CN109816234A - Service access method, service access device, electronic equipment and storage medium - Google Patents
Service access method, service access device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN109816234A CN109816234A CN201910044398.1A CN201910044398A CN109816234A CN 109816234 A CN109816234 A CN 109816234A CN 201910044398 A CN201910044398 A CN 201910044398A CN 109816234 A CN109816234 A CN 109816234A
- Authority
- CN
- China
- Prior art keywords
- serviced
- result
- risk
- debt
- probability
- 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.)
- Pending
Links
Landscapes
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
Present disclose provides a kind of service access method, service access device, electronic equipment and computer readable storage mediums, belong to field of computer technology.This method comprises: obtaining the characteristic of object to be serviced;Credit evaluation is carried out to the object to be serviced based on the characteristic, obtains the credit evaluation result of the object to be serviced;Response assessment is carried out to the object to be serviced based on the characteristic, obtains the response assessment result of the object to be serviced;Object to be serviced described in access is determined whether according to the credit evaluation result of the object to be serviced and response assessment result.The accuracy of object access result to be serviced can be improved in the disclosure, reduces service risk, and reduce dependence of the access review process for external data.
Description
Technical field
This disclosure relates to which field of computer technology more particularly to a kind of service access method, service access device, electronics are set
Standby and computer readable storage medium.
Background technique
With the rapid development of computer technology, in order to adapt to economic new demand, internet is also got in every field
To be more widely applied.Wherein, financial product, application loan are bought, credit card is handled or receives other services by internet
User it is more and more, in order to ensure the interests of each mechanism, it is necessary to be audited to the access of service object.
Existing service access method, generally use manual examination and verification mode judge service object whether can access, but
It is this method human cost with higher, and cannot be guaranteed the objectivity and accuracy of auditing result, to service organization
Cause risk;In addition, need to treat service object by external data toward contact and assess when auditing service object, from
And the time needed for increasing audit, review efficiency is reduced, the enthusiasm that user receives service is influenced.
It should be noted that information is only used for reinforcing the reason to the background of the disclosure disclosed in above-mentioned background technology part
Solution, therefore may include the information not constituted to the prior art known to persons of ordinary skill in the art.
Summary of the invention
Present disclose provides a kind of service access method, service access device, electronic equipment and computer-readable storage mediums
Matter, and then overcome existing service access method that there is risk and strong to external data dependency at least to a certain extent
The problem of.
Other characteristics and advantages of the disclosure will be apparent from by the following detailed description, or partially by the disclosure
Practice and acquistion.
According to one aspect of the disclosure, a kind of service access method is provided, comprising: obtain the characteristic of object to be serviced
According to;Credit evaluation is carried out to the object to be serviced based on the characteristic, obtains the credit evaluation of the object to be serviced
As a result;Response assessment is carried out to the object to be serviced based on the characteristic, the response for obtaining the object to be serviced is commented
Estimate result;According to the credit evaluation result of the object to be serviced with response assessment result determine whether it is to be serviced right described in access
As.
It is described that the object to be serviced is carried out based on the characteristic in a kind of exemplary embodiment of the disclosure
Credit evaluation obtains the credit evaluation result of the object to be serviced, comprising: passes through risk analysis model and analysis of being in debt respectively
Model handles the characteristic, obtains the risk analysis result of the object to be serviced and analysis result of being in debt;
The credit evaluation knot of the object to be serviced is obtained according to the risk analysis result of the object to be serviced and analysis result of being in debt
Fruit.
In a kind of exemplary embodiment of the disclosure, the method also includes: obtain multiple sample objects and each institute
State the sample characteristics data of sample object;Obtain the first tag along sort and the second tag along sort of the sample object;Using institute
Sample characteristics data and first tag along sort are stated, the first machine learning model of training obtains the risk analysis model;
Using the sample characteristics data and second tag along sort, the second machine learning model of training obtains described be in debt and divides
Analyse model;Wherein, first tag along sort includes high risk object or non-high risk object, and second tag along sort includes
High liabilities object or non-high liabilities object.
In a kind of exemplary embodiment of the disclosure, first machine learning model is with the second machine learning model
Gradient promotes decision-tree model, and the risk analysis result includes high risk probability, and the debt analysis result includes high liabilities
Probability;It is described that the characteristic is handled by risk analysis model and debt analysis model respectively, obtain it is described to
The risk analysis result of service object and analysis result of being in debt include: the N decision in the face of risk by the risk analysis model
Tree handles the characteristic, obtains N number of 1 or 0 high risk classification value, wherein 1 indicates that prediction is described to be serviced right
As indicating to predict that the object to be serviced is non-high risk object for high risk object, 0;Pass through formula
The high risk probability is calculated, wherein PrFor the high risk probability, WriFor the weight coefficient of i-th decision in the face of risk tree, TriFor
The high risk classification value of i-th decision in the face of risk tree output;By M debt decision tree of the debt analysis model to the spy
Sign data are handled, and M 1 or 0 high liabilities classification value is obtained, wherein 1 indicates to predict that the object to be serviced is high liabilities
Object, 0 indicates to predict that the object to be serviced is non-high liabilities object;Pass through formulaDescribed in calculating
High liabilities probability, wherein PdFor the high liabilities probability, WdjFor the weight coefficient of jth debt decision tree, TdjFor jth debt
The high liabilities classification value of decision tree output.
In a kind of exemplary embodiment of the disclosure, the credit evaluation result includes credit appraisal value, the response
Assessment result includes response rate;The credit evaluation result according to the object to be serviced determines whether with response assessment result
Object to be serviced described in access includes: the promise breaking that the object to be serviced is calculated according to the credit appraisal value and the response rate
Prediction probability;If the violation correction probability of the object to be serviced reaches a probability threshold value, refuse to be serviced described in access
Object;If the violation correction probability of the object to be serviced is not up to the probability threshold value, object to be serviced described in access.
It is described that institute is calculated according to the credit appraisal value and the response rate in a kind of exemplary embodiment of the disclosure
The violation correction probability for stating object to be serviced includes: the violation correction probability for being calculated by the following formula the object to be serviced:
Pq=Pm a·Pn b, wherein PqFor the violation correction probability, PmFor the credit appraisal value, PnFor the response rate, a is equal with b
For the constant parameter greater than 0.
It is described that the object to be serviced is carried out based on the characteristic in a kind of exemplary embodiment of the disclosure
Response assessment, the response assessment result for obtaining the object to be serviced includes: by be serviced described in response prediction model prediction
Object obtains the response assessment result of the object to be serviced;Wherein, the response assessment result include high response object with it is non-
High response object.
