merchant is abnormal, determine the target abnormal parameter corresponding to the target indicator data;
  • the merchant risk prevention and control method provided in the first aspect of the present application determines at least one linkage category based on the target dimension and the target category, including:
  • the at least one linkage category associated with the target category is obtained.
  • the merchant risk prevention and control method provided in the first aspect of the present application based on the target abnormal parameter and the at least one linkage category, performing merchant risk prevention and control includes:
  • the merchant risk prevention and control method provided in the first aspect of the present application based on the target abnormal parameter, the at least one linkage category, and the first indicator set, performing merchant risk prevention and control includes:
  • each abnormal parameter in the first parameter set corresponds one-to-one with index data of each index in the first index set
  • the first index set and the first parameter set, a second index set and a second parameter set are determined, each abnormal parameter in the second parameter set and the second index set
  • the indicator data of each indicator corresponds to each other;
  • the merchant risk prevention and control method provided in the first aspect of the present application, based on the at least one linkage category, determines an abnormal cause that causes abnormality in the target indicator data, including:
  • the cause of the abnormality is determined according to the target linkage category.
  • the merchant risk prevention and control method provided in the first aspect of the present application based on the target abnormal parameter, the first indicator set and the first parameter set, determining the second indicator set and the second parameter set includes: :
  • the second indicator set is constructed based on indicators corresponding to each abnormal parameter in the second parameter set.
  • the merchant risk prevention and control method provided in the first aspect of the present application based on the abnormal reason, the second indicator set, and the second parameter set, performing merchant risk prevention and control includes:
  • the merchant risk prevention and control method provided in the first aspect of the present application, based on the risk prevention and control indicators and the risk prevention and control parameters, carries out merchant risk prevention and control, including:
  • the merchant risk prevention and control method provided in the first aspect of the present application before performing merchant risk prevention and control, the method further includes:
  • the merchant risk prevention and control method provided in the first aspect of the present application, based on the risk prevention and control indicators and the risk prevention and control parameters, carries out merchant risk prevention and control, including:
  • the initial model is trained based on the first training sample to obtain the first merchant risk identification model
  • the missing indicator is included in the historical risk indicator
  • the first training sample includes the risk prevention and control indicator, the risk prevention and control parameter, the missing indicator and the missing abnormal parameter corresponding to the missing indicator .
  • the merchant risk prevention and control method provided in the first aspect of the present application, based on the first training sample, trains the initial model to obtain the first merchant risk identification model, including:
  • the merchant risk prevention and control method provided in the first aspect of the present application based on the risk prevention and control indicators and the risk prevention and control parameters, performing risk prevention and control on the merchant includes:
  • the initial model is trained to obtain a second merchant risk identification model
  • the missing indicator is included in the historical risk indicator
  • the second training sample includes the second risk prevention and control indicator and the second risk prevention and control parameter.
  • the merchant risk prevention and control method provided in the first aspect of the present application, based on the second training sample, trains the initial model to obtain a second merchant risk identification model, including:
  • the method further includes:
  • the merchant risk prevention and control method provided in the first aspect of the present application, based on a time series anomaly detection model, detects whether the merchant's target indicator data is abnormal, including:
  • the detection result indicates that the target index data is normal.
  • the step of determining the timing outliers includes:
  • the merchant risk prevention and control method provided in the first aspect of the present application includes at least one of the following:
  • the dimension includes at least one of a geographic dimension, a merchant type dimension, and a payment interface dimension;
  • the category includes at least one of a merchant registration category and a merchant operation category.
  • the merchant risk prevention and control method provided in the first aspect of the present application includes at least one of the following:
  • the merchant registration category includes at least one of the following indicators: the nature of the merchant, maintenance records, business content, and industry;
  • the merchant operation category includes at least one of the following indicators: transaction amount, number of complaints, product quality, service quality, and payment interface usage.
  • an embodiment of the present application further provides a merchant risk prevention and control device, including:
  • the first determining module is used to determine the target abnormal parameter corresponding to the target index data when the target index data of the merchant is abnormal;
  • An obtaining module configured to obtain the target category to which the target indicator corresponding to the target indicator data belongs and the target dimension to which the target category belongs based on the correspondence between the indicators, categories and dimensions;
  • a second determination module configured to determine at least one linkage category based on the target dimension and the target category, and the at least one linkage category includes an indicator that causes abnormality of the target indicator data;
  • the processing module is configured to perform merchant risk prevention and control based on the target abnormal parameter and the at least one linkage category.
  • an embodiment of the present application further provides an electronic device, including:
  • a memory arranged to store computer-executable instructions, which when executed, causes the processor to perform the following operations:
  • the target indicator data of the merchant is abnormal, determine the target abnormal parameter corresponding to the target indicator data;
  • an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores one or more programs, and when the one or more programs are electronically including a plurality of application programs When the device is executed, the electronic device is caused to perform the following operations:
  • the target indicator data of the merchant is abnormal, determine the target abnormal parameter corresponding to the target indicator data;
  • indicators for merchant risk prevention and control are comprehensively set from different dimensions, and the indicators are classified to obtain corresponding categories, and the correspondence between indicators, categories, and dimensions is established to detect the existence of indicator data.
  • indicators for merchant risk prevention and control are comprehensively set from different dimensions, and the indicators are classified to obtain corresponding categories, and the correspondence between indicators, categories, and dimensions is established to detect the existence of indicator data.
  • For abnormal target indicators based on the corresponding relationship, all linkage categories to which the index that caused the abnormal target data belonged are obtained, and then merchant risk prevention and control are carried out based on the obtained linkage categories and target abnormal parameters that characterize the abnormality of the target index data.
  • the merchant risk prevention and control will be carried out in time based on the anomaly of the indicator, so that the risk of the merchant before the risk confirmation is identified in advance and improved Timeliness of merchant risk prevention and control.
  • FIG. 1 is a schematic flowchart of a merchant risk prevention and control method provided in an embodiment of this application;
  • FIG. 2 is a schematic diagram of the flow of merchant risk prevention and control information provided in an embodiment of the present application.
  • FIG. 3 is a schematic structural diagram of a merchant risk prevention and control device provided in an embodiment of this application.
  • FIG. 4 is a schematic structural diagram of an electronic device provided in an embodiment of this application.
  • the prevention and control scheme based on the merchant strategy requires the strategy operator to extract the risk characteristics through manual trial and analysis of the case, and then configure the online prevention and control strategy based on the extracted risk characteristics to allow the merchant to access beforehand and in the event Conduct layer-by-layer prevention and control in transactions and ex-post management.
  • the prevention and control scheme based on the risk detection model needs to analyze and design the corresponding risk characteristics for specific merchant risks, such as fraud, gambling, cash out and other illegal and prohibited behaviors, and then establish a risk detection model based on the risk characteristics. Furthermore, based on the risk value output by the risk detection model, risk prevention and control can be carried out in combination with corresponding prevention and control strategies.
  • the prevention and control plan based on user complaints when implementing the prevention and control plan, mainly review and determine the merchants with user complaints, punish the corresponding risk merchants after verifying the risks, and punish the risk merchants together Of other merchants.
  • the embodiments of the present application provide a merchant risk prevention and control solution, which can identify merchant risks in advance, realize the dynamic prevention and control of merchant risks, improve the timeliness of merchant risk prevention and control, and have strong risk dynamics Adversarial ability, suitable for different types of merchant risks, good versatility.
  • an embodiment of the present application provides a method for merchant risk prevention and control.
  • the method may include the following steps:
  • S103 Acquire the target category to which the target indicator corresponding to the target indicator data belongs, and the target dimension to which the target category belongs, based on the correspondence between the indexes, categories, and dimensions.
  • the following embodiments can be used to realize anomaly detection of the merchant's target indicator data and determine target anomaly parameters:
  • time series anomaly detection models include ARIMA model (Autoregressive Integrated Moving Average model), HOLT-WINTERS model (exponential smoothing model) and LSTM (Long Short-Term Memory, long-short-term memory network) model
  • ARIMA model Autoregressive Integrated Moving Average model
  • HOLT-WINTERS model Exponential smoothing model
  • LSTM Long Short-Term Memory, long-short-term memory network
  • the step of detecting whether the target indicator data of the merchant is abnormal may include:
  • the detection result indicates that the target index data is normal.
  • the step of determining the timing outliers may include:
  • the target index data can reflect the value of the target index as a parameter variable, and the target abnormal parameter can reflect the abnormal degree when the target index data is abnormal.
  • the target abnormal data corresponding to the target index data can be multiple The average value of time series outliers.
  • the time series measured value can reflect the actual detection value of the target indicator data at a detection time point in the current period
  • the time series prediction value can reflect the target index data according to the normal estimate of the estimated value at the detection time point
  • the time series actual measurement can reflect the gap between the actual measurement value and the normal estimated value of the target indicator data at the detection time point
  • the time difference can measure the target indicator data in the detection Whether the detection value at the moment changes.
  • the values of the above current time period and the preset value can be determined based on the experience of the merchant performing the anomaly detection and the actual situation of the target indicator; for example, the average value of multiple time series anomaly values that can be detected within 3 consecutive days is greater than Or equal to the preset value, it is determined that the detection result indicates that the target index data is abnormal.
  • the dimension may include at least one of a geographic dimension, a merchant type dimension, and a payment interface dimension.
  • the regional dimension from thick to thin can include autonomous regions, provinces, cities, flag counties, and streets.
  • the merchant type dimension from superior to inferior may include: service providers, business development specialists, offline merchants, merchant transactions, etc.
  • the payment interface dimension may include whether to sign a third-party payment interface, whether the third-party payment interface is Alipay, and so on.
  • the category under the dimension may include at least one of a merchant registration category and a merchant operation category.
  • the merchant registration category may include at least one of the following indicators: the nature of the merchant, maintenance records, business content, and industry; the merchant operation category may include at least one of the following indicators: transaction amount, number of complaints, product quality, service quality and Payment interface usage.
  • the nature of merchants can include individual merchants and corporate merchants; dimension records include business development specialists (BD, Business Development) to regularly record the merchant's dimensions, and the maintenance content can include monitoring the merchant's tax payment and monitoring whether the merchant's business license expires Whether the due date is updated, monitor whether the merchant handles customer complaints in a timely manner, etc.
  • BD business development specialists
  • the maintenance content can include monitoring the merchant's tax payment and monitoring whether the merchant's business license expires Whether the due date is updated, monitor whether the merchant handles customer complaints in a timely manner, etc.
  • the BD generates corresponding maintenance records after each dimension of the merchant;
  • the business content can refer to the merchandise operated by the merchant when the merchant is registered, for example, the fruit shop’s business content can be It is fruit, and the business content of the vegetable shop can be vegetables;
  • the industry to which the merchant belongs can refer to the industry filled in when the merchant is registered, or can refer to the industry derived from the business content of the merchant;
  • the transaction amount can include the merchant's transaction within the set time The total amount;
  • the number of complaints may include the number of complaints by the merchant in the most recent cycle;
  • the usage of the payment interface may include whether the merchant's third-party payment interface is used by other merchants, and the geographic location information of the third-party payment interface to complete the transaction.
  • anomaly detection can be performed on specific indicators under a specified category in a specified dimension to achieve the capture of target indicator data that has changed in time series during the risk prevention and control process of the merchant; all indicators designed in advance can also be captured Anomaly detection is carried out to capture the target indicator data that has changed in time series during the risk prevention and control process of merchants.
  • the process of anomaly detection may be performed periodically or continuously, and in the case of periodic anomaly detection, the polling period for different indicators may be the same or different.
  • S105 Determine at least one linkage category based on the target dimension and the target category, and the at least one linkage category includes indicators that cause abnormal target indicator data.
  • At least one linkage category is further obtained based on the target dimension and target category, and the at least one linkage category includes data that can cause the target indicator to be abnormal Indicators, so as to find out related abnormal indicators more comprehensively.
  • the step of determining at least one linkage category based on the target dimension and the target category may include:
  • Under at least one linkage dimension obtain at least one linkage category associated with the target category.
  • At least one linkage category can be scrolled up and down to search for the correlation between the dimensions and the associations between the categories and the categories to more comprehensively find relevant abnormal indicators.
  • the relationship between dimensions and dimensions, categories and categories can be recorded and maintained in the correspondence between indicators, categories and dimensions, or the association between dimensions and dimensions, categories and categories can be established and maintained separately.
  • the linkage dimension may refer to a dimension that has an association relationship with the target dimension to which the target index corresponding to the abnormal target index data belongs
  • the linkage category may refer to the association category to the target category to which the target index corresponding to the abnormal target index data belongs to Category
  • each linkage category is subordinate to the corresponding linkage dimension
  • all indicators under each linkage category have indicators that can cause the target indicator data to be abnormal, in other words, each indicator data corresponding to each indicator under each linkage category can cause Indicator data with abnormal target indicator data.
  • merchant risk prevention and control can be performed accordingly, that is, the following steps are performed:
  • S107 Perform merchant risk prevention and control based on target abnormal parameters and at least one linkage category.
  • this step S107 may include:
  • At least one linkage category and the first indicator set carry out merchant risk prevention and control.
  • anomaly detection can be performed on the index data corresponding to the index under the at least one linkage category, and optionally, anomaly detection can be performed on the index data within a specified time period to Improve the degree of association with abnormal target index data, improve the accuracy and reliability of index acquisition in the first index set, and use the above-mentioned similar detection scheme for detecting target index data for abnormality to at least one linkage category
  • Anomaly detection is performed on the index data under, that is, whether the index data corresponding to the index under at least one linkage category is abnormal based on the time-series anomaly detection model, and when the detection result indicates abnormality, it is determined that the index under at least one linkage category corresponds to There is an abnormality in the indicator data of, and at least one indicator corresponding to the at least one indicator data that has an anomaly can be formed into a first indicator set for use in merchant risk prevention and control together with the target anomaly parameter and at least one linkage category.
  • merchant risk prevention and control may specifically include the following steps:
  • the cause of the abnormality of the target indicator data may be determined according to at least one linkage category.
  • the linkage category that the user is more concerned about in the at least one linkage category is determined as the cause of abnormality , which can include the following steps:
  • the target linkage category determine the cause of the abnormality.
  • the cause of the abnormality can be determined based on at least one linkage category in other ways, such as setting a category screening rule to automatically filter out categories that meet the requirements of the rule in at least one linkage category based on the category screening rule Abnormal causes, etc.
  • each abnormal parameter in the first parameter set has a one-to-one correspondence with the index data of each index in the first index set.
  • the abnormal parameter of each index data can be determined accordingly, so that the first parameter set corresponding to the first index set can be obtained, wherein the determination of each abnormal parameter can be Refer to the above embodiment for determining the target abnormal parameter, which will not be repeated here.
  • each abnormal parameter in the second parameter set is one-to-one with the index data of each index in the second index set correspond.
  • the first indicator set and the first parameter set may be converted into the second indicator set and the second parameter set based on the target abnormal parameter, so as to achieve reasonable control of the strength of merchant risk prevention and control, where
  • the second indicator set includes at least one indicator in the first indicator set, and the second parameter set includes at least one abnormal parameter in the first parameter set.
  • the process of determining the second indicator set and the second parameter set may include the following steps:
  • the second index set is constructed based on the indexes corresponding to the abnormal parameters in the second parameter set.
  • the abnormal parameters whose abnormality level indicated in the first parameter set is above the abnormality level corresponding to the target index data can be classified into the second parameter set, that is, when the first parameter set When the difference between the abnormal parameter and the target abnormal parameter is within a certain specified range, the abnormal parameter is classified into the second parameter set, and then the indicators in the first index set corresponding to each abnormal parameter in the second parameter set are classified as The second set of indicators.
  • the steps of merchant risk prevention and control based on the cause of abnormality, the second indicator set and the second parameter set may be performed.
  • Optional may include:
  • the risk prevention and control parameters are obtained, and the risk prevention and control parameters correspond to the risk prevention and control indicators one-to-one;
  • the risk prevention and control index that matches the cause of the abnormality can be obtained in the second index set, and the Obtain the risk prevention and control parameters corresponding to the aforementioned risk prevention and control indicators that match the cause of the abnormality in the two parameter sets, so as to carry out more reasonable merchant risk prevention and control.
  • merchant risk prevention and control can be achieved through the following specific embodiments:
  • a solution for merchant risk prevention and control may include:
  • risk prevention and control strategies include fines for merchants, return of fraud gains, order to suspend business operations for rectification, and order to improve operational qualifications, etc., to prohibit merchants from performing any risk-causing actions.
  • the merchant risk identification model can be trained to update the merchant risk identification model based on the latest indicator abnormality, while detecting the indicator abnormality in a timely manner, and improving the timeliness of merchant risk prevention and control.
  • a step of obtaining historical risk indicators and corresponding historical abnormal parameters needs to be performed, and then the historical risk indicators are compared with the current risk prevention and control indicators to implement corresponding merchant risk prevention and control schemes based on the results of different comparisons, to avoid situations where historical risks are missed, and to achieve more comprehensive merchant risk prevention and control coverage.
  • the merchant risk prevention and control scheme may include the following two aspects:
  • the initial model is trained based on the first training sample to obtain the first merchant risk identification model
  • the missing indicators are included in the historical risk indicators
  • the first training sample includes risk prevention and control indicators, risk prevention and control parameters, missing indicators, and missing abnormal parameters corresponding to the missing indicators.
  • the above initial model may be an untrained original model constructed using the isolated forest iForest algorithm or the symbolic regression (Symbolic Regression) algorithm, or a merchant risk identification model that is being trained based on historical risk indicators and other data. .
  • the process of training the initial model to obtain the first merchant risk identification model may include:
  • the input vector of the initial model is generated
  • the risk identification threshold can be determined based on this, so that the risk identification standard is reliable and accurate.
  • the training set and the verification set during model training can be determined.
  • the number of samples in the training set and the verification set can be flexibly allocated. For example, 70% of the training samples are used as the training set to train the merchant risk identification model, and the remaining 30% of the training samples are used as the verification set to verify the merchant risk identification. Whether the output of the model meets the requirements in order to realize the model effect evaluation.
  • risk prevention and control strategies include fines for merchants, return of fraud gains, order to suspend business operations for rectification, and order to improve operational qualifications, etc., to prohibit merchants from performing any risk-causing actions.
  • the initial model is trained to obtain the second merchant risk identification model
  • the missing indicator is included in the historical risk indicator
  • the second training sample includes the second risk prevention and control indicator and the second risk prevention and control parameter.
  • the latest and comprehensive risk prevention and control indicators may be considered.
  • the corresponding risk prevention and control parameters are used as training samples for the second merchant risk identification model, so that the latest second merchant risk identification model based on this training can identify historical and latest merchant risks, and has good timeliness.
  • the above initial model may be an untrained original model constructed using the isolated forest iForest algorithm or the symbolic regression (Symbolic Regression) algorithm, or a merchant risk identification model that is being trained based on historical risk indicators and other data. .
  • the process of training the initial model to obtain the second merchant risk identification model may include:
  • the training set and the verification set during model training can be determined.
  • the number of samples in the training set and the verification set can be flexibly allocated. For example, 70% of the training samples are used as the training set to train the second merchant risk identification model, and the remaining 30% of the training samples are used as the verification set. Whether the output of the two merchants' risk identification model meets the requirements in order to realize the model effect evaluation
  • the merchant risk identification model training can be directly performed based on the risk prevention and control indicators and their corresponding risk prevention and control parameters.
  • risk prevention and control strategies include fines for merchants, return of fraud gains, order to suspend business operations for rectification, and order to improve operational qualifications, etc., to prohibit merchants from performing any risk-causing actions.
  • indicators for merchant risk prevention and control are comprehensively set from different dimensions, and the indicators are classified to obtain corresponding categories, and the correspondence between indicators, categories, and dimensions is established to
  • an abnormal target indicator of the indicator data is detected, based on the corresponding relationship, all linkage categories to which the indicator that caused the abnormal target indicator data belongs to are obtained, and then the merchant is conducted based on the obtained linkage category and the target abnormal parameter characterizing the abnormality of the target indicator data.
  • the merchant risk prevention and control will be carried out in time based on the anomaly of the indicator, so that the risk of the merchant before the risk confirmation is identified in advance and improved Timeliness of merchant risk prevention and control.
  • the merchant risk prevention and control method described in the embodiments of the present application can also be displayed through the information flow shown in FIG. 2.
  • Abnormality detection that is, abnormality detection
  • monitoring that is, monitoring where abnormality occurs.
  • Reasonable dimension division and indicator design can be adopted, such as dividing dimensions according to contract information, merchant type, payment institution, etc., and separately designing indicators in the account opening, transaction, operation and other aspects of the merchant, and classifying the indicators, and testing over time
  • the time series change of the indicator in each dimension or specific dimension, according to the difference between the detected time series actual value and the time series forecast value to determine whether the indicator has changed, if there is a change, the target change category to which the target change indicator belongs is output
  • the abnormality dimension and the target abnormality degree corresponding to the target abnormality index that is, the target abnormality parameter).
  • the degree of abnormality of the indicator can be calculated as follows:
  • Degree of change (time series actual value-time series predicted value) / time series predicted value.
  • the identified historical risk indicators and the transaction degree can be considered to avoid missing risks.
  • the above-mentioned risk identification model is output to the risk prevention and control platform for model deployment to realize the assessment of the true results of the model, and the risk prevention and control strategy can be matched with the results to quickly prevent and control new risks.
  • the city dimension as the main dimension to detect transaction abnormalities or anomaly detection for merchants
  • the data corresponding to the indicator B of city A is perceived to be abnormal, compared with the normal time-series forecast value, the transaction has occurred for many consecutive days, and the average transaction
  • the degree reaches a certain value
  • the indicator category C to which the indicator B belongs, the average transaction degree corresponding to the data of the indicator B, and the city dimension A are output.
  • the invention does not target any merchant's figurative risk. Through comprehensive index design, it can detect merchant anomalies caused by different risks.
  • an embodiment of the present application further provides a merchant risk prevention and control device, which may include:
  • the first determining module 301 is configured to determine target abnormal parameters corresponding to the target index data when the target index data of the merchant is abnormal;
  • the obtaining module 303 is used to obtain the target category to which the target indicator corresponding to the target indicator data belongs and the target dimension to which the target category belongs based on the correspondence between the indicators, categories and dimensions;
  • the second determining module 305 is configured to determine at least one linkage category based on the target dimension and the target category, and the at least one linkage category includes indicators that cause abnormal target indicator data;
  • the processing module 307 is configured to perform merchant risk prevention and control based on the target abnormal parameter and at least one linkage category.
  • the merchant risk prevention and control device shown in FIG. 3 can implement various steps of the merchant risk prevention and control method described in FIG. 1, and the relevant explanations on the merchant risk prevention and control methods in the foregoing embodiments are applicable to merchant risk prevention and control The device will not be repeated here.
  • the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory.
  • the memory may include a memory, such as a high-speed random access memory (Random-Access Memory, RAM), or may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.
  • RAM Random-Access Memory
  • non-volatile memory such as at least one disk memory.
  • the electronic device may also include hardware required for other services.
  • the processor, network interface and memory can be connected to each other through an internal bus, which can be an ISA (Industry Standard Architecture, Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry, Standard Architecture, extended industry standard structure) bus, etc.
  • the bus can be divided into an address bus, a data bus, and a control bus. For ease of representation, only one bidirectional arrow is used in FIG. 4, but it does not mean that there is only one bus or one type of bus.
  • the program may include program code, and the program code includes a computer operation instruction.
  • the memory may include memory and non-volatile memory, and provide instructions and data to the processor.
  • the processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to form a merchant risk prevention and control device at a logical level.
  • the processor executes the programs stored in the memory and is specifically used to perform the following operations:
  • the target indicator data of the merchant is abnormal, determine the target abnormal parameter corresponding to the target indicator data;
  • At least one linkage category includes indicators that cause abnormal target indicator data
  • the above method performed by the merchant risk prevention and control device disclosed in the foregoing corresponding embodiments of the present application may be applied to or implemented by a processor.
  • the processor may be an integrated circuit chip with signal processing capabilities.
  • each step of the above method may be completed by an integrated logic circuit of hardware in the processor or instructions in the form of software.
  • the aforementioned processor may be a general-purpose processor, including a central processor (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; it may also be a digital signal processor (Digital Signal Processor, DSP), dedicated integration Circuit (Application Specific Integrated Circuit, ASIC), field programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.
  • the methods, steps, and logical block diagrams disclosed in the embodiments of the present application may be implemented or executed.
  • the general-purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
  • the steps of the method disclosed in conjunction with the embodiments of the present application may be directly embodied and executed by a hardware decoding processor, or may be executed and completed by a combination of hardware and software modules in the decoding processor.
  • the software module may be located in a mature storage medium in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, or electrically erasable programmable memory, and registers.
  • the storage medium is located in the memory.
  • the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.
  • the electronic device may also execute the method performed by the corresponding merchant risk prevention and control device, and implement the functions of the merchant risk prevention and control device in the aforementioned corresponding embodiments, and the embodiments of the present application will not be repeated here.
  • An embodiment of the present application also provides a computer-readable storage medium that stores one or more programs, and the one or more programs include instructions, which are executed by an electronic device that includes multiple application programs At this time, it can enable the electronic device to execute the method executed by the merchant risk prevention and control device in the embodiment shown in FIG. 1, and is specifically used to execute:
  • the target indicator data of the merchant is abnormal, determine the target abnormal parameter corresponding to the target indicator data;
  • At least one linkage category includes indicators that cause abnormal target indicator data
  • the embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.
  • computer usable storage media including but not limited to disk storage, CD-ROM, optical storage, etc.
  • each flow and/or block in the flowchart and/or block diagram and a combination of the flow and/or block in the flowchart and/or block diagram may be implemented by computer program instructions.
  • These computer program instructions can be provided to the processor of a general-purpose computer, special-purpose computer, embedded processing machine, or other programmable data processing device to produce a machine that enables the generation of instructions executed by the processor of the computer or other programmable data processing device
  • These computer program instructions may also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, the instructions
  • the device implements the functions specified in one block or multiple blocks of the flowchart one flow or multiple flows and/or block diagrams.
  • These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operating steps are performed on the computer or other programmable device to produce computer-implemented processing, which is executed on the computer or other programmable device
  • the instructions provide steps for implementing the functions specified in one block or multiple blocks of the flowchart one flow or multiple flows and/or block diagrams.
  • the computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
  • processors CPUs
  • input/output interfaces network interfaces
  • memory volatile and non-volatile memory
  • Memory may include non-permanent memory, random access memory (RAM) and/or non-volatile memory in computer-readable media, such as read only memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.
  • RAM random access memory
  • ROM read only memory
  • flash RAM flash random access memory
  • Computer-readable media including permanent and non-permanent, removable and non-removable media, can store information by any method or technology.
  • the information may be computer readable instructions, data structures, modules of programs, or other data.
  • Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, read-only compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, Magnetic tape cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media can be used to store information that can be accessed by computing devices.
  • computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.
  • the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.
  • computer usable storage media including but not limited to disk storage, CD-ROM, optical storage, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Educational Administration (AREA)
  • General Business, Economics & Management (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Game Theory and Decision Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Evolutionary Computation (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种商户风险防控方法,包括:在检测到商户的目标指标数据异常的情况下,确定目标指标数据对应的目标异常参数(S101);基于指标、类别和维度之间的对应关系,获取目标指标数据对应的目标指标所属的目标类别,及目标类别所属的目标维度(S103);基于目标维度和目标类别,确定至少一个联动类别,至少一个联动类别下包括引起目标指标数据异常的指标(S105);基于目标异常参数和至少一个联动类别,进行商户风险防控(S107)。该方法通过对商户风险防控相关的指标进行异常检测,并每当检测到指标数据存在异常的指标时,基于指标异常情况及时进行商户风险防控,实现商户风险确认前的风险提前识别,提高商户风险防控的时效性。

