CN108399509A - Determine the method and device of the risk probability of service request event - Google Patents
Determine the method and device of the risk probability of service request event Download PDFInfo
- Publication number
- CN108399509A CN108399509A CN201810327337.1A CN201810327337A CN108399509A CN 108399509 A CN108399509 A CN 108399509A CN 201810327337 A CN201810327337 A CN 201810327337A CN 108399509 A CN108399509 A CN 108399509A
- Authority
- CN
- China
- Prior art keywords
- user
- service request
- illustrative plates
- request event
- specific crowd
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION 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
- G06Q30/00—Commerce
- G06Q30/018—Certifying business or products
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/02—Banking, e.g. interest calculation or account maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION 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
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Abstract
This specification embodiment provides a kind of method and apparatus of the risk probability of determining service request event, and method includes obtaining the affair character of service request event, obtaining the individual subscriber feature of user involved by the service request event.Also, based on the human relation collection of illustrative plates for including the above-mentioned specific crowd for being related to user, determine the relationship characteristic for being related to user.Relationship characteristic as a result, based on above-mentioned affair character, individual subscriber feature and user, determines the risk probability of the service request event.In this way, can comprehensively be assessed the risk of service request event.
Description
Technical field
This specification one or more embodiment is related to field of computer technology, more particularly to determines business by computer
The method and apparatus of the risk probability of request event.
Background technology
With the development of computer and Internet technology, more and more business are realized by computing platform, such as quotient
Product transaction, debt payment, finance debt-credit, settlement of insurance claim etc..However, in perhaps multiple services execution and processing, if not right
The background of service request people and requested business are audited, and are just likely to bring greater risk, such as some illegal point
Son implements financial swindling, debt-credit arbitrage, insurance insurance fraud etc. possibly also with e-platform.
In routine techniques, in order to prevent with the above-mentioned risk of reduction, risk audit is carried out often through artificial.In some platforms
In, some simple rules can be also set, and indirect labor judges.However, such mode efficiency is very low, it is difficult to meet industry
The fast-developing needs of business；Also, the accuracy of identification high risk user and high risk event depends on the business of manual examination and verification
The experience of member, the difference of the experience of different business person also bring along operational risk so that audit accuracy is difficult to obtain
Ensure, usually omits.
Accordingly, it would be desirable to have improved plan, by effectively and accurately determining the risk probability of service request event, drop
Low business executes risk.
Invention content
This specification one or more embodiment describes a kind of method and apparatus, for efficiently determining service request thing
The risk probability of part.
According in a first aspect, provide a kind of method of the risk probability of determining service request event, including：
Obtain the affair character of service request event；
Obtain the individual subscriber feature of at least one user involved by the service request event；
Human relation collection of illustrative plates based on specific crowd determines the relationship characteristic of at least one user, wherein the spy
It includes at least one user to determine crowd；
According to the affair character, the individual subscriber feature of at least one user and at least one user
Relationship characteristic, determine the risk probability of the service request event.
In one embodiment, above-mentioned affair character includes at least one of the following：The requested service amount of money, service log-on
Time, Time To Event, the time difference of service log-on time and Time To Event, venue location point.
In one embodiment, above-mentioned at least one user includes that the claimant of the service request event and business are asked
The beneficiary asked.
In one embodiment, above-mentioned individual subscriber feature include it is following in it is one or more, user essential attribute is special
Sign, user behavior characteristics, user location feature.
According to a kind of embodiment, the relationship characteristic vector for determining above-mentioned at least one user, specifically includes：Acquisition includes
The specific crowd of at least one user；Obtain the human relation collection of illustrative plates of the specific crowd；And it is based on the people
Group relation collection of illustrative plates determines the relationship characteristic of at least one user.
In one embodiment, obtaining above-mentioned specific crowd includes again, in the multiple user's subsets divided in advance, determines
User's subset belonging at least one user, using user's subset as above-mentioned specific crowd；Alternatively, by described at least
One user is added in the user's set being pre-selected, and the user is gathered and is used as the specific crowd.
In one embodiment, the human relation collection of illustrative plates for obtaining specific crowd further comprises：It obtains to be directed to and be pre-selected
User gather structure the first relation map；It obtains at least one user and the user's set being pre-selected
The incidence relation of user；The incidence relation is added to first relation map, the crowd as the specific crowd is closed
It is collection of illustrative plates.