According to one aspect of the disclosure, a kind of service access device is provided, comprising: data acquisition module, for obtaining
The characteristic of object to be serviced;First evaluation module, for carrying out letter to the object to be serviced based on the characteristic
With assessment, the credit evaluation result of the object to be serviced is obtained;Second evaluation module, for being based on the characteristic to institute
It states object to be serviced and carries out response assessment, obtain the response assessment result of the object to be serviced;Access judgment module is used for root
Object to be serviced described in access is determined whether according to the credit evaluation result and response assessment result of the object to be serviced.
In a kind of exemplary embodiment of the disclosure, the first evaluation module includes: model analysis unit, for leading to respectively
It crosses risk analysis model and debt analysis model handles the characteristic, obtain the risk point of the object to be serviced
Analyse result and analysis result of being in debt;As a result acquiring unit, for according to the risk analysis result of the object to be serviced and negative
Debt analysis result obtains the credit evaluation result of the object to be serviced.
In a kind of exemplary embodiment of the disclosure, access device is serviced further include: sample object obtains module, is used for
Obtain the sample characteristics data of multiple sample objects and each sample object;Label acquisition module, for obtaining the sample
First tag along sort and the second tag along sort of this object;First training module, for using the sample characteristics data and
First tag along sort, the first machine learning model of training, obtains the risk analysis model;Second training module, is used for
Using the sample characteristics data and second tag along sort, the second machine learning model of training obtains described be in debt and divides
Analyse model;Wherein first tag along sort includes high risk object or non-high risk object, and second tag along sort includes
High liabilities object or non-high liabilities object.
In a kind of exemplary embodiment of the disclosure, first machine learning model is with the second machine learning model
Gradient promotes decision-tree model, and the risk analysis result includes high risk probability, and the debt analysis result includes high liabilities
Probability;Model analysis unit includes: the first processing subelement, for the N decision in the face of risk tree by the risk analysis model
The characteristic is handled, N number of 1 or 0 high risk classification value is obtained, wherein 1 indicates to predict the object to be serviced
It indicates to predict that the object to be serviced is non-high risk object for high risk object, 0;First computation subunit, for passing through public affairs
FormulaThe high risk probability is calculated, wherein PrFor the high risk probability, WriFor i-th decision in the face of risk
The weight coefficient of tree, TriFor the high risk classification value of i-th decision in the face of risk tree output;Second processing subelement, for passing through
M debt decision tree for stating debt analysis model handles the characteristic, obtains M 1 or 0 high liabilities classification
Value, wherein 1 indicates to predict that the object to be serviced is high liabilities object, 0 indicates to predict that the object to be serviced is non-high liabilities
Object;Second computation subunit, for passing through formulaThe high liabilities probability is calculated, wherein PdFor
The high liabilities probability, WdjFor the weight coefficient of jth debt decision tree, TdjFor the high liabilities of jth debt decision tree output
Classification value.
In a kind of exemplary embodiment of the disclosure, the credit evaluation result includes credit appraisal value, the response
Assessment result includes response rate;Access judgment module includes: probability calculation unit, for according to the credit appraisal value with it is described
Response rate calculates the violation correction probability of the object to be serviced;Threshold decision unit, if for the object to be serviced
Violation correction probability reaches a probability threshold value, then refuses object to be serviced described in access;If the promise breaking of the object to be serviced
Prediction probability is not up to the probability threshold value, then object to be serviced described in access.
In a kind of exemplary embodiment of the disclosure, probability calculation unit is described wait take for being calculated by the following formula
The violation correction probability of business object: Pq=Pm a·Pn b, wherein PqFor the violation correction probability, PmFor the credit appraisal value,
PnFor the response rate, a and b is the constant parameter greater than 0.
In a kind of exemplary embodiment of the disclosure, the second evaluation module is used for by described in response prediction model prediction
Object to be serviced obtains the response assessment result of the object to be serviced;Wherein, the response assessment result includes high response pair
As with non-high response object.
According to one aspect of the disclosure, a kind of electronic equipment is provided, comprising: processor;And memory, for storing
The executable instruction of the processor;Wherein, the processor is configured to above-mentioned to execute via the executable instruction is executed
Method described in any one.
According to one aspect of the disclosure, a kind of computer readable storage medium is provided, computer program is stored thereon with,
The computer program realizes method described in above-mentioned any one when being executed by processor.
The exemplary embodiment of the disclosure has the advantages that
In the example embodiments that the disclosure provides, the spy of service object is treated by using credit evaluation and response assessment
Sign data are handled, and obtain the credit evaluation result and response assessment result of object to be serviced, and based thereon determine whether access
Object to be serviced.It on the one hand, can be by credit standing or response by analyzing the credit standing and responsive status of object to be serviced
The object range that the poor object of situation removes destination service reduces service wind to improve the accuracy of auditing result
Danger, has ensured the interests of service organization.On the other hand, the row based on object to be serviced in the website of this service organization or App
For log statistic characteristic, and credit and response analysis are carried out, dependence of the review process for external data can be reduced, and
Shorten auditing flow, promotes user experience.
It should be understood that above general description and following detailed description be only it is exemplary and explanatory, not
The disclosure can be limited.
Detailed description of the invention
The drawings herein are incorporated into the specification and forms part of this specification, and shows the implementation for meeting the disclosure
Example, and together with specification for explaining the principles of this disclosure.It should be evident that the accompanying drawings in the following description is only the disclosure
Some embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.
Fig. 1 schematically shows a kind of flow chart for servicing access method in the present exemplary embodiment;
Fig. 2 schematically shows a kind of sub-process figure for servicing access method in the present exemplary embodiment;
Fig. 3 schematically shows the flow chart of another service access method in the present exemplary embodiment;
Fig. 4 schematically shows a kind of structural block diagram for servicing access device in the present exemplary embodiment;
Fig. 5 schematically shows a kind of electronic equipment for realizing the above method in the present exemplary embodiment;
Fig. 6 schematically shows a kind of computer-readable storage medium for realizing the above method in the present exemplary embodiment
Matter.
Specific embodiment
Example embodiment is described more fully with reference to the drawings.However, example embodiment can be with a variety of shapes
Formula is implemented, and is not understood as limited to example set forth herein;On the contrary, thesing embodiments are provided so that the disclosure will more
Fully and completely, and by the design of example embodiment comprehensively it is communicated to those skilled in the art.Described feature, knot
Structure or characteristic can be incorporated in any suitable manner in one or more embodiments.