Description

商户风险防控方法及装置 技术领域
本申请涉及计算机技术领域,尤其涉及一种商户风险防控方法及装置。
背景技术
随着互联网技术的快速发展,移动支付越来越普及,消费者越来越习惯于使用第三方支付平台进行便捷的移动支付。
对于移动支付而言,选择使用第三方支付平台的用户和商户是最重要的两个主体。随着商户数量级的急剧增长,商户风险发生的机率也随之增大,如何及时有效地对数以万计的商户进行风险防控成为亟待解决的问题。
因此,亟需一种商户风险防控方法,以提高商户风险防控的时效性。
发明内容
本申请实施例提供了一种商户风险防控方法及装置,旨在通过对商户风险防控相关的指标进行异常检测,并每当检测到指标数据存在异常的指标时,基于指标异常情况及时进行商户风险防控,实现商户风险确认前的风险提前识别,提高商户风险防控的时效性。
本申请实施例采用下述技术方案:
第一方面,本申请实施例提供一种商户风险防控方法,包括:
在检测到商户的目标指标数据异常的情况下,确定所述目标指标数据对应的目标异常参数;
基于指标、类别和维度之间的对应关系,获取所述目标指标数据对应的目标指标所属的目标类别,及所述目标类别所属的目标维度;
基于所述目标维度和所述目标类别,确定至少一个联动类别,所述至少一个联动类别下包括引起所述目标指标数据异常的指标;
基于所述目标异常参数和所述至少一个联动类别,进行商户风险防控。
可选的,本申请第一方面提供的商户风险防控方法,基于所述目标维度和所述目标 类别,确定至少一个联动类别,包括:
获取与所述目标维度关联的至少一个联动维度;
在所述至少一个联动维度下,获取与所述目标类别关联的所述至少一个联动类别。
可选的,本申请第一方面提供的商户风险防控方法,基于所述目标异常参数和所述至少一个联动类别,进行商户风险防控,包括:
在所述至少一个联动类别下,获取存在异常的指标数据对应的第一指标集合;
基于所述目标异常参数、所述至少一个联动类别和所述第一指标集合,进行商户风险防控。
可选的,本申请第一方面提供的商户风险防控方法,基于所述目标异常参数、所述至少一个联动类别和所述第一指标集合,进行商户风险防控,包括:
基于所述至少一个联动类别,确定造成所述目标指标数据出现异常的异常原因;
确定所述第一指标集合对应的第一参数集合,所述第一参数集合中各异常参数与所述第一指标集合中各指标的指标数据一一对应;
基于所述目标异常参数、所述第一指标集合和所述第一参数集合,确定第二指标集合和第二参数集合,所述第二参数集合中各异常参数与所述第二指标集合中各指标的指标数据一一对应;
基于所述异常原因、所述第二指标集合和所述第二参数集合,进行商户风险防控。
可选的,本申请第一方面提供的商户风险防控方法,基于所述至少一个联动类别,确定造成所述目标指标数据出现异常的异常原因,包括:
接收类别选择指令;
在所述至少一个联动类别中,确定所述类别选择指令对应的目标联动类别;
根据所述目标联动类别,确定所述异常原因。
可选的,本申请第一方面提供的商户风险防控方法,基于所述目标异常参数、所述第一指标集合和所述第一参数集合,确定第二指标集合和第二参数集合,包括:
确定所述目标异常参数与所述第一参数集合中的各异常参数间的差值;
在所述差值处于预设阈值范围内的情况下,基于所述差值对应的异常参数构建所述第二参数集合;
基于所述第二参数集合中各异常参数对应的指标构建所述第二指标集合。
可选的,本申请第一方面提供的商户风险防控方法,基于所述异常原因、所述第二指标集合和所述第二参数集合,进行商户风险防控,包括:
在所述第二指标集合中,获取风险防控指标,所述风险防控指标与所述异常原因相匹配;
在所述第二参数集合中,获取风险防控参数,所述风险防控参数与所述风险防控指标一一对应;
基于所述风险防控指标和所述风险防控参数,进行商户风险防控。
可选的,本申请第一方面提供的商户风险防控方法,基于所述风险防控指标和所述风险防控参数,进行商户风险防控,包括:
在所述风险防控参数大于异常参数阈值的情况下,获取与所述风险防控指标对应的风险防控策略;
基于所述风险防控策略进行商户风险防控。
可选的,本申请第一方面提供的商户风险防控方法,在进行商户风险防控前,所述方法还包括:
获取历史风险指标及对应的历史异常参数。
可选的,本申请第一方面提供的商户风险防控方法,基于所述风险防控指标和所述风险防控参数,进行商户风险防控,包括:
在所述风险防控指标与所述历史风险指标相比存在遗漏指标的情况下,基于第一训练样本,对初始模型进行训练,得到第一商户风险识别模型;
部署所述第一商户风险识别模型以用于商户风险识别,并输出风险识别结果;
获取与所述风险识别结果对应的风险防控策略,以基于所述风险防控策略进行风险防控;
其中,所述遗漏指标包含在所述历史风险指标中,所述第一训练样本包括所述风险防控指标、所述风险防控参数、所述遗漏指标及所述遗漏指标对应的遗漏异常参数。
可选的,本申请第一方面提供的商户风险防控方法,基于第一训练样本,对初始模型进行训练,得到第一商户风险识别模型,包括:
基于所述风险防控参数和所述遗漏异常参数,确定风险识别阈值;
基于所述风险防控指标和所述遗漏指标,生成所述初始模型的输入向量;
将所述输入向量输入所述初始模型,得到所述初始模型的输出;
根据所述初始模型的输出与所述风险识别阈值之间的差距,调整所述初始模型的参数;
重复以上步骤,直至所述差距满足预设条件,得到所述第一商户风险识别模型。
可选的,本申请第一方面提供的商户风险防控方法,基于所述风险防控指标和所述风险防控参数,对所述商户进行风险防控,包括:
在所述风险防控指标与所述历史风险指标相比不存在遗漏指标的情况下,基于所述第二训练样本,对初始模型进行训练,得到第二商户风险识别模型;
部署所述第二商户风险识别模型以用于商户风险识别,并输出风险识别结果;
获取与所述风险识别结果对应的风险防控策略,以基于所述风险防控策略进行风险防控;
其中,所述遗漏指标包含在所述历史风险指标中,所述第二训练样本包括所述第二风险防控指标和所述第二风险防控参数。
可选的,本申请第一方面提供的商户风险防控方法,基于所述第二训练样本,对初始模型进行训练,得到第二商户风险识别模型,包括:
基于所述风险防控参数,确定风险识别阈值;
基于所述风险防控指标,生成所述初始模型的输入向量;
将所述输入向量输入所述初始模型,得到所述初始模型的输出;
根据所述初始模型的输出与所述风险识别阈值之间的差距,调整所述初始模型的参数;
重复以上步骤,直至所述差距满足预设条件,得到所述第二商户风险识别模型。
可选的,本申请第一方面提供的商户风险防控方法,所述方法还包括:
基于时序异常检测模型,检测所述商户的目标指标数据是否存在异常;
在检测结果指示所述目标指标数据异常的情况下,确定所述目标指标数据存在异常。
可选的,本申请第一方面提供的商户风险防控方法,基于时序异常检测模型,检测所述商户的目标指标数据是否存在异常,包括:
确定所述目标指标数据在当前时段内对应的多个时序实测值;
基于时序预测值与所述多个时序实测值,计算得到与所述多个时序实测值对应的多个时序异常值;
在所述多个时序异常值的平均值大于或等于预设值的情况下,确定所述检测结果指示所述目标指标数据异常;
在所述多个时序异常值的平均值小于所述预设值的情况下,确定所述检测结果指示所述目标指标数据正常。
可选的,本申请第一方面提供的商户风险防控方法,确定所述时序异常值的步骤,包括:
确定所述时序实测值与所述时序预测值间的时序差值;
计算所述时序差值与所述时序预测值间的比值,得到所述时序异常值。
可选的,本申请第一方面提供的商户风险防控方法,所述方法包括以下至少一项:
所述维度包括地域维度、商户类型维度、支付接口维度中的至少一个;
所述类别包括商户注册类别和商户运营类别中的至少一个。
可选的,本申请第一方面提供的商户风险防控方法,所述方法包括以下至少一项:
所述商户注册类别中包括以下至少一项指标:商户性质、维护记录、经营内容、所属行业;
所述商户运营类别包括以下至少一项指标:交易金额、投诉次数、商品质量、服务质量和支付接口使用情况。
第二方面,本申请实施例还提供一种商户风险防控装置,包括:
第一确定模块,用于在检测到商户的目标指标数据异常的情况下,确定所述目标指标数据对应的目标异常参数;
获取模块,用于基于指标、类别和维度之间的对应关系,获取所述目标指标数据对应的目标指标所属的目标类别,及所述目标类别所属的目标维度;
第二确定模块,用于基于所述目标维度和所述目标类别,确定至少一个联动类别, 所述至少一个联动类别下包括引起所述目标指标数据异常的指标;
处理模块,用于基于所述目标异常参数和所述至少一个联动类别,进行商户风险防控。
第三方面,本申请实施例还提供一种电子设备,包括:
处理器;以及
被安排成存储计算机可执行指令的存储器,所述可执行指令在被执行时使所述处理器执行以下操作:
在检测到商户的目标指标数据异常的情况下,确定所述目标指标数据对应的目标异常参数;
基于指标、类别和维度之间的对应关系,获取所述目标指标数据对应的目标指标所属的目标类别,及所述目标类别所属的目标维度;
基于所述目标维度和所述目标类别,确定至少一个联动类别,所述至少一个联动类别下包括引起所述目标指标数据异常的指标;
基于所述目标异常参数和所述至少一个联动类别,进行商户风险防控。
第四方面,本申请实施例还提供一种计算机可读存储介质,其中,所述计算机可读存储介质存储一个或多个程序,所述一个或多个程序当被包括多个应用程序的电子设备执行时,使得所述电子设备执行以下操作:
在检测到商户的目标指标数据异常的情况下,确定所述目标指标数据对应的目标异常参数;
基于指标、类别和维度之间的对应关系,获取所述目标指标数据对应的目标指标所属的目标类别,及所述目标类别所属的目标维度;
基于所述目标维度和所述目标类别,确定至少一个联动类别,所述至少一个联动类别下包括引起所述目标指标数据异常的指标;
基于所述目标异常参数和所述至少一个联动类别,进行商户风险防控。
本申请实施例采用的上述至少一个技术方案能够达到以下有益效果:
本申请实施例中,从不同的维度全面设置用于商户风险防控的指标,并将指标进行分类得到相应的类别,建立指标、类别和维度之间的对应关系,以在检测到指标数据存 在异常的目标指标时,基于该对应关系,获取引起目标指标数据异常的指标所属的所有联动类别,进而基于获取到的联动类别和表征目标指标数据异常程度的目标异常参数进行商户风险防控。因此,通过对商户风险防控相关的指标进行异常检测,并每当检测到指标数据存在异常的指标时,基于指标异常情况及时进行商户风险防控,实现商户风险确认前的风险提前识别,提高商户风险防控的时效性。
附图说明
此处所说明的附图用来提供对本申请的进一步理解,构成本申请的一部分,本申请的示意性实施例及其说明用于解释本申请,并不构成对本申请的不当限定。在附图中:
图1为本申请实施例中提供的商户风险防控方法的流程示意图;
图2为本申请实施例中提供的商户风险防控信息流向示意图;
图3为本申请实施例中提供的商户风险防控装置的结构示意图;