According to a kind of embodiment, the human relation collection of illustrative plates of above-mentioned specific crowd is built based on one or more of relationship
It is vertical：Transaction relationship, device relationships, fund relationship, social networks.
In one embodiment, it determines that the relationship characteristic of user includes, is calculated using node-vector network structure feature extraction
Relation map is converted to the vectorial factor by method, and the relationship characteristic vector of user is determined based on the vectorial factor.
In one embodiment, the risk probability of service request event is determined using assessment models trained in advance, it is described
Assessment models promote decision Tree algorithms based on gradient and train.
According to second aspect, a kind of device of the risk probability of determining service request event is provided, including：
Affair character acquiring unit is configured to obtain the affair character of service request event；
Personal characteristics acquiring unit is configured to obtain the user of at least one user involved by the service request event
Personal characteristics；
Relationship characteristic acquiring unit is configured to the human relation collection of illustrative plates of specific crowd, determines at least one use
The relationship characteristic at family, wherein the specific crowd includes at least one user；
Risk determination unit, is configured to according to the affair character, the individual subscriber feature of at least one user, with
And the relationship characteristic of at least one user, determine the risk probability of the service request event.
According to the third aspect, a kind of computer readable storage medium is provided, computer program is stored thereon with, when described
When computer program executes in a computer, enable computer execute first aspect method.
According to fourth aspect, a kind of computing device, including memory and processor are provided, which is characterized in that described to deposit
It is stored with executable code in reservoir, when the processor executes the executable code, the method for realizing first aspect.
The method and apparatus provided by this specification embodiment, it is involved based on the affair character of service request event
The individual subscriber feature of user and the relationship characteristic of involved user, the comprehensive risk probability for determining service request event, from
And it is more efficient and accurate so that risk determines.
Description of the drawings
In order to illustrate the technical solution of the embodiments of the present invention more clearly, required use in being described below to embodiment
Attached drawing be briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for this
For the those of ordinary skill of field, without creative efforts, others are can also be obtained according to these attached drawings
Attached drawing.
Fig. 1 shows the implement scene schematic diagram of one embodiment that this specification discloses；
Fig. 2 shows the method flow diagrams according to the risk probability of the determination service request event of one embodiment；
Fig. 3 shows the step flow of the relationship characteristic of the determination associated user according to one embodiment；
Fig. 4 shows the example of the human relation collection of illustrative plates according to one embodiment；
Fig. 5 shows the schematic block diagram of the risk determining device according to one embodiment.
Specific implementation mode
Below in conjunction with the accompanying drawings, the scheme provided this specification is described.
Fig. 1 is the implement scene schematic diagram of one embodiment that this specification discloses.In the implement scene, pass through calculating
Platform is audited to execute the risk of service request event.User can send out service request event to computing platform, such as apply
Loan, application settlement of insurance claim etc..After computing platform gets such service request, various information are obtained, with to this
The risk probability of event is fully assessed.This various information includes the event information and business of service request event
The individual subscriber feature of user involved by request event.In addition, the user involved by event is also put into particular person by computing platform
In group, to obtain relationship characteristic of the user in human relation collection of illustrative plates.On this basis, according to above-mentioned affair character, user
The relationship characteristic of people's feature and user, the comprehensive risk probability for comprehensively assessing service request event.Above-mentioned field is described below
The specific implementation procedure of scape.
Fig. 2 shows the method flow diagrams according to the risk probability of the determination service request event of one embodiment.This method
Executive agent can any there is calculating, the system of processing capacity, unit, platform or server, such as Fig. 1 institutes
The computing platform shown more specifically, being, for example, the various background servers for needing to carry out business risk analysis management and control, for example is propped up
Pay precious server, insurance business server, finance examination ＆ approval server etc..As shown in Fig. 2, this approach includes the following steps：Step
21, obtain the affair character of service request event；Step 22, the use of at least one user involved by service request event is obtained
Family personal characteristics；Step 23, the human relation collection of illustrative plates based on specific crowd determines the relationship characteristic of at least one user,
The wherein described specific crowd includes at least one user；Step 24, according to the affair character, at least one user
Individual subscriber feature and at least one user relationship characteristic, determine the risk probability of the service request event.
The executive mode of above each step is described below.