The exemplary embodiment of the disclosure provides firstly a kind of service access method, wherein service can be each service
Mechanism provides the service with certain default risk, such as electronic money management, online wallet, debt-credit service, credit by internet
Card application etc.;Correspondingly, above-mentioned service towards personal user or enterprise customer can be the service object of the present embodiment.This
Embodiment can be applied to also can be applied to in the scene of object progress access audit for applying for above-mentioned service as above-mentioned clothes
In the scene of business screening potential object.The executing subject of the present embodiment can be the server of service organization.
1 pair of present exemplary embodiment is described further with reference to the accompanying drawing, as shown in Figure 1, the service access method can
To include the following steps S110~S140:
Step S110 obtains the characteristic of object to be serviced.
Wherein, object to be serviced can be the user for having applied for destination service, be also possible to the user by other accesses
The user etc. recommended;Characteristic refers to many aspects such as reflection object behavior habit to be serviced, consumption habit, asset transition
The data of feature usually have multiple indexs, such as: it is counted according to the daily login time of object to be serviced, downtime
Attribute is logged in, its product preference, consumption frequency, spending limit etc. are counted according to the history consumer record of object to be serviced.At this
In exemplary embodiment, the statistics of characteristic can be carried out according to the user behaviors log of object to be serviced, wherein user behaviors log is
Refer to the original record of object to be serviced all or part of behavior in the website of this service organization or App (application program), such as:
The daily login time of object to be serviced, downtime, history consumer record etc..
Step S120 treats service object based on characteristic and carries out credit evaluation, and the credit for obtaining object to be serviced is commented
Estimate result.
Credit evaluation can be the credit record for treating service object, economic situation, debt paying ability, contract compliance ability etc.
Many aspects relevant to credit are assessed, and the available performance of credit evaluation object credit level to be serviced and credit are passed through
Situation as a result, i.e. credit evaluation result.In the present example embodiment, the characteristic based on object to be serviced, Ke Yitong
It crosses specific mathematical formulae or credit evaluation is calculated as a result, can also handling by machine learning model in function model
To credit evaluation result.Furthermore, it is possible to assessed respectively according to many aspects of above-mentioned credit evaluation, then comprehensive various aspects
Assessment result obtains final credit evaluation as a result, can for example distinguish the credit record of object to be serviced, economic situation, payment of debts
Ability, contract compliance ability etc. evaluate level value, then each level value weighted calculation is obtained credit evaluation result.Credit evaluation
As a result it can be expressed as numeric form, such as credit value, Default Probability etc., the form of classification results can also be expressed as, such as
Whether be high credit object, credit standing be it is excellent/good/in/difference etc., this implementation is not specially limited this.
Step S130 treats service object based on characteristic and carries out response assessment, and the response for obtaining object to be serviced is commented
Estimate result;
Response assessment can be the assessment treated service object's responsive status and enliven the information such as situation, to a certain extent
Reflect that object to be serviced can assess the viscosity of service organization based on characteristic relevant to response, such as clear
Look at service organization's correlation pushed information the frequency, using service App using duration or service App in mutual dynamic frequency
Deng.In the present example embodiment, it can be responded by specific mathematical formulae, function model or machine learning model etc.
Assessment.Response assessment result can be expressed as numeric form, such as response rate, liveness etc., can also be expressed as classification results
Form, for example whether be high response object, responsive status be it is active/general/inactive etc., this implementation is not done this especially
It limits.
Step S140 determines whether that access is to be serviced according to the credit evaluation result of object to be serviced and response assessment result
Object.
In the present example embodiment, credit evaluation result can be integrated with response assessment result to determine whether that access waits for
Service object.For example, if credit evaluation result and response results are two classification results, it can be concluded that at least four points
Class result: Gao Xinyong object increases response object, Gao Xinyong object Jia Feigao response object, and non-high credit object increases response pair
As non-high credit object Jia Feigao response object.It can set only when object to be serviced increases response object as high credit object
When, allow to service access, can also set when object to be serviced belongs to above-mentioned rear three kinds of classification, obtain servicing access accordingly
As a result.In practical application, different judgment criterias is selected according to the concrete condition of service organization, such as the risk being related to when service
When lower, the object to be serviced standard that high credit object increases response object and high credit object Jia Feigao response object can permit
Enter, with the relevance factor for the service of improving.If credit evaluation result and response assessment result are numerical value, judgment criteria can be pass
When the credit evaluation result and response assessment result of the standard of numerical value, such as object to be serviced are all higher than certain threshold value, really
It fixes into object to be serviced.
You need to add is that credit evaluation result and response assessment result can be it is various forms of as a result, such as credit
Assessment result is numeric type as a result, response assessment result is classification results, then the judgment criteria of two kinds of forms can be respectively adopted
Judged respectively, when credit evaluation result and response assessment result all meet judgment criteria when, it can be determined that access is to be serviced
Object.
In the present example embodiment, since credit evaluation and response are evaluated as two different evaluation systems, credit is commented
Estimate result and response assessment result is unaffected mutually, therefore, can first carry out the credit evaluation in step S120, then walked
Response assessment in rapid S130 can also first carry out the response assessment in step S130, then the credit carried out in step S120 is commented
Estimate, step S120 and step S130 can also be carried out simultaneously, is according to credit evaluation result and response assessment result finally
The result of no access object to be serviced.The sequence of step S120 and step S130 is not specifically limited herein.
Based on above description, the feature of service object is treated by credit evaluation and response assessment in the present exemplary embodiment
Data are handled, and obtain the credit evaluation result and response assessment result of object to be serviced, and based thereon determine whether that access waits for
Service object's access.It on the one hand, can be from credit and response two by analyzing the credit standing and responsive status of object to be serviced
A angle carries out assessment audit, according to actual needs, the poor object of credit standing or responsive status is excluded service object's model
It encloses, to improve the accuracy of access result, reduces service risk, ensured the interests of service organization.On the other hand, special
Sign data are usually the data of the object to be serviced of service organization's internal control, such as can be from the behavior day in website or App
Will counts to obtain characteristic, and the present exemplary embodiment is based on characteristic and carries out access audit, can reduce for external number
According to dependence, and shorten the auditing flow of access, promote user experience.
In one exemplary embodiment, step S120 may comprise steps of:
Characteristic is handled by risk analysis model and debt analysis model respectively, obtains object to be serviced
Risk analysis result and analysis result of being in debt;
The credit evaluation knot of object to be serviced is obtained according to the risk analysis result of object to be serviced and analysis result of being in debt
Fruit.