图4为本申请实施例中提供的电子设备的结构示意图。
具体实施方式
为使本申请的目的、技术方案和优点更加清楚,下面将结合本申请具体实施例及相应的附图对本申请技术方案进行清楚、完整地描述。显然,所描述的实施例仅是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本申请保护的范围。
随着移动支付方式的普及,人们越来越习惯于使用第三方支付平台进行便捷的移动支付。而对于移动支付而言,选择使用第三方支付平台的用户和商户是最重要的两个主体,且这两个群体相互之间存在着积极的影响:越多的商户提供可以使用第三方支付平台进行移动支付的场景和服务,会积极拉动更多的用户使用该第三方支付平台,反之,用户数量级的不断增长,同样也会带动商户数量级的不断增长。
然而,随着商户数量的不断增长,问题商户的占比也会有所增长,当问题商户进行欺诈、赌博、套现及其他违规违禁行为时,会带来相应的风险,严重时可能会对用户的利益、商户的正常运营造成危害。因此,如何及时有效地对数以万计的商户进行风险识别和防控成为亟待解决的问题。而且,考虑到商户风险往往涉及的资金数量级大,容易引起舆情和监管部门的注意,因此,商户风险防控的及时性显得尤为重要。
为了实现对商户风险的有效防控,可以采用以下所述的防控方案:
(1)基于商户策略的防控方案,需要策略运营人员通过人工审理及分析案件的方式提取风险特征,然后基于提取的风险特征配置线上防控策略,以在商户的事前准入、事中交易和事后管控环节进行层层防控。
(2)基于风险检测模型的防控方案,需要针对特定的商户风险,比如欺诈、赌博、套现等违规违禁行为引发的商户风险,分析设计相应的风险特征,然后基于风险特征建立风险检测模型,进而可以基于该风险检测模型输出的风险值,结合相应的防控策略进行风险防控。
(3)基于用户投诉的防控方案,实施该防控方案时,主要是对有用户投诉的商户进行审理定性,核定风险后对相应的风险商户进行处罚,同时一并处罚与该风险商户相关的其他商户。
然而,上述几种商户风险防控方案一般存在以下缺陷:
(1)在基于商户策略的防控方案中,不仅需要花费较多的人力审理及分析案件以进行风险特征的提取,而且审理及分析的案件一般是针对已经发生的风险,如此,在有新型风险出现的情况下,需要重新审理及分析案件和提取风险特征以重新配置防控策略,很明显会降低风险防控的时效性,另外,新策略的防控效果很多时候需要在策略上线后才能进行评价。
(2)在基于风险检测模型的防控方案中,建立风险检测模型时,需要利用历史数据中的黑白样本,确定输入风险检测模型中的变量,然后训练上线模型,建模周期比较长,而且,当风险检测模型上线后,其对于历史上没有出现过的商户风险的识别能力有限,会出现模型检测能力衰退的情况,为此需要定期重新训练模型,显然会影响风险防控的时效性。
(3)在基于用户投诉的防控方案中,由于该方案属于事后防控,所以不仅需要商户风险已经发生,并且要求在该商户风险伤害了用户利益的情况下,有用户主动进行投诉,进而基于用户投诉对商户风险定性之后,才能对该类商户风险进行相应的管控,时效性不好。
因此,在实际应用场景中,以上几种防控方案的风险防控结果往往无法满足商户风险防控的实际需求,影响用户体验。
鉴于此,本申请实施例提供了一种商户风险防控方案,能够对商户风险进行提 前识别,实现商户风险的动态防控,能够提高商户风险防控的时效性,且具有较强的风险动态对抗能力,适用于不同类型的商户风险,通用性好。
以下结合附图,详细说明本申请各实施例提供的技术方案。
参见图1所示,本申请实施例提供一种商户风险防控方法,该方法可包括以下步骤:
S101:在检测到商户的目标指标数据异常的情况下,确定目标指标数据对应的目标异常参数。
S103:基于指标、类别和维度之间的对应关系,获取目标指标数据对应的目标指标所属的目标类别,及目标类别所属的目标维度。
可以理解,在本申请实施例中,需要对商户的目标指标数据进行异常检测,并在检测到目标指标数据存在异常时,一方面需要基于预先建立的指标、类别和维度之间的对应关系,查找获取该存在异常的目标指标数据对应的目标指标所属的目标类别以及该目标类别所属的目标维度,另一方面需要确定该目标指标数据对应的目标异常参数。
可选的,可以通过以下实施例实现对商户的目标指标数据的异常检测并确定目标异常参数:
基于时序异常检测模型,检测商户的目标指标数据是否存在异常。
可选的,时序异常检测模型包括ARIMA模型(Autoregressive Integrated Moving Average Model,自回归积分滑动平均模型)、HOLT-WINTERS模型(指数平滑模型)和LSTM(Long Short-Term Memory,长短期记忆网络)模型中的一个,当然也可以采用其他能够实现异常检测的模型。
可选的,基于时序异常检测模型,检测商户的目标指标数据是否存在异常的步骤可以包括:
确定目标指标数据在当前时段内对应的多个时序实测值;
基于时序预测值与多个时序实测值,计算得到与多个时序实测值对应的多个时序异常值;
在多个时序异常值的平均值大于或等于预设值的情况下,确定检测结果指示目标指标数据异常;
在多个时序异常值的平均值小于预设值的情况下,确定检测结果指示目标指标 数据正常。
可选的,确定时序异常值的步骤,可以包括:
确定时序实测值与时序预测值间的时序差值;
计算时序差值与时序预测值间的比值,得到时序异常值。
可选的,目标指标数据可以反映作为参变量的目标指标的取值情况,目标异常参数可以反映目标指标数据存在异常时的异常程度,优选的,目标指标数据对应的目标异常参数可以为多个时序异常值的平均值。
其中,时序实测值可以反映目标指标数据在当前时段的一个检测时间点上对应的实际检测值,时序预测值可以反映目标指标数据按正常走向在该检测时间点上的预估值,而时序实测值与时序预测值间的时序差值可以反映目标指标数据在该检测时间点上,实际测量值与正常的预估值之间的差距,通过该时序差值则可以衡量目标指标数据在该检测时刻点的检测值是否异动。
可以理解,上述当前时段和预设值的取值可以根据进行异常检测的商户和目标指标的实际情况确定的经验值;比如,可以在连续3天内检测到的多个时序异常值的平均值大于或等于预设值时,确定检测结果指示目标指标数据异常。
进一步地,在检测结果指示目标指标数据异常的情况下,确定目标指标数据存在异常。
可选的,在建立指标、类别和维度之间的对应关系时,可以从不同的维度全面设置用于商户风险防控的指标,并将指标进行分类得到相应的类别,其中,在设置指标的维度时,可以考虑从粗到细和/或从上级到下级等角度进行多层级的设置,以使指标的设计的覆盖范围更加全面、更加具体。
其中,维度可以包括地域维度、商户类型维度、支付接口维度中的至少一个。
例如,地域维度从粗到细可以包括自治区、省份、城市、旗县和街道等。又例如,商户类型维度从上级到下级可以包括:服务商、业务拓展专员、线下商户、商户交易等。还例如,支付接口维度可以包括是否签约第三方支付接口、第三方支付接口是否为支付宝等。
可以理解,在本申请实施例中,考虑到在商户分布于不同地域的情况下,其经营等方面可能会受一定地域特性的影响,以及不同类型的商户各自的经营等方面也会有 所区别,而采用的支付接口不同,其资金流转权限、操作流程等方面也会有所区别,则可能引起的商户风险也会存在区别,因此,可以通过设置不同的维度对商户进行异常检测,以实现更加全面的商户风险监控。
另外,维度下的类别可以包括商户注册类别和商户运营类别中的至少一个。
可选的,商户注册类别可以包括以下至少一项指标:商户性质、维护记录、经营内容、所属行业;商户运营类别可以包括以下至少一项指标:交易金额、投诉次数、商品质量、服务质量和支付接口使用情况。
其中,商户性质可以包括个人商户和企业商户;维度记录包括业务拓展专员(BD,Business Development)对商户进行定期维度的记录,维护的内容可以包括监控商户缴税、监控商户经营许可证是否到期、到期是否更新,监控商户是否及时处理客户投诉等,BD每次维度商户后对应生成相应的维护记录;经营内容可以指商户注册时填写的商户经营的商品,比如,水果店的经营内容可以是水果,蔬菜店的经营内容可以是蔬菜;商户所属的行业可以指商户注册时填写的行业,也可以指根据商户的经营内容推导出的行业;交易金额可以包括商户在设定时长内的交易总金额;投诉次数可以包括商户最近周期内的投诉次数;支付接口使用情况可以包括商户的第三方支付接口是否被其他商户使用、第三方支付接口完成交易的地理位置信息。
当然,用于对商户进行风险防控的维度、类别、指标等的设置不限于上述内容,可以根据具体的、实际的防控需求进行相应调整。
在本申请实施例中,可以通过对指定维度的指定类别下的特定指标进行异常检测,实现对商户风险防控过程中时序上发生异动的目标指标数据进行捕捉;也可以对预先设计的所有指标进行异常检测,实现对商户风险防控过程中时序上发生异动的目标指标数据进行捕捉。其中,异常检测的过程可以使周期性进行的,也可以是不间断持续进行的,以及在进行周期性异常检测的情况下,对不同指标的轮询周期,可以相同,也可以不同。
S105:基于目标维度和目标类别,确定至少一个联动类别,至少一个联动类别下包括引起目标指标数据异常的指标。
可以理解,在基于检测到存在异常的目标指标锁定对应的目标维度和目标类别后,进一步基于目标维度和目标类别进行至少一个联动类别的获取,该至少一个联动类别下包括能够引起目标指标数据异常的指标,从而更加全面地找出相关的异常指标。
可选的,基于目标维度和目标类别,确定至少一个联动类别的步骤可以包括:
获取与目标维度关联的至少一个联动维度;
在至少一个联动维度下,获取与目标类别关联的至少一个联动类别。
可以理解,在本申请实施例中,可以维度与维度、类别与类别间的关联关系进行至少一个联动类别的上卷、下探式查找,以更加全面地查找出相关的异常指标,可选的,维度与维度、类别与类别间的关联关系可以记载在指标、类别和维度之间的对应关系中进行维护,也可以单独建立并维护维度与维度、类别与类别间的关联关系。
能够理解,联动维度可以指与存在异常的目标指标数据对应的目标指标所属的目标维度存在关联关系的维度,联动类别可以指与存在异常的目标指标数据对应的目标指标所属的目标类别存在关联关系的类别,各联动类别从属于相应的联动维度,以及各联动类别下的所有指标中存在能够引起目标指标数据异常的指标,换言之,各联动类别下的各指标对应的各指标数据中存在能够引起目标指标数据异常的指标数据。
在确定目标指标数据对应的目标异常参数和获取到至少一个联动类别后,则可以据此进行商户风险防控,即执行如下步骤:
S107:基于目标异常参数和至少一个联动类别,进行商户风险防控。
可选的,该步骤S107可以包括:
在至少一个联动类别下,获取存在异常的指标数据对应的第一指标集合;
基于目标异常参数、至少一个联动类别和第一指标集合,进行商户风险防控。