First, in step 21, the affair character of service request event to be assessed is obtained.It is appreciated that industry to be assessed
Business request event can be the event made requests on for the various business that there may be risk, for example, application loan, application
Credit services, application settlement of insurance claim etc..Correspondingly, may include with the relevant affair character of service request event it is following in one
Item is multinomial：When the type of service of request, the request amount of money, the time of origin of request, service log-on time, registion time are with request
Between time difference, venue location point etc..More specifically, in a specific example, above-mentioned service request event is that application is protected
The event nearly settled a claim, correspondingly, affair character may include：The insurance kind of request, the application Claims Resolution time, is thrown request amount for which loss settled
Protect time difference, the scene etc. of time, the time of policy purchase and Claims Resolution time.In another example, above-mentioned service request event is
Apply for the event of loan, correspondingly, affair character may include：Ask the amount of money, the application time, registion time, registion time with
Time difference, scene of application time etc..
In addition, in step 22, the individual subscriber feature of the associated user involved by service request event is also obtained.At one
In embodiment, the associated user involved by service request event is service request people.In another embodiment, service request thing
Associated user involved by part further includes other Stakeholders in addition to claimant.For example, the event of application loan transaction,
The associated user being related to, can also be including guarantor etc. in addition to including loan requests people.The event for applying for settlement of insurance claim, is related to
Associated user can also include beneficiary etc. in addition to including Claims Resolution claimant.Therefore, involved by service request event
Associated user can be multiple users.The user of these users is obtained in step 22 for involved each associated user
People's feature.
In one embodiment, individual subscriber feature includes user's essential attribute feature, such as：Gender, age, when registration
It is long, contact method etc. essential information.
In one embodiment, individual subscriber feature includes user behavior characteristics.More specifically, user behavior characteristics can be with
Include operating relevant behavioural information with the history service of user, for example, transaction count, average transaction amount, application Claims Resolution time
Number, granted number of settling a claim, average amount for which loss settled etc..
In one embodiment, individual subscriber feature further includes user location feature, such as every history service operation hair
Raw position, range of position change, etc..
In more embodiments, individual subscriber feature can also include more various user characteristics.It is appreciated that user
Personal characteristics is to only rely upon some features of some user individual, portrays attribute feature, operating feature of the user itself etc..
Also user is put into certain crowd other than obtaining the personal characteristics of user's individual according to the embodiment of this specification,
And then excavate out relationship characteristic of the user in human relation network, so as to be based on the relationship characteristic carry out more fully analysis and
Assessment.
Then, in step 23, for each associated user referred in step 22, the human relation figure based on specific crowd
Spectrum, determines the relationship characteristic of each user, wherein the specific crowd includes above-mentioned associated user.Fig. 3 is shown according to a reality
Apply the step flow of the relationship characteristic of the determination associated user of example, the i.e. sub-step of step 23.As shown in figure 3, each in order to determine
The relationship characteristic of a associated user obtains the specific crowd for including associated user in step 31.
In one embodiment, sufficiently large user's set is predefined so that user set is comprising to be assessed
Service request event associated user, then can by the user gather be used as specific crowd.For example, in service request event
It, can be using the set of all personnel that insure as above-mentioned specific crowd in the case of for application settlement of insurance claim.
In one embodiment, according to certain features of user, the set of full dose user is divided into multiple user's subsets.
In step 31, user's subset belonging to the associated user involved by service request event is judged, using user's subset as upper
State specific crowd.
In one embodiment, it is pre-selected and constitutes user's collection with the certain customers of certain similitude or relevance
It closes.For example, in the case where service request event is application settlement of insurance claim, all use for once applying for Claims Resolution can be pre-selected
Family constitutes user's set.Then in step 31, the associated user of current event is judged whether in above-mentioned user gathers, such as
Fruit does not exist, then is added in user set, the user after addition is gathered and is used as the specific crowd.
Above-mentioned specific crowd can also be obtained by other means, as long as so that the specific crowd includes phase to be analyzed
Close user.
Then, in step 32, the human relation collection of illustrative plates of above-mentioned specific crowd is obtained.
In one embodiment, which includes, for above-mentioned specific crowd, rebuilding human relation collection of illustrative plates.
In another embodiment, above-mentioned specific crowd is gathered selected from scheduled user, and system has been the use in advance
Family set constructs human relation collection of illustrative plates.For example, in foregoing example, specific crowd can be selected from full dose user, or
Based on some user's subset that full dose user divides, and system may establish human relation collection of illustrative plates for full dose user in advance, or
Person establishes human relation collection of illustrative plates for each user's subset.At this point, in the step 32, the people built in advance can be directly acquired
Group relation collection of illustrative plates, or from human relation collection