Wherein, risk analysis model can be handled characteristic, obtain the risk status of object to be serviced, i.e. wind
Danger analysis result.Such as: risk analysis model can filter out index relevant to risk from characteristic, and refer to these
Target data are calculated, and risk analysis result is obtained;Object to be serviced can also be predicted based on the variation tendency of characteristic
Following behavior, and risk analysis result is obtained according to behavior;Debt analysis model can also be handled characteristic, be obtained
To the debt situation of object to be serviced, that is, analysis of being in debt is as a result, its processing mode can be similar with risk analysis model.This implementation
Example for above-mentioned model specific processing mode without limitation.Risk analysis result can be classification results, such as analyze wait take
Object be engaged in as high risk or non-high risk, is also possible to special value, such as analyze the risk index of object to be serviced, Gao Feng
Dangerous probability etc..Analysis result of being in debt is similar with risk analysis result, can be classification results, such as analyzes object to be serviced as height
Perhaps non-high liabilities etc. of being in debt can be special value, such as analyze the debt index of object to be serviced, high liabilities probability
Deng.
In the present example embodiment, the characteristic application risk analysis model of service object is treated and analysis mould of being in debt
Type is handled, if obtained risk analysis result and analysis result of being in debt are classification results, it can be concluded that at least four
Classification results: high risk object increases debt object, high risk object Jia Fei high liabilities object, and non-high risk object increases debt
Object, non-high risk object Jia Fei high liabilities object, in addition, risk analysis result and analysis result of being in debt are also possible to certain number
Value.
You need to add is that risk analysis result and be in debt analysis result can be it is various forms of as a result, such as risk
Analysis result is risk index, and the classification results that analysis result is high liabilities object or non-high liabilities object of being in debt can be distinguished
Credit evaluation result is obtained using the standard of two kinds of forms.
In one exemplary embodiment, refering to what is shown in Fig. 2, service access method can also include step S210~S240:
Step S210 obtains the sample characteristics data of multiple sample objects and each sample object.
Step S220 obtains the first tag along sort and the second tag along sort of sample object.
Step S230, using sample characteristics data and the first tag along sort, the first machine learning model of training obtains wind
Dangerous analysis model.
Step S240, using sample characteristics data and the second tag along sort, the second machine learning model of training is born
Debt analysis model.
Wherein, the first tag along sort includes high risk object or non-high risk object, the second tag along sort include high liabilities
Object or non-high liabilities object.
In the present embodiment, sample object can be the use for having the external data of reflection risk status and debt situation
Family, such as the user of people's row collage-credit data has been provided, the user for having licensed other App data, has been filed on assets proof
User, the user for having carried out risk questionnaire investigation etc..The user behaviors log of sample object refers to above-mentioned user in this service organization
Website or App in behavior original record, identical index can therefrom be counted by mode identical with step S110
Sample characteristics data.
It can be that sample object manually marks the first tag along sort and the second classification according to external data in step S220
Label.Wherein, the first tag along sort may include high risk object or non-high risk object.High risk object can be payment energy
Power or credit are poor, cause to have not in the user for founding the interests of service organization high risk, such as people's row collage-credit data
The user of good record, by the user that other service organizations evaluation credit is poor, the very relatively low or higher user of risk partiality is lost
Industry, nothing are just when unstable user of occupation or revenue source etc..Second tag along sort may include high liabilities object or non-height
Debt object, high liabilities object can be the current or more serious user of debt situation in a short time, for example, bear great number housing loan,
The user that vehicle is borrowed has the user of borrowing plan in a short time, and the frozen user of property, assets ownership has the user of dispute under one's name
Etc..It, can also in advance externally when marking the first tag along sort and the second tag along sort to sample object by external data
Portion's data carry out certain cleaning, mapping and statistics, can also be by being different from above-mentioned first machine learning model or the second machine
The other model to external data of device learning model are analyzed and processed, and obtain the first tag along sort and second of sample object
Tag along sort etc., the present embodiment are not specially limited this.
According to the sample characteristics data and the first tag along sort of sample object, the first machine learning model can be trained, is instructed
Practicing process may include: the first machine learning model with sample characteristics data for input, and output sample object is high risk object
Or the classification results of non-high risk object can make the classification results of output become closer to classify by adjusting model parameter
Label, until the accuracy rate of model reaches certain standard, it is believed that training is completed.Second machine learning model was trained
Journey is similar with the training process of the first machine learning model.To obtain the first machine learning model and the second machine learning mould
Type can call directly during servicing access.
In one exemplary embodiment, the first machine learning model and the second machine learning model may include that gradient is promoted
Decision-tree model, Random Forest model or Logic Regression Models.
Wherein, the first machine learning model and the second machine learning model can be same type of machine learning model,
It is also possible to different types of machine learning model.Above-mentioned three classes machine learning model can carry out the characteristic of multivariable
Processing, obtains continuous output numerical value or discrete classification results
Further, in one exemplary embodiment, the first machine learning model can be with the second machine learning model
Gradient promotes decision-tree model, and risk analysis result can be high risk probability, and analysis result of being in debt can be high liabilities probability;
The risk analysis result of object to be serviced and analysis result of being in debt can be obtained by following steps in step S120:
Characteristic is handled by N decision in the face of risk tree of risk analysis model, obtains N number of 1 or 0 high risk
Classification value, wherein 1 indicates to predict that object to be serviced is high risk object, 0 indicates to predict that object to be serviced is non-high risk object;
Pass through formulaHigh risk probability is calculated, wherein PrFor high risk probability, WriFor i-th wind
The weight coefficient of dangerous decision tree, TriFor the high risk classification value of i-th decision in the face of risk tree output;
Characteristic is handled by M debt decision tree of debt analysis model, obtains M 1 or 0 high liabilities
Classification value, wherein 1 indicates to predict that object to be serviced is high liabilities object, 0 indicates to predict that object to be serviced is non-high liabilities object;
Pass through formulaHigh liabilities probability is calculated, wherein PdFor high liabilities probability, WdjFor jth
The weight coefficient of debt decision tree, TdjFor the high liabilities classification value of jth debt decision tree output.