可以理解,基于与目标指标关联的至少一个联动类别,可以对该至少一个联动类别下的指标对应的指标数据进行异常检测,可选的,可以对指定时间段内的指标数据进行异常检测,以提高与存在异常的目标指标数据间的关联程度,提高第一指标集合中指标获取的准确性和可靠性,以及可以采用上述检测目标指标数据是否存在异常的检测方案类似的方案对至少一个联动类别下的指标数据进行异常检测,也就是说,基于时序异常检测模型检测至少一个联动类别下的指标对应的指标数据是否存在异常,并在检测结果指示异常时,确定至少一个联动类别下的指标对应的指标数据存在异常,进而可以将存在异常的至少一个指标数据对应的至少一个指标形成第一指标集合,以与目标异常参数、至少一个联动类别一同用于商户风险防控。
可选的,基于目标异常参数、至少一个联动类别和第一指标集合,进行商户风 险防控,可以具体包括以下步骤:
基于至少一个联动类别,确定造成目标指标数据出现异常的异常原因。
可以理解,在本申请实施例中,可以根据至少一个联动类别确定导致目标指标数据出现异常的原因,可选的,基于用户的选择将至少一个联动类别中用户较为关注的联动类别确定为异常原因,可以包括以下步骤:
接收类别选择指令;
在至少一个联动类别中,确定类别选择指令对应的目标联动类别;
根据目标联动类别,确定异常原因。
当然,除了基于用户的选择外,还可以通过其他方式基于至少一个联动类别确定异常原因,比如设置类别筛选规则,以基于该类别筛选规则自动在至少一个联动类别中筛选出符合规则要求的类别形成异常原因,等等。
进一步地,确定第一指标集合对应的第一参数集合,第一参数集合中各异常参数与第一指标集合中各指标的指标数据一一对应。
可以理解,对于至少一个联动类别下指标数据存在异常的指标,相应地能够确定各指标数据的异常参数,从而能够得到与第一指标集合对应的第一参数集合,其中,各异常参数的确定可以参照上述确定目标异常参数的实施例执行,在此不再赘述。
进一步地,基于目标异常参数、第一指标集合和第一参数集合,确定第二指标集合和第二参数集合,第二参数集合中各异常参数与第二指标集合中各指标的指标数据一一对应。
在本申请实施例中,可以基于目标异常参数将第一指标集合和第一参数集合转换为第二指标集合和第二参数集合,以实现对商户风险防控的力度的合理控制,其中,第二指标集合包括第一指标集合中的至少一个指标,第二参数集合包括第一参数集合中的至少一个异常参数。
可选的,确定第二指标集合和第二参数集合的过程可以包括以下步骤:
确定目标异常参数与第一参数集合中的各异常参数间的差值;
在差值处于预设阈值范围内的情况下,基于差值对应的异常参数构建第二参数集合;
基于第二参数集合中各异常参数对应的指标构建第二指标集合。
在该实施例中,可以将第一参数集合中指示的异常程度在目标指标数据对应的异常程度(即目标异常参数)以上的异常参数划入第二参数集合,也就是当第一参数集合中的异常参数与目标异常参数的差距在一定指定的范围内时,将该异常参数划入第二参数集合,继而将第二参数集合中各异常参数对应的第一指标集合中的指标划归为第二指标集合。
进一步则可以执行基于异常原因、第二指标集合和第二参数集合,进行商户风险防控的步骤,可选的,可以包括:
在第二指标集合中,获取风险防控指标,风险防控指标与异常原因相匹配;
在第二参数集合中,获取风险防控参数,风险防控参数与风险防控指标一一对应;
基于风险防控指标和风险防控参数,进行商户风险防控。
可以理解,在基于指标数据存在异常的指标所属类别(即至少一个联动类别)的分布情况确定异常原因后,可以在第二指标集合中获取与该异常原因匹配的风险防控指标,以及在第二参数集合中获取与该异常原因匹配的前述风险防控指标对应的风险防控参数,以进行更加合理的商户风险防控。
上述基于风险防控指标和风险防控参数,进行商户风险防控可以通过以下具体实施例实现:
在具体实施例一中,进行商户风险防控的方案可以包括:
在风险防控参数大于异常参数阈值的情况下,获取与风险防控指标对应的风险防控策略;
基于风险防控策略进行商户风险防控。
可以理解,当基于第二参数集合中的异常参数和异常原因得到的风险防控参数的值大于异常参数阈值时,说明指标的异常程度已经足够大,此时,可以采用直接部署相应风险防控策略的方式及时进行商户风险防控,即在及时检测到指标异常情况的同时,提高商户风险防控的时效性。
可选的,风险防控策略包括对商户进行罚款、返还欺诈所得利益、责令停业整顿、责令完善运营资质等等,以禁止商户进行任何导致风险的行为。
在具体实施例二中,可以通过进行商户风险识别模型训练的方式,以基于最新的指标异常情况更新商户风险识别模型,在及时检测到指标异常情况的同时,提高商户风险防控的时效性。
可选的,在进行商户风险防控前,在本申请实施例中,还需要执行获取历史风险指标及对应的历史异常参数的步骤,进而将历史风险指标与当前的风险防控指标进行比对,以基于不同的比对结果执行相应的商户风险防控方案,避免出现漏过历史风险的情况,以实现覆盖范围更加全面的商户风险防控。
可选的,基于风险防控指标和风险防控参数,进行商户风险防控的方案可以包括以下两个方面:
第一方面,在风险防控指标与历史风险指标相比存在遗漏指标的情况下,基于第一训练样本,对初始模型进行训练,得到第一商户风险识别模型;
其中,遗漏指标包含在历史风险指标中,第一训练样本包括风险防控指标、风险防控参数、遗漏指标及遗漏指标对应的遗漏异常参数。
可以理解,在基于当前的指标异常情况确定的风险防控指标相较于历史风险指标,存在待防控的指标覆盖范围不全的情况时,为了商户风险识别的时效性,即涵盖可能出现的新风险,也确保在风险识别时不会漏掉已经确认的历史风险指标,则可以考虑基于最新的风险防控指标及其对应风险防控参数、遗漏指标及其对应的遗漏异常参数作为第一商户风险识别模型的训练样本,以使基于此训练出的最新的第一商户风险识别模型能够识别历史的和最新的商户风险,时效性好。
可选的,上述初始模型可以为采用孤立森林iForest算法或符号回归(Symbolic Regression)算法构建的未经训练的原始模型,也可以为基于历史风险指标等数据训练好的正在使用的商户风险识别模型。
可选的,基于第一训练样本,对初始模型进行训练得到第一商户风险识别模型的过程,可以包括:
基于风险防控参数和遗漏异常参数,确定风险识别阈值;
基于风险防控指标和遗漏指标,生成初始模型的输入向量;
将输入向量输入初始模型,得到初始模型的输出;
根据初始模型的输出与风险识别阈值之间的差距,调整初始模型的参数;
重复以上步骤,直至差距满足预设条件,得到第一商户风险识别模型。
可以理解,考虑到风险防控参数和遗漏异常参数反映相应的指标数据的异常程度,则可以基于此确定风险识别阈值,以使风险识别标准可靠而准确。
进一步地,部署第一商户风险识别模型以用于商户风险识别,并输出风险识别结果;
获取与风险识别结果对应的风险防控策略,以基于风险防控策略进行风险防控。
可以理解,在当前的风险防控指标与历史风险指标相比存在遗漏指标的情况下,进行商户风险识别模型训练时,需要将遗漏的部分历史指标及其对应的异常参数考虑在内。
需要说明的是,基于上述训练样本,可以确定模型训练时的训练集和验证集。可选的,训练集和验证集的样本数量可以灵活调配,例如,70%的训练样本作为训练集,用于训练商户风险识别模型,而剩余30%的训练样本作为验证集,验证商户风险识别模型的输出是否满足要求,以实现模型效果评估。可选的,风险防控策略包括对商户进行罚款、返还欺诈所得利益、责令停业整顿、责令完善运营资质等等,以禁止商户进行任何导致风险的行为。
第二方面,在风险防控指标与历史风险指标相比不存在遗漏指标的情况下,基于第二训练样本,对初始模型进行训练,得到第二商户风险识别模型;
其中,遗漏指标包含在历史风险指标中,第二训练样本包括第二风险防控指标和第二风险防控参数。
可以理解,在基于当前的指标异常情况确定的风险防控指标相较于历史风险指标,不存在待防控的指标覆盖范围不全的情况时,可以考虑基于最新的且全面的风险防控指标及其对应风险防控参数作为第二商户风险识别模型的训练样本,以使基于此训练出的最新的第二商户风险识别模型能够识别历史的和最新的商户风险,时效性好。
可选的,上述初始模型可以为采用孤立森林iForest算法或符号回归(Symbolic Regression)算法构建的未经训练的原始模型,也可以为基于历史风险指标等数据训练好的正在使用的商户风险识别模型。
可选的,基于第二训练样本,对初始模型进行训练得到第二商户风险识别模型的过程,可以包括:
基于风险防控参数,确定风险识别阈值;
基于风险防控指标,生成初始模型的输入向量;
将输入向量输入初始模型,得到初始模型的输出;
根据初始模型的输出与风险识别阈值之间的差距,调整初始模型的参数;
重复以上步骤,直至差距满足预设条件,得到第二商户风险识别模型。
需要说明的是,基于上述第二训练样本,可以确定模型训练时的训练集和验证集。可选的,训练集和验证集的样本数量可以灵活调配,例如,70%的训练样本作为训练集,用于训练第二商户风险识别模型,而剩余30%的训练样本作为验证集,验证第二商户风险识别模型的输出是否满足要求,以实现模型效果评估。
进一步地,部署第二商户风险识别模型以用于商户风险识别,并输出风险识别结果;
获取与风险识别结果对应的风险防控策略,以基于风险防控策略进行风险防控。
可以理解,在当前的风险防控指标与历史风险指标相比不存在遗漏指标的情况下,进行商户风险识别模型训练时,可以直接基于风险防控指标及其对应的风险防控参数执行。
可选的,风险防控策略包括对商户进行罚款、返还欺诈所得利益、责令停业整顿、责令完善运营资质等等,以禁止商户进行任何导致风险的行为。
综上,在本申请的实施例中,从不同的维度全面设置用于商户风险防控的指标,并将指标进行分类得到相应的类别,建立指标、类别和维度之间的对应关系,以在检测到指标数据存在异常的目标指标时,基于该对应关系,获取引起目标指标数据异常的指标所属的所有联动类别,进而基于获取到的联动类别和表征目标指标数据异常程度的目标异常参数进行商户风险防控。因此,通过对商户风险防控相关的指标进行异常检测,并每当检测到指标数据存在异常的指标时,基于指标异常情况及时进行商户风险防控,实现商户风险确认前的风险提前识别,提高商户风险防控的时效性。
本申请实施例所述的商户风险防控方法,还可以通过如图2所示的信息流进行展示。
(1)异动感知(即异常检测),即监控哪里出现异动。
可以通过合理的维度划分和指标设计,比如根据签约信息、商户类型、支付机 构等划分维度,以及在商户开户、交易、运营等环节分别设计指标,并进行指标分类,并随着时间推移,检测各个维度或特定维度上指标的时序变化,根据检测到的时序实际值与时序预测值之间的差值来判断指标是否出现异动,如果出现异动,则输出目标异动指标所属的目标异动类别、目标异动维度及目标异动指标对应的目标异动程度(即目标异常参数)。
其中,可以通过如下方式计算指标的异动程度:
异动程度=(时序实际值-时序预测值)/时序预测值。
(2)异动下探与上卷,即寻找异动原因与相关的异动指标。
基于上述输出的目标异动类别、目标异动维度进行下探和/或上卷关联类别、关联维度下的明细指标数据,根据上述输出的目标异动程度,确定合适的防控力度,异动程度越大需要的防控措施越严格,以对关联类别、关联维度下的明细指标数据进行筛选,找出所有相关的异动指标,并基于与上述相同的方式得到指标的异常程度,同时,基于各异动指标所属类别的分布情况可以确定异动原因(即异常原因),以输出异动原因、与该异动原因匹配的异动指标列表及对应的指标异动程度。
可选的,在输出指标列表及对应的异动程度时,可以考虑已识别出的历史风险指标及其异动程度,以避免漏过风险。
(3)异动响应
基于上述输出的异动指标列表及对应的指标异动程度,以及采用孤立森林iForest算法、符号回归算法等快速建立模型,并在上述异动明细数据基础上进行模型评估,得到模型的预估效果,输出风险识别模型,并推荐风险防控策略。
(4)风险防控
将上述风险识别模型输出到风险防控平台进行模型部署,实现对模型真实结果的评估,并可以结合结果匹配风险防控策略,以快速防控新风险。
举例来说,以城市维度为主维度对商户进行异动感知或异常检测,当感知到城市A的指标B对应的数据上涨异常,与正常的时序预测值相比连续多天出现异动,且平均异动程度达到一定值时,输出该指标B所属的指标类别C,指标B的数据对应的平均异动程度以及城市维度A。
进一步地,基于指标、类别和维度之间预先建立的关系,查找确定与城市维度 A关联的其他城市维度、旗县维度、城镇维度等,比如在城市A被异动感知到商户在地域存在连锁机构、分支机构或服务商等等,以及查找确定与类别C关联的其他类别,比如在其他地域下该商户对应的相应类别等等,则可以通过对确定的各类别下指标的指标数据进行异动感知挖掘,找出所有相关异动指标,并基于确定的类别即相关异动指标的分布情况分析确定异动原因,进而依据与异动原因匹配的指标及对应的异动程度等进行模型训练组建、模型评估以及防控策略推荐等,实现商户风险防控。
从整个过程来看,通过异动感知可以快速及时的发掘异常,并且通过异动下探与上卷的方式能够快速定位原因,输出有效的异动指标,进行模型训练和风险防控,保证了防控的时效性,另外风险的转移会引起其他指标的异动,可以动态捕捉风险,增强了动态对抗性,可见整个系统不关注具体风险,只关注异动,任何形势的风险只要引起异动就能被及时检测和防控,具体很好的通用性。
可见,本申请实施例的商户风险防控方法具有以下优点:
(1)时效性更好:基于实时风险指标计算和监控可以更加及时地发现和定位风险;
(2)动态对抗性更强:商户风险的转移和变异都可以被及时发现,并且可以进行策略和模型的快速组装,完成新风险的防控;
(3)通用性更好:该发明不针对任何商户具象风险,通过全面的指标设计,可以检测不同风险带来的商户异常。
参见图3所示,本申请实施例还提供了一种商户风险防控装置,该装置可包括:
第一确定模块301,用于在检测到商户的目标指标数据异常的情况下,确定目标指标数据对应的目标异常参数;
获取模块303,用于基于指标、类别和维度之间的对应关系,获取目标指标数据对应的目标指标所属的目标类别,及目标类别所属的目标维度;
第二确定模块305,用于基于目标维度和目标类别,确定至少一个联动类别,至少一个联动类别下包括引起目标指标数据异常的指标;
处理模块307,用于基于目标异常参数和至少一个联动类别,进行商户风险防控。
能够理解,图3给出的商户风险防控装置能够实现图1中所述的商户风险防控方法的各个步骤,前述实施例中关于商户风险防控方法的相关阐述均适用于商户风险防 控装置,此处不再赘述。
图4是本申请的一个实施例电子设备的结构示意图。请参考图4,在硬件层面,该电子设备包括处理器,可选地还包括内部总线、网络接口、存储器。其中,存储器可能包含内存,例如高速随机存取存储器(Random-Access Memory,RAM),也可能还包括非易失性存储器(non-volatile memory),例如至少1个磁盘存储器等。当然,该电子设备还可能包括其他业务所需要的硬件。
处理器、网络接口和存储器可以通过内部总线相互连接,该内部总线可以是ISA(Industry Standard Architecture,工业标准体系结构)总线、PCI(Peripheral Component Interconnect,外设部件互连标准)总线或EISA(Extended Industry Standard Architecture,扩展工业标准结构)总线等。所述总线可以分为地址总线、数据总线、控制总线等。为便于表示,图4中仅用一个双向箭头表示,但并不表示仅有一根总线或一种类型的总线。
存储器,用于存放程序。具体地,程序可以包括程序代码,所述程序代码包括计算机操作指令。存储器可以包括内存和非易失性存储器,并向处理器提供指令和数据。
处理器从非易失性存储器中读取对应的计算机程序到内存中然后运行,在逻辑层面上形成商户风险防控装置。处理器,执行存储器所存放的程序,并具体用于执行以下操作:
在检测到商户的目标指标数据异常的情况下,确定目标指标数据对应的目标异常参数;
基于指标、类别和维度之间的对应关系,获取目标指标数据对应的目标指标所属的目标类别,及目标类别所属的目标维度;
基于目标维度和目标类别,确定至少一个联动类别,至少一个联动类别下包括引起目标指标数据异常的指标;
基于目标异常参数和至少一个联动类别,进行商户风险防控。
上述如本申请前述对应实施例揭示的商户风险防控装置执行的方法可以应用于处理器中,或者由处理器实现。处理器可能是一种集成电路芯片,具有信号的处理能力。在实现过程中,上述方法的各步骤可以通过处理器中的硬件的集成逻辑电路或者软件形式的指令完成。上述的处理器可以是通用处理器,包括中央处理器(Central Processing Unit,CPU)、网络处理器(Network Processor,NP)等;还可以是数字信号处理器(Digital Signal Processor,DSP)、专用集成电路(Application Specific Integrated Circuit,ASIC)、 现场可编程门阵列(Field-Programmable Gate Array,FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合本申请实施例所公开的方法的步骤可以直接体现为硬件译码处理器执行完成,或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器,闪存、只读存储器,可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器,处理器读取存储器中的信息,结合其硬件完成上述方法的步骤。
该电子设备还可执行前述对应商户风险防控装置执行的方法,并实现商户风险防控装置在前述对应实施例中的功能,本申请实施例在此不再赘述。
本申请实施例还提出了一种计算机可读存储介质,该计算机可读存储介质存储一个或多个程序,该一个或多个程序包括指令,该指令当被包括多个应用程序的电子设备执行时,能够使该电子设备执行图1所示实施例中商户风险防控装置执行的方法,并具体用于执行:
在检测到商户的目标指标数据异常的情况下,确定目标指标数据对应的目标异常参数;
基于指标、类别和维度之间的对应关系,获取目标指标数据对应的目标指标所属的目标类别,及目标类别所属的目标维度;
基于目标维度和目标类别,确定至少一个联动类别,至少一个联动类别下包括引起目标指标数据异常的指标;
基于目标异常参数和至少一个联动类别,进行商户风险防控。
本领域内的技术人员应明白,本发明的实施例可提供为方法、系统、或计算机程序产品。因此,本发明可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且,本发明可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本发明是参照根据本发明实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供 这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器,使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中,使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品,该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上,使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理,从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
在一个典型的配置中,计算设备包括一个或多个处理器(CPU)、输入/输出接口、网络接口和内存。
内存可能包括计算机可读介质中的非永久性存储器,随机存取存储器(RAM)和/或非易失性内存等形式,如只读存储器(ROM)或闪存(flash RAM)。内存是计算机可读介质的示例。
计算机可读介质包括永久性和非永久性、可移动和非可移动媒体可以由任何方法或技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机的存储介质的例子包括,但不限于相变内存(PRAM)、静态随机存取存储器(SRAM)、动态随机存取存储器(DRAM)、其他类型的随机存取存储器(RAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(CD-ROM)、数字多功能光盘(DVD)或其他光学存储、磁盒式磁带,磁带磁磁盘存储或其他磁性存储设备或任何其他非传输介质,可用于存储可以被计算设备访问的信息。按照本文中的界定,计算机可读介质不包括暂存电脑可读媒体(transitory media),如调制的数据信号和载波。
还需要说明的是,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、商品或者设备不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、商品或者设备所固 有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括所述要素的过程、方法、商品或者设备中还存在另外的相同要素。
本领域技术人员应明白,本申请的实施例可提供为方法、系统或计算机程序产品。因此,本申请可采用完全硬件实施例、完全软件实施例或结合软件和硬件方面的实施例的形式。而且,本申请可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
以上所述仅为本申请的实施例而已,并不用于限制本申请。对于本领域技术人员来说,本申请可以有各种更改和变化。凡在本申请的精神和原理之内所作的任何修改、等同替换、改进等,均应包含在本申请的权利要求范围之内。

Claims (21)

  1. 一种商户风险防控方法,包括:
    在检测到商户的目标指标数据异常的情况下,确定所述目标指标数据对应的目标异常参数;
    基于指标、类别和维度之间的对应关系,获取所述目标指标数据对应的目标指标所属的目标类别,及所述目标类别所属的目标维度;
    基于所述目标维度和所述目标类别,确定至少一个联动类别,所述至少一个联动类别下包括引起所述目标指标数据异常的指标;
    基于所述目标异常参数和所述至少一个联动类别,进行商户风险防控。
  2. 根据权利要求1所述方法,基于所述目标维度和所述目标类别,确定至少一个联动类别,包括:
    获取与所述目标维度关联的至少一个联动维度;
    在所述至少一个联动维度下,获取与所述目标类别关联的所述至少一个联动类别。
  3. 根据权利要求2所述方法,基于所述目标异常参数和所述至少一个联动类别,进行商户风险防控,包括:
    在所述至少一个联动类别下,获取存在异常的指标数据对应的第一指标集合;
    基于所述目标异常参数、所述至少一个联动类别和所述第一指标集合,进行商户风险防控。
  4. 根据权利要求3所述方法,基于所述目标异常参数、所述至少一个联动类别和所述第一指标集合,进行商户风险防控,包括:
    基于所述至少一个联动类别,确定造成所述目标指标数据出现异常的异常原因;
    确定所述第一指标集合对应的第一参数集合,所述第一参数集合中各异常参数与所述第一指标集合中各指标的指标数据一一对应;
    基于所述目标异常参数、所述第一指标集合和所述第一参数集合,确定第二指标集合和第二参数集合,所述第二参数集合中各异常参数与所述第二指标集合中各指标的指标数据一一对应;
    基于所述异常原因、所述第二指标集合和所述第二参数集合,进行商户风险防控。
  5. 根据权利要求4所述方法,基于所述至少一个联动类别,确定造成所述目标指标数据出现异常的异常原因,包括:
    接收类别选择指令;
    在所述至少一个联动类别中,确定所述类别选择指令对应的目标联动类别;
    根据所述目标联动类别,确定所述异常原因。
  6. 根据权利要求4所述方法,基于所述目标异常参数、所述第一指标集合和所述第一参数集合,确定第二指标集合和第二参数集合,包括:
    确定所述目标异常参数与所述第一参数集合中的各异常参数间的差值;
    在所述差值处于预设阈值范围内的情况下,基于所述差值对应的异常参数构建所述第二参数集合;
    基于所述第二参数集合中各异常参数对应的指标构建所述第二指标集合。
  7. 根据权利要求6所述方法,基于所述异常原因、所述第二指标集合和所述第二参数集合,进行商户风险防控,包括:
    在所述第二指标集合中,获取风险防控指标,所述风险防控指标与所述异常原因相匹配;
    在所述第二参数集合中,获取风险防控参数,所述风险防控参数与所述风险防控指标一一对应;
    基于所述风险防控指标和所述风险防控参数,进行商户风险防控。
  8. 根据权利要求7所述方法,基于所述风险防控指标和所述风险防控参数,进行商户风险防控,包括:
    在所述风险防控参数大于异常参数阈值的情况下,获取与所述风险防控指标对应的风险防控策略;
    基于所述风险防控策略进行商户风险防控。
  9. 根据权利要求7所述方法,在进行商户风险防控前,所述方法还包括:
    获取历史风险指标及对应的历史异常参数。
  10. 根据权利要求9所述方法,基于所述风险防控指标和所述风险防控参数,进行商户风险防控,包括:
    在所述风险防控指标与所述历史风险指标相比存在遗漏指标的情况下,基于第一训练样本,对初始模型进行训练,得到第一商户风险识别模型;
    部署所述第一商户风险识别模型以用于商户风险识别,并输出风险识别结果;
    获取与所述风险识别结果对应的风险防控策略,以基于所述风险防控策略进行风险防控;
    其中,所述遗漏指标包含在所述历史风险指标中,所述第一训练样本包括所述风险防控指标、所述风险防控参数、所述遗漏指标及所述遗漏指标对应的遗漏异常参数。
  11. 根据权利要求10所述方法,基于第一训练样本,对初始模型进行训练,得到 第一商户风险识别模型,包括:
    基于所述风险防控参数和所述遗漏异常参数,确定风险识别阈值;
    基于所述风险防控指标和所述遗漏指标,生成所述初始模型的输入向量;
    将所述输入向量输入所述初始模型,得到所述初始模型的输出;
    根据所述初始模型的输出与所述风险识别阈值之间的差距,调整所述初始模型的参数;
    重复以上步骤,直至所述差距满足预设条件,得到所述第一商户风险识别模型。
  12. 根据权利要求9所述方法,基于所述风险防控指标和所述风险防控参数,对所述商户进行风险防控,包括:
    在所述风险防控指标与所述历史风险指标相比不存在遗漏指标的情况下,基于所述第二训练样本,对初始模型进行训练,得到第二商户风险识别模型;
    部署所述第二商户风险识别模型以用于商户风险识别,并输出风险识别结果;
    获取与所述风险识别结果对应的风险防控策略,以基于所述风险防控策略进行风险防控;
    其中,所述遗漏指标包含在所述历史风险指标中,所述第二训练样本包括所述第二风险防控指标和所述第二风险防控参数。
  13. 根据权利要求12所述方法,基于所述第二训练样本,对初始模型进行训练,得到第二商户风险识别模型,包括:
    基于所述风险防控参数,确定风险识别阈值;
    基于所述风险防控指标,生成所述初始模型的输入向量;
    将所述输入向量输入所述初始模型,得到所述初始模型的输出;
    根据所述初始模型的输出与所述风险识别阈值之间的差距,调整所述初始模型的参数;
    重复以上步骤,直至所述差距满足预设条件,得到所述第二商户风险识别模型。
  14. 根据权利要求1~13之任一所述方法,还包括:
    基于时序异常检测模型,检测所述商户的目标指标数据是否存在异常;
    在检测结果指示所述目标指标数据异常的情况下,确定所述目标指标数据存在异常。
  15. 根据权利要求14所述方法,基于时序异常检测模型,检测所述商户的目标指标数据是否存在异常,包括:
    确定所述目标指标数据在当前时段内对应的多个时序实测值;
    基于时序预测值与所述多个时序实测值,计算得到与所述多个时序实测值对应的多 个时序异常值;
    在所述多个时序异常值的平均值大于或等于预设值的情况下,确定所述检测结果指示所述目标指标数据异常;
    在所述多个时序异常值的平均值小于所述预设值的情况下,确定所述检测结果指示所述目标指标数据正常。
  16. 根据权利要求15所述方法,确定所述时序异常值的步骤,包括:
    确定所述时序实测值与所述时序预测值间的时序差值;
    计算所述时序差值与所述时序预测值间的比值,得到所述时序异常值。
  17. 根据权利要求1~13之任一所述方法,包括以下至少一项:
    所述维度包括地域维度、商户类型维度、支付接口维度中的至少一个;
    所述类别包括商户注册类别和商户运营类别中的至少一个。
  18. 根据权利要求17所述方法,包括以下至少一项:
    所述商户注册类别中包括以下至少一项指标:商户性质、维护记录、经营内容、所属行业;
    所述商户运营类别包括以下至少一项指标:交易金额、投诉次数、商品质量、服务质量和支付接口使用情况。
  19. 一种商户风险防控装置,包括:
    第一确定模块,用于在检测到商户的目标指标数据异常的情况下,确定所述目标指标数据对应的目标异常参数;
    获取模块,用于基于指标、类别和维度之间的对应关系,获取所述目标指标数据对应的目标指标所属的目标类别,及所述目标类别所属的目标维度;
    第二确定模块,用于基于所述目标维度和所述目标类别,确定至少一个联动类别,所述至少一个联动类别下包括引起所述目标指标数据异常的指标;
    处理模块,用于基于所述目标异常参数和所述至少一个联动类别,进行商户风险防控。
  20. 一种电子设备,包括:
    处理器;以及
    被安排成存储计算机可执行指令的存储器,所述可执行指令在被执行时使所述处理器执行以下操作:
    在检测到商户的目标指标数据异常的情况下,确定所述目标指标数据对应的目标异常参数;
    基于指标、类别和维度之间的对应关系,获取所述目标指标数据对应的目标指标所属的目标类别,及所述目标类别所属的目标维度;
    基于所述目标维度和所述目标类别,确定至少一个联动类别,所述至少一个联动类别下包括引起所述目标指标数据异常的指标;
    基于所述目标异常参数和所述至少一个联动类别,进行商户风险防控。
  21. 一种计算机可读存储介质,其中,所述计算机可读存储介质存储一个或多个程序,所述一个或多个程序当被包括多个应用程序的电子设备执行时,使得所述电子设备执行以下操作:
    在检测到商户的目标指标数据异常的情况下,确定所述目标指标数据对应的目标异常参数;
    基于指标、类别和维度之间的对应关系,获取所述目标指标数据对应的目标指标所属的目标类别,及所述目标类别所属的目标维度;
    基于所述目标维度和所述目标类别,确定至少一个联动类别,所述至少一个联动类别下包括引起所述目标指标数据异常的指标;
    基于所述目标异常参数和所述至少一个联动类别,进行商户风险防控。
PCT/CN2019/112126 2018-12-04 2019-10-21 商户风险防控方法及装置 Ceased WO2020114110A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811476079.X 2018-12-04
CN201811476079.XA CN110046781B (zh) 2018-12-04 2018-12-04 商户风险防控方法及装置

Publications (1)

Publication Number Publication Date
WO2020114110A1 true WO2020114110A1 (zh) 2020-06-11

Family

ID=67273602

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/112126 Ceased WO2020114110A1 (zh) 2018-12-04 2019-10-21 商户风险防控方法及装置

Country Status (3)

Country Link
CN (1) CN110046781B (zh)
TW (1) TWI714262B (zh)
WO (1) WO2020114110A1 (zh)

Cited By (24)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112686521A (zh) * 2020-12-25 2021-04-20 江苏通付盾科技有限公司 一种风控规则调优方法和系统
CN112950222A (zh) * 2021-04-08 2021-06-11 腾讯科技(深圳)有限公司 资源处理异常检测方法、装置、电子设备及存储介质
CN113254542A (zh) * 2021-04-21 2021-08-13 泰康保险集团股份有限公司 数据可视化处理方法及装置、电子设备
CN113626421A (zh) * 2021-08-02 2021-11-09 浪潮软件股份有限公司 一种数据校验的数据质量管控方法
CN113642637A (zh) * 2021-08-12 2021-11-12 北京沃东天骏信息技术有限公司 数据检测方法、装置及存储介质
CN113836360A (zh) * 2021-09-24 2021-12-24 支付宝(杭州)信息技术有限公司 数据检测方法及装置
CN114321872A (zh) * 2022-01-07 2022-04-12 神华神东电力有限责任公司 锅炉保护方法及锅炉保护装置
CN114328490A (zh) * 2021-12-29 2022-04-12 浪潮卓数大数据产业发展有限公司 一种直播电商分析展现的构建方法、设备及介质
CN114612104A (zh) * 2020-12-09 2022-06-10 支付宝(杭州)信息技术有限公司 风险识别方法、装置和电子设备
CN114782192A (zh) * 2022-04-27 2022-07-22 中国银行股份有限公司 银行开户的风险控制方法及系统
CN115115359A (zh) * 2021-03-17 2022-09-27 腾讯科技(深圳)有限公司 一种异常用户识别方法、打击方法、装置、设备及介质
CN115438967A (zh) * 2022-09-06 2022-12-06 河南中原消费金融股份有限公司 一种基于超图的密集子图检测方法、装置及终端设备
CN115689343A (zh) * 2022-10-24 2023-02-03 万翼科技有限公司 供应商鉴别方法、装置、设备及存储介质
CN116860811A (zh) * 2022-03-25 2023-10-10 腾讯科技(深圳)有限公司 数据处理方法、装置、计算机设备及计算机可读存储介质
CN117215948A (zh) * 2022-10-24 2023-12-12 支付宝(杭州)信息技术有限公司 变更风险防控系统、方法、电子设备及存储介质
CN118277866A (zh) * 2024-04-24 2024-07-02 中电山河数字科技(南通)有限公司 一种用于制造行业的数据分级分类的方法和系统
CN118656708A (zh) * 2024-08-19 2024-09-17 长沙市长财科技有限公司 一种用户行为分析的支付风险评估与控制方法
CN118673445A (zh) * 2024-08-21 2024-09-20 浙江有数数智科技有限公司 基于大语言模型的异常程度识别方法、装置、介质及设备
CN119294840A (zh) * 2024-12-11 2025-01-10 蚂蚁智信(杭州)信息技术有限公司 一种数据处理方法、装置及设备
CN119379425A (zh) * 2024-12-31 2025-01-28 杭银消费金融股份有限公司 一种用于逾期指标变化的归因分析方法及系统
CN119420499A (zh) * 2024-09-24 2025-02-11 联想(北京)有限公司 一种数据处理方法、装置、设备及存储介质
CN119722320A (zh) * 2024-12-19 2025-03-28 广东通莞科技股份有限公司 一种基于聚合支付交易路由的移动支付监管方法与系统
CN119850328A (zh) * 2024-12-31 2025-04-18 交通银行股份有限公司 信用卡风险类别确定方法、装置、设备、介质及程序产品
CN120198157A (zh) * 2025-03-17 2025-06-24 浙江瑞耀信息技术有限公司 一种智能商户数据分析方法与系统