Wherein, high risk probability and high liabilities probability, which can be, reflects that high risk or high liabilities situation occurs in object to be serviced
A possibility that size measurement.Gradient promotes decision-tree model and generally comprises in more decision trees, such as the present embodiment, risk point
It, can be according to formula for model is analysed comprising N decision in the face of risk treeCalculate the high wind of object to be serviced
Dangerous probability.The weight coefficient W of each decision in the face of risk treeriIt can specifically be calculated by a variety of methods, lift two explanations below:
(1)、Wri=1/N, wherein WriFor weight, N is the total quantity of decision in the face of risk tree, i.e., each decision in the face of risk tree can be waited
Weight;
(2)、Wherein WriFor the weight of i-th decision in the face of risk tree, R (i) and R (k) are respectively i-th
The accuracy rate of decision in the face of risk tree and kth decision in the face of risk tree, N are the total quantity of decision in the face of risk tree, i, k ∈ [1, N];
The present embodiment is not specially limited the method for calculating each decision in the face of risk tree weight coefficient.
Pass through formulaHigh liabilities probability can be calculated, with above-mentioned high risk probability calculation
It is similar, therefore repeat no more.
In one exemplary embodiment, credit evaluation result may include credit appraisal value, and response assessment result can wrap
Include response rate;Step S140 may comprise steps of:
The violation correction probability of object to be serviced is calculated according to credit appraisal value and response rate;
If the violation correction probability of object to be serviced reaches a probability threshold value, refuse access object to be serviced;
If the violation correction probability of object to be serviced is not up to probability threshold value, access object to be serviced.
Wherein, credit appraisal value is used to reflect the credit level and credit standing of object to be serviced, can be with scoring or hundred
Divide the performance of the numeric forms such as ratio.Response rate is used to reflect the responsive status of object to be serviced or enlivens situation, usually with percentage
Form embodies.It has been generally acknowledged that credit appraisal value and response rate are index independent of each other, two indices can be based respectively on and treated
Service object assesses, but in some cases, this mode possibly can not make effective judgement, such as setting credit
Evaluation of estimate and response rate are above 80% object access, then the user that credit appraisal value is 90%, response rate is 70% cannot
Access, although the comprehensive condition of latter user may be 80% user better than credit appraisal value and response rate.Therefore, may be used
To integrate credit appraisal value and response rate, by calculate violation correction probability, come judge object to be serviced whether access.In addition,
The Default Probability threshold value of judgment criteria can be provided as in violation correction probabilistic determination mechanism, if such as setting Default Probability
Threshold value is 80%, can be with access object to be serviced when the resulting violation correction probability of calculating is not up to 80%.Default Probability threshold
Value can rule of thumb initialization, and adjusting is optimized according to result feedback in use, allows to standard
Whether true prediction object to be serviced can break a contract, to obtain accurate access result.
Have much by the specific method that credit appraisal value and response rate calculate violation correction probability, such as can be by high wind
Dangerous probability and high liabilities probability multiplication, can also average to two probability.In one exemplary embodiment, according to credit
The violation correction probability that evaluation of estimate calculates object to be serviced with response rate can also include:
It is calculated by the following formula the violation correction probability of object to be serviced:
Pq=Pm a·Pn b, wherein PqFor violation correction probability, PmFor credit appraisal value, PnFor response rate, a and b are big
In 0 constant parameter.
In the violation correction of object to be serviced, the difference of credit appraisal value and response rate, it is to be serviced right to will affect
The no judging result by audit is liked, since the credit appraisal value economic situation current with object to be serviced and risk status have
Close, and response rate be treat service object whether a kind of active judgement, therefore, credit appraisal value has more compared to response rate
There is reference value.Such as the object and credit appraisal value that credit appraisal value is 80%, response rate is 50% are 50%, response rate is
80% compares, and the probability of the latter's promise breaking may be larger.According to above-mentioned analysis, formula P can useq=Pm a·Pn bIt is pre- to calculate promise breaking
Probability is surveyed, is credit appraisal value PmWith response rate PnThe constant parameter a and b of exponential form is respectively set, wherein a and b needs are greater than
0, numerical value both usually, can be in order to control importance of the credit appraisal value in violation correction probability calculation 0.5 or so
A < b is set.It in practical applications, can also be according to the numerical value of result feedback optimizing regulation a and b.
In one exemplary embodiment, after obtaining credit appraisal value and response rate, two numerical value can be carried out respectively
Normalized is in same numerical value level, is conducive to subsequent calculating violation correction probability.
In one exemplary embodiment, step S130 may include:
By response prediction model prediction object to be serviced, the response assessment result of object to be serviced is obtained;
Wherein, response assessment result includes high response object and non-high response object.
Response prediction model is the model that its response rate is analyzed by the characteristic of object to be serviced, such as can be mind
Through machine learning models such as network model, Logic Regression Models, Random Forest models, it is also possible to function of many variables model etc., this
Embodiment is not specially limited this.It is similar with above-mentioned risk analysis model or debt analysis model, sample number can be passed through
According to the parameter of training response prediction model or calculating response prediction model, existing response prediction model can also be directly used
Deng this will not be repeated here.
High response object can be the relatively high user of the liveness in service organization, such as often browsing service organization's phase
The user of pass pushed information uses service App time longer user or the higher user of mutual dynamic frequency in service App
Deng.In view of high response object indicates that the object is higher to the viscosity of service organization to a certain extent, in this exemplary implementation
, can be in the case where passing through credit evaluation in example, then service object is treated by response prediction model and is further judged,
I.e. eventually by object have that credit standing is preferable and the preferable condition of responsive status, auditing result can be further increased
Accuracy rate.
In one exemplary embodiment, it is also possible to treat service object first with response prediction model and carries out response and comments
Estimate, after filtering out high response object, then carry out credit evaluation, after credit evaluation, available responsive status is good, and believes
It is first since good responsive status is easier to reach compared to good credit standing with object to be serviced in order
The high response user of screening, for access, more satisfactory object provides possibility.
Fig. 3 show in the present exemplary embodiment it is a kind of service access method flow chart, step S310, based on obtain from
The characteristic of object to be serviced, application risk analysis model and debt point can be distinguished by carrying out step S320 and step S330
Analysis model is calculated, and the high risk probability and high liabilities probability of object to be serviced in step S321 and step S331 are obtained,
Pass through the credit appraisal value of the high risk probability and high liabilities probability calculation object to be serviced that obtain, step in step S340
In S350, then judge whether credit appraisal value reaches a preset threshold.It, can be with if credit appraisal value is not up to preset threshold
Step S380 is carried out, access object to be serviced is refused;If credit appraisal value reaches preset threshold, step S360 can be carried out
The characteristic of object to be serviced is inputted into response prediction model, predicts whether object to be serviced is high response in step S361
Object.If prediction result be it is no, refuse access object to be serviced;If prediction result be it is yes, carry out step S370, it is quasi-
Enter object to be serviced, so as to complete the overall process of audit.
The example embodiments of the disclosure additionally provide a kind of service access device.Referring to Fig. 4, which can
To include data acquisition module 410, the first evaluation module 420, the second evaluation module 430 and access judgment module 440.Wherein,
Data acquisition module 410, for obtaining the characteristic of object to be serviced;First evaluation module 420, for being based on characteristic
It treats service object and carries out credit evaluation, obtain the credit evaluation result of object to be serviced;Second evaluation module 430 is used for base
Service object is treated in characteristic and carries out response assessment, obtains the response assessment result of object to be serviced;Access judgment module
440, for determining whether access object to be serviced according to the credit evaluation result and response assessment result of object to be serviced.
In one exemplary embodiment, the first evaluation module may include: model analysis unit, for passing through risk respectively
Analysis model and debt analysis model handle characteristic, obtain the risk analysis result and debt of object to be serviced
Analyze result;As a result acquiring unit, for being obtained with analysis result of being in debt wait take according to the risk analysis result of object to be serviced
The credit evaluation result of business object.
In one exemplary embodiment, service access device can also include: that sample object obtains module, more for obtaining
The sample characteristics data of a sample object and each sample object;Label acquisition module, for obtaining first point of sample object
Class label and the second tag along sort;First training module, for utilizing sample characteristics data and the first tag along sort, training the
One machine learning model, obtains risk analysis model;Second training module, for utilizing sample characteristics data and the second classification
Label, the second machine learning model of training, obtains debt analysis model;Wherein the first tag along sort includes high risk object or non-
High risk object, the second tag along sort include high liabilities object or non-high liabilities object.
In one exemplary embodiment, the first machine learning model and the second machine learning model are that gradient promotes decision tree
Model, risk analysis result include high risk probability, and analysis result of being in debt includes high liabilities probability;Model analysis unit can wrap
Include: the first processing subelement obtains N for handling by N decision in the face of risk tree of risk analysis model characteristic
A 1 or 0 high risk classification value, wherein 1 indicates to predict that object to be serviced is high risk object, 0 indicates to predict object to be serviced
For non-high risk object;First computation subunit, for passing through formulaHigh risk probability is calculated, wherein
PrFor high risk probability, WriFor the weight coefficient of i-th decision in the face of risk tree, TriFor the high risk of i-th decision in the face of risk tree output
Classification value;Second processing subelement, for being handled by M debt decision tree of debt analysis model characteristic,
M 1 or 0 high liabilities classification value is obtained, wherein 1 indicates to predict that object to be serviced is high liabilities object, 0 indicates that prediction is to be serviced
Object is non-high liabilities object;Second computation subunit, for passing through formulaIt is general to calculate high liabilities
Rate, wherein PdFor high liabilities probability, WdjFor the weight coefficient of jth debt decision tree, TdjFor the output of jth debt decision tree
High liabilities classification value.
In one exemplary embodiment, credit evaluation result may include credit appraisal value, and response assessment result includes ringing
It should rate;Access judgment module includes: probability calculation unit, for calculating object to be serviced according to credit appraisal value and response rate
Violation correction probability;Threshold decision unit is refused if the violation correction probability for object to be serviced reaches a probability threshold value
Exhausted access object to be serviced;If the violation correction probability of object to be serviced is not up to probability threshold value, access object to be serviced.
In one exemplary embodiment, probability calculation unit can be used for being calculated by the following formula disobeying for object to be serviced
About prediction probability: Pq=Pm a·Pn b, wherein PqFor violation correction probability, PmFor credit appraisal value, PnFor response rate, a is with b
Constant parameter greater than 0.
In one exemplary embodiment, the second evaluation module can be used for be serviced right by response prediction model prediction
As obtaining the response assessment result of object to be serviced;Wherein, response assessment result includes high response object and non-high response pair
As.
The detail of above-mentioned each module/unit has been described in detail in the embodiment of method part, therefore no longer superfluous
It states.
The exemplary embodiment of the disclosure additionally provides a kind of electronic equipment that can be realized the above method.
Person of ordinary skill in the field it is understood that various aspects of the disclosure can be implemented as system, method or
Program product.Therefore, various aspects of the disclosure can be with specific implementation is as follows, it may be assumed that complete hardware embodiment, complete
The embodiment combined in terms of full Software Implementation (including firmware, microcode etc.) or hardware and software, can unite here
Referred to as circuit, " module " or " system ".
The electronic equipment 500 of this exemplary embodiment according to the disclosure is described referring to Fig. 5.What Fig. 5 was shown
Electronic equipment 500 is only an example, should not function to the embodiment of the present disclosure and use scope bring any restrictions.
As shown in figure 5, electronic equipment 500 is showed in the form of universal computing device.The component of electronic equipment 500 can wrap
It includes but is not limited to: at least one above-mentioned processing unit 510, at least one above-mentioned storage unit 520, the different system components of connection
The bus 530 of (including storage unit 520 and processing unit 510), display unit 540.
Wherein, storage unit is stored with program code, and program code can be executed with unit 510 processed, so that processing is single
Member 510 executes the step described in above-mentioned " illustrative methods " part of this specification according to the various illustrative embodiments of the disclosure
Suddenly.For example, processing unit 510 can execute step S110~S140 shown in FIG. 1, step shown in Fig. 2 can also be executed
S210~S240 etc..
Storage unit 520 may include the readable medium of volatile memory cell form, such as Random Access Storage Unit
(RAM) 521 and/or cache memory unit 522, it can further include read-only memory unit (ROM) 523.
Storage unit 520 can also include program/utility 524 with one group of (at least one) program module 525,
Such program module 525 includes but is not limited to: operating system, one or more application program, other program modules and
It may include the realization of network environment in program data, each of these examples or certain combination.
Bus 530 can be to indicate one of a few class bus structures or a variety of, including storage unit bus or storage
Cell controller, peripheral bus, graphics acceleration port, processing unit use any bus structures in a variety of bus structures
Local bus.
Electronic equipment 5 can also be with one or more external equipments 700 (such as keyboard, sensing equipment, bluetooth equipment etc.)