Families Citing this family (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110046781B (zh) * 2018-12-04 2020-07-07 阿里巴巴集团控股有限公司 商户风险防控方法及装置
CN110610290B (zh) * 2019-08-07 2023-06-30 创新先进技术有限公司 间联商户风险管控方法及其系统
CN110717653B (zh) * 2019-09-17 2023-05-12 创新先进技术有限公司 风险识别方法及装置和电子设备
CN111275547B (zh) * 2020-03-19 2023-07-18 重庆富民银行股份有限公司 基于孤立森林的风控系统及方法
CN111444060B (zh) * 2020-03-25 2023-06-30 北京奇艺世纪科技有限公司 异常检测模型训练方法、异常检测方法及相关装置
CN111489074A (zh) * 2020-04-07 2020-08-04 支付宝(杭州)信息技术有限公司 数据处理方法、装置、设备及存储介质
CN111798274B (zh) * 2020-07-03 2022-01-11 中国烟草总公司湖南省公司 零售户识别方法、终端及可读存储介质
CN111967779B (zh) * 2020-08-19 2022-12-16 支付宝(杭州)信息技术有限公司 一种风险评估方法、装置及设备
CN112738630A (zh) * 2020-12-29 2021-04-30 北京达佳互联信息技术有限公司 用户数据处理方法及装置、电子设备、存储介质
CN113159933A (zh) * 2021-05-20 2021-07-23 中国工商银行股份有限公司 一种风险控制方法、系统、设备及介质
CN113052516B (zh) * 2021-05-31 2022-01-04 深圳高灯计算机科技有限公司 一种基于流式计算的风控方法、系统及设备
CN113313575B (zh) * 2021-06-08 2022-06-03 支付宝(杭州)信息技术有限公司 一种风险识别模型的确定方法及装置
CN115063153A (zh) * 2022-06-22 2022-09-16 浙江惠瀜网络科技有限公司 商户风险管理方法及装置
CN115238289B (zh) * 2022-07-27 2025-09-19 中国银行股份有限公司 银行内部风险的控制方法及装置
CN115439928A (zh) * 2022-08-12 2022-12-06 中国银联股份有限公司 一种操作行为识别方法及装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101976419A (zh) * 2010-10-19 2011-02-16 中国工商银行股份有限公司 交易数据的风险监控处理方法和系统
US20170039479A1 (en) * 2015-08-07 2017-02-09 Wistron Corporation Risk assessment system and data processing method
CN107993143A (zh) * 2017-11-23 2018-05-04 深圳大管加软件与技术服务有限公司 一种信贷风险评估方法及系统
CN108154430A (zh) * 2017-12-28 2018-06-12 上海氪信信息技术有限公司 一种基于机器学习和大数据技术的信用评分构建方法
CN110046781A (zh) * 2018-12-04 2019-07-23 阿里巴巴集团控股有限公司 商户风险防控方法及装置

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050216411A1 (en) * 2004-03-29 2005-09-29 Mcquary Patrick T Payor identification at the point of transaction
US7716125B2 (en) * 2005-08-10 2010-05-11 Axcessnet Innovations Llc Networked loan market and lending management system
CN103577987A (zh) * 2012-07-20 2014-02-12 阿里巴巴集团控股有限公司 一种风险用户的识别方法和装置
US20140358789A1 (en) * 2013-05-30 2014-12-04 B. Scott Boding Acquirer facing fraud management system and method
CN104809110A (zh) * 2014-01-23 2015-07-29 安徽海汇金融投资集团有限公司 一种大数据系统
US9600819B2 (en) * 2015-03-06 2017-03-21 Mastercard International Incorporated Systems and methods for risk based decisioning
CN106295382B (zh) * 2015-05-20 2019-06-14 阿里巴巴集团控股有限公司 一种信息风险防控方法及装置
CN106611120B (zh) * 2015-10-26 2019-10-01 阿里巴巴集团控股有限公司 一种风险防控系统的评估方法及装置
TWI680427B (zh) * 2017-02-24 2019-12-21 富邦產物保險股份有限公司 企業風險評估與保險規劃系統及方法
CN107958341A (zh) * 2017-12-12 2018-04-24 阿里巴巴集团控股有限公司 风险识别方法及装置和电子设备
CN107958346A (zh) * 2017-12-14 2018-04-24 北京小度信息科技有限公司 异常行为的识别方法及装置
CN108364137A (zh) * 2018-03-12 2018-08-03 广东省科技创新监测研究中心 高新技术企业的监测方法、装置、计算机设备及存储介质
CN108564386B (zh) * 2018-04-28 2020-06-02 腾讯科技(深圳)有限公司 商户识别方法及装置、计算机设备及存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101976419A (zh) * 2010-10-19 2011-02-16 中国工商银行股份有限公司 交易数据的风险监控处理方法和系统
US20170039479A1 (en) * 2015-08-07 2017-02-09 Wistron Corporation Risk assessment system and data processing method
CN107993143A (zh) * 2017-11-23 2018-05-04 深圳大管加软件与技术服务有限公司 一种信贷风险评估方法及系统
CN108154430A (zh) * 2017-12-28 2018-06-12 上海氪信信息技术有限公司 一种基于机器学习和大数据技术的信用评分构建方法
CN110046781A (zh) * 2018-12-04 2019-07-23 阿里巴巴集团控股有限公司 商户风险防控方法及装置

Cited By (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114612104A (zh) * 2020-12-09 2022-06-10 支付宝(杭州)信息技术有限公司 风险识别方法、装置和电子设备
CN112686521A (zh) * 2020-12-25 2021-04-20 江苏通付盾科技有限公司 一种风控规则调优方法和系统
CN115115359B (zh) * 2021-03-17 2025-09-26 腾讯科技(深圳)有限公司 一种异常用户识别方法、处理方法、装置、设备及介质
CN115115359A (zh) * 2021-03-17 2022-09-27 腾讯科技(深圳)有限公司 一种异常用户识别方法、打击方法、装置、设备及介质
CN112950222A (zh) * 2021-04-08 2021-06-11 腾讯科技(深圳)有限公司 资源处理异常检测方法、装置、电子设备及存储介质
CN113254542B (zh) * 2021-04-21 2023-10-27 泰康保险集团股份有限公司 数据可视化处理方法及装置、电子设备
CN113254542A (zh) * 2021-04-21 2021-08-13 泰康保险集团股份有限公司 数据可视化处理方法及装置、电子设备
CN113626421A (zh) * 2021-08-02 2021-11-09 浪潮软件股份有限公司 一种数据校验的数据质量管控方法
CN113642637A (zh) * 2021-08-12 2021-11-12 北京沃东天骏信息技术有限公司 数据检测方法、装置及存储介质
CN113836360A (zh) * 2021-09-24 2021-12-24 支付宝(杭州)信息技术有限公司 数据检测方法及装置
CN113836360B (zh) * 2021-09-24 2024-06-11 支付宝(杭州)信息技术有限公司 数据检测方法及装置
CN114328490A (zh) * 2021-12-29 2022-04-12 浪潮卓数大数据产业发展有限公司 一种直播电商分析展现的构建方法、设备及介质
CN114321872A (zh) * 2022-01-07 2022-04-12 神华神东电力有限责任公司 锅炉保护方法及锅炉保护装置
CN114321872B (zh) * 2022-01-07 2024-01-19 神华神东电力有限责任公司 锅炉保护方法及锅炉保护装置
CN116860811A (zh) * 2022-03-25 2023-10-10 腾讯科技(深圳)有限公司 数据处理方法、装置、计算机设备及计算机可读存储介质
CN114782192A (zh) * 2022-04-27 2022-07-22 中国银行股份有限公司 银行开户的风险控制方法及系统
CN115438967A (zh) * 2022-09-06 2022-12-06 河南中原消费金融股份有限公司 一种基于超图的密集子图检测方法、装置及终端设备
CN117215948A (zh) * 2022-10-24 2023-12-12 支付宝(杭州)信息技术有限公司 变更风险防控系统、方法、电子设备及存储介质
CN115689343A (zh) * 2022-10-24 2023-02-03 万翼科技有限公司 供应商鉴别方法、装置、设备及存储介质
CN118277866A (zh) * 2024-04-24 2024-07-02 中电山河数字科技(南通)有限公司 一种用于制造行业的数据分级分类的方法和系统
CN118656708A (zh) * 2024-08-19 2024-09-17 长沙市长财科技有限公司 一种用户行为分析的支付风险评估与控制方法
CN118673445A (zh) * 2024-08-21 2024-09-20 浙江有数数智科技有限公司 基于大语言模型的异常程度识别方法、装置、介质及设备
CN119420499A (zh) * 2024-09-24 2025-02-11 联想(北京)有限公司 一种数据处理方法、装置、设备及存储介质
CN119294840A (zh) * 2024-12-11 2025-01-10 蚂蚁智信(杭州)信息技术有限公司 一种数据处理方法、装置及设备
CN119722320A (zh) * 2024-12-19 2025-03-28 广东通莞科技股份有限公司 一种基于聚合支付交易路由的移动支付监管方法与系统
CN119379425A (zh) * 2024-12-31 2025-01-28 杭银消费金融股份有限公司 一种用于逾期指标变化的归因分析方法及系统
CN119850328A (zh) * 2024-12-31 2025-04-18 交通银行股份有限公司 信用卡风险类别确定方法、装置、设备、介质及程序产品
CN119850328B (zh) * 2024-12-31 2025-11-07 交通银行股份有限公司 信用卡风险类别确定方法、装置、设备、介质及程序产品
CN120198157A (zh) * 2025-03-17 2025-06-24 浙江瑞耀信息技术有限公司 一种智能商户数据分析方法与系统

Also Published As

Publication number Publication date
TWI714262B (zh) 2020-12-21
TW202022749A (zh) 2020-06-16
CN110046781B (zh) 2020-07-07
CN110046781A (zh) 2019-07-23

Similar Documents

Publication Publication Date Title
WO2020114110A1 (zh) 商户风险防控方法及装置
US11977995B2 (en) Machine learning artificial intelligence system for predicting hours of operation
Kogan et al. Design and evaluation of a continuous data level auditing system
Kahyaoğlu et al. Continuous auditing as a strategic tool in public sector internal audit: The Turkish case
JP2020523715A (ja) 不正取引の改善された検出
WO2020103560A1 (zh) 风险控制方法、装置、服务器及存储介质
CN108876188B (zh) 一种间连服务商风险评估方法及装置
CN103123712A (zh) 一种网络行为数据的监控方法和系统
CN117714722B (zh) 一种电商直播购物的数据分析方法及系统
CN117237126B (zh) 一种保险平台及保险数据处理方法
CN113095931A (zh) 贷后风险监控方法、装置及计算机设备
CN114757765A (zh) 一种基于评级模型的信用评级方法及装置
CN117151882B (zh) 一种基于多品种电力交易的风险评估方法及系统
CN118552327A (zh) 一种汽车保险业务的数据分析方法、系统及存储介质
CN117132383A (zh) 一种信贷数据处理方法、装置、设备及可读存储介质
CN110288038A (zh) 一种企业的分类方法及装置
CN119250859B (zh) 一种基于大数据的供应商供应数据模型构建方法及系统
CN113298575A (zh) 一种商标价值批量评估的方法、系统、设备及存储介质
US12567116B2 (en) Tax-advantaged business ownership transfer system
Suratova EMPIRICAL ANALYSIS OF TOURISM FLOW FORECASTING IN CENTRAL ASIA BASED REGRESSION MODELS
Khang 3 Internet of Things
CN120634734A (zh) 客户风险等级评估方法、装置、计算机设备及存储介质
Li et al. Quantifying data manipulation of food inspection agencies: evidence from food safety supervision sampling in China
CN120707273A (zh) 信用卡审批方法、装置、设备、介质和程序产品
CN120106862A (zh) 一种对象异常经营状态的预测方法、装置及设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19892839

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19892839

Country of ref document: EP

Kind code of ref document: A1