Communication, can also be enabled a user to one or more equipment interact with the electronic equipment 500 communicate, and/or with make this
Any equipment (such as the router, modem that electronic equipment 500 can be communicated with one or more of the other calculating equipment
Etc.) communication.This communication can be carried out by input/output (I/O) interface 550.Also, electronic equipment 500 can also lead to
Cross network adapter 560 and one or more network (such as local area network (LAN), wide area network (WAN) and/or public network, example
Such as internet) communication.As shown, network adapter 560 is communicated by bus 530 with other modules of electronic equipment 500.It answers
When understanding, although not shown in the drawings, other hardware and/or software module can be used in conjunction with electronic equipment 500, including but unlimited
In: microcode, device driver, redundant processing unit, external disk drive array, RAID system, tape drive and number
According to backup storage system etc..
Through the above description of the embodiments, those skilled in the art is it can be readily appreciated that example described herein is implemented
Mode can also be realized by software realization in such a way that software is in conjunction with necessary hardware.Therefore, according to the disclosure
The technical solution of embodiment can be embodied in the form of software products, which can store non-volatile at one
Property storage medium (can be CD-ROM, USB flash disk, mobile hard disk etc.) in or network on, including some instructions are so that a calculating
Equipment (can be personal computer, server, terminal installation or network equipment etc.) is executed according to the exemplary implementation of the disclosure
The method of example.
The exemplary embodiment of the disclosure additionally provides a kind of computer readable storage medium, and being stored thereon with can be realized
The program product of this specification above method.In some possible embodiments, various aspects of the disclosure can also be realized
For a kind of form of program product comprising program code, when program product is run on the terminal device, program code is used for
Execute terminal device described in above-mentioned " illustrative methods " part of this specification according to the various exemplary embodiment party of the disclosure
The step of formula.
It is produced refering to what is shown in Fig. 6, describing the program according to the exemplary embodiment of the disclosure for realizing the above method
Product 600, can be using portable compact disc read only memory (CD-ROM) and including program code, and can set in terminal
It is standby, such as run on PC.However, the program product of the disclosure is without being limited thereto, in this document, readable storage medium storing program for executing can
With to be any include or the tangible medium of storage program, the program can be commanded execution system, device or device use or
Person is in connection.
Program product can be using any combination of one or more readable mediums.Readable medium can be readable signal Jie
Matter or readable storage medium storing program for executing.Readable storage medium storing program for executing for example can be but be not limited to electricity, magnetic, optical, electromagnetic, infrared ray or partly lead
System, device or the device of body, or any above combination.More specific example (the non exhaustive column of readable storage medium storing program for executing
Table) it include: the electrical connection with one or more conducting wires, portable disc, hard disk, random access memory (RAM), read-only storage
Device (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read only memory (CD-
ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.
Computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
In carry readable program code.The data-signal of this propagation can take various forms, including but not limited to electromagnetic signal,
Optical signal or above-mentioned any appropriate combination.Readable signal medium can also be any readable Jie other than readable storage medium storing program for executing
Matter, the readable medium can send, propagate or transmit for by instruction execution system, device or device use or and its
The program of combined use.
The program code for including on readable medium can transmit with any suitable medium, including but not limited to wirelessly, have
Line, optical cable, RF etc. or above-mentioned any appropriate combination.
Can with any combination of one or more programming languages come write for execute the disclosure operation program
Code, programming language include object oriented program language-Java, C++ etc., further include conventional process
Formula programming language-such as " C " language or similar programming language.Program code can be calculated fully in user
It executes in equipment, partly execute on a user device, executing, as an independent software package partially in user calculating equipment
Upper part executes on a remote computing or executes in remote computing device or server completely.It is being related to remotely counting
In the situation for calculating equipment, remote computing device can pass through the network of any kind, including local area network (LAN) or wide area network
(WAN), it is connected to user calculating equipment, or, it may be connected to external computing device (such as utilize ISP
To be connected by internet).
In addition, above-mentioned attached drawing is only the schematic theory of the processing according to included by the method for disclosure exemplary embodiment
It is bright, rather than limit purpose.It can be readily appreciated that the time that above-mentioned processing shown in the drawings did not indicated or limited these processing is suitable
Sequence.In addition, be also easy to understand, these processing, which can be, for example either synchronously or asynchronously to be executed in multiple modules.
It should be noted that although being referred to several modules or list for acting the equipment executed in the above detailed description
Member, but this division is not enforceable.In fact, according to an exemplary embodiment of the present disclosure, above-described two or
More multimode or the feature and function of unit can embody in a module or unit.Conversely, above-described one
A module or the feature and function of unit can be to be embodied by multiple modules or unit with further division.
Those skilled in the art after considering the specification and implementing the invention disclosed here, will readily occur to its of the disclosure
His embodiment.This application is intended to cover any variations, uses, or adaptations of the disclosure, these modifications, purposes or
Adaptive change follow the general principles of this disclosure and including the undocumented common knowledge in the art of the disclosure or
Conventional techniques.The description and examples are only to be considered as illustrative, and the true scope and spirit of the disclosure are by claim
It points out.
It should be understood that the present disclosure is not limited to the precise structures that have been described above and shown in the drawings, and
And various modifications and changes may be made without departing from the scope thereof.The scope of the present disclosure is only limited by the attached claims.
Claims (10)
1. a kind of service access method characterized by comprising
Obtain the characteristic of object to be serviced;
Credit evaluation is carried out to the object to be serviced based on the characteristic, obtains the credit evaluation of the object to be serviced
As a result;
Response assessment is carried out to the object to be serviced based on the characteristic, obtains the response assessment of the object to be serviced
As a result;
Object to be serviced described in access is determined whether according to the credit evaluation result of the object to be serviced and response assessment result.
2. the method according to claim 1, wherein described be based on the characteristic to the object to be serviced
Credit evaluation is carried out, the credit evaluation result of the object to be serviced is obtained, comprising:
The characteristic is handled by risk analysis model and debt analysis model respectively, it is described to be serviced right to obtain
The risk analysis result of elephant and analysis result of being in debt;
It is commented according to the credit that the risk analysis result of the object to be serviced obtains the object to be serviced with analysis result of being in debt
Estimate result.
3. according to the method described in claim 2, it is characterized in that, the method also includes:
Obtain the sample characteristics data of multiple sample objects and each sample object;
Obtain the first tag along sort and the second tag along sort of the sample object;
Using the sample characteristics data and first tag along sort, the first machine learning model of training obtains the wind
Dangerous analysis model;
Using the sample characteristics data and second tag along sort, the second machine learning model of training is obtained described negative
Debt analysis model;
Wherein, first tag along sort includes high risk object or non-high risk object, and second tag along sort includes height
Debt object or non-high liabilities object.
4. according to the method described in claim 3, it is characterized in that, first machine learning model and the second machine learning mould
Type is that gradient promotes decision-tree model, and the risk analysis result includes high risk probability, and the debt analysis result includes height
Debt probability;
It is described that the characteristic is handled by risk analysis model and debt analysis model respectively, it obtains described wait take
The risk analysis result of object of being engaged in and analysis result of being in debt include:
The characteristic is handled by N decision in the face of risk tree of the risk analysis model, obtains N number of 1 or 0 height
Classification of risks value, wherein 1 indicates to predict that the object to be serviced is high risk object, 0 indicates to predict that the object to be serviced is
Non- high risk object;
Pass through formulaThe high risk probability is calculated, wherein PrFor the high risk probability, WriIt is i-th
The weight coefficient of decision in the face of risk tree, TriFor the high risk classification value of i-th decision in the face of risk tree output;
The characteristic is handled by M debt decision tree of the debt analysis model, obtains M 1 or 0 height
Classification of liabilities value, wherein 1 indicates to predict that the object to be serviced is high liabilities object, 0 indicates to predict that the object to be serviced is
Non- high liabilities object;
Pass through formulaThe high liabilities probability is calculated, wherein PdFor the high liabilities probability, WdjIt is
The weight coefficient of j debt decision tree, TdjFor the high liabilities classification value of jth debt decision tree output.
5. described the method according to claim 1, wherein the credit evaluation result includes credit appraisal value
Responding assessment result includes response rate;
The credit evaluation result according to the object to be serviced determines whether described to be serviced with response assessment result
Object access includes:
The violation correction probability of the object to be serviced is calculated according to the credit appraisal value and the response rate;
If the violation correction probability of the object to be serviced reaches a probability threshold value, refuse object to be serviced described in access;
If the violation correction probability of the object to be serviced is not up to the probability threshold value, object to be serviced described in access.
6. according to the method described in claim 5, it is characterized in that, described according to the credit appraisal value and the response rate meter
The violation correction probability for calculating the object to be serviced includes:
It is calculated by the following formula the violation correction probability of the object to be serviced:
Pq=Pm a·Pn b, wherein PqFor the violation correction probability, PmFor the credit appraisal value, PnFor the response rate, a with
B is the constant parameter greater than 0.
7. the method according to claim 1, wherein described be based on the characteristic to the object to be serviced
Response assessment is carried out, the response assessment result for obtaining the object to be serviced includes:
By object to be serviced described in response prediction model prediction, the response assessment result of the object to be serviced is obtained;
Wherein, the response assessment result includes high response object and non-high response object.
8. a kind of service access device characterized by comprising
Data acquisition module, for obtaining the characteristic of object to be serviced;
First evaluation module, for carrying out credit evaluation to the object to be serviced based on the characteristic, obtain it is described to
The credit evaluation result of service object;
Second evaluation module, for carrying out response assessment to the object to be serviced based on the characteristic, obtain it is described to
The response assessment result of service object;
Access judgment module, for determining whether standard according to the credit evaluation result and response assessment result of the object to be serviced
Enter the object to be serviced.
9. a kind of electronic equipment characterized by comprising
Processor;And
Memory, for storing the executable instruction of the processor;
Wherein, the processor is configured to require 1-7 described in any item via executing the executable instruction and carry out perform claim
Method.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
Claim 1-7 described in any item methods are realized when being executed by processor.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910044398.1A CN109816234A (en) | 2019-01-17 | 2019-01-17 | Service access method, service access device, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910044398.1A CN109816234A (en) | 2019-01-17 | 2019-01-17 | Service access method, service access device, electronic equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109816234A true CN109816234A (en) | 2019-05-28 |
Family
ID=66603404
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910044398.1A Pending CN109816234A (en) | 2019-01-17 | 2019-01-17 | Service access method, service access device, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109816234A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110399818A (en) * | 2019-07-15 | 2019-11-01 | 联动优势科技有限公司 | A kind of method and apparatus of risk profile |
CN110400072A (en) * | 2019-07-24 | 2019-11-01 | 阿里巴巴集团控股有限公司 | A kind of authorization method of credit services, device and equipment |
-
2019
- 2019-01-17 CN CN201910044398.1A patent/CN109816234A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110399818A (en) * | 2019-07-15 | 2019-11-01 | 联动优势科技有限公司 | A kind of method and apparatus of risk profile |
CN110400072A (en) * | 2019-07-24 | 2019-11-01 | 阿里巴巴集团控股有限公司 | A kind of authorization method of credit services, device and equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20210173711A1 (en) | Integrated value chain risk-based profiling and optimization | |
TWI712981B (en) | Risk identification model training method, device and server | |
US11568480B2 (en) | Artificial intelligence derived anonymous marketplace | |
Sherrick et al. | Crop insurance valuation under alternative yield distributions | |
CN109685645A (en) | User credit methods of risk assessment and device, storage medium | |
CN113822488B (en) | Risk prediction method and device for financing lease, computer equipment and storage medium | |
CN111695938B (en) | Product pushing method and system | |
CN111179051A (en) | Financial target customer determination method and device and electronic equipment | |
KR20190134934A (en) | Predictive device for customer churn using Deep Learning and Boosted Decision Trees and method of predicting customer churn using it | |
Li et al. | Stock trading strategies based on deep reinforcement learning | |
CN114819967A (en) | Data processing method and device, electronic equipment and computer readable storage medium | |
CN109816234A (en) | Service access method, service access device, electronic equipment and storage medium | |
KR20210157767A (en) | Systems and methods for financial management | |
CN115760332A (en) | Risk prediction method, system, medium and device based on enterprise data analysis | |
CN117235608B (en) | Risk detection method, risk detection device, electronic equipment and storage medium | |
CN113962567A (en) | Information recommendation method and device, electronic equipment and storage medium | |
Deng et al. | Stock price change rate prediction by utilizing social network activities | |
WO2022015488A1 (en) | Event-driven model refinement for insurance and risk management | |
CN116664306A (en) | Intelligent recommendation method and device for wind control rules, electronic equipment and medium | |
KR102609681B1 (en) | Method for determining product planning reflecting user feedback and Apparatus thereof | |
WO2020000782A1 (en) | Financial product recommendation method and apparatus, computer device, and readable storage medium | |
KR20060033976A (en) | System and method for forecasting and informing real estate mortgage risk | |
CN109523296A (en) | User behavior probability analysis method and device, electronic equipment, storage medium | |
CN115907898A (en) | Method for recommending financial products to reinsurance client and related equipment | |
CN112712270B (en) | Information processing method, device, equipment and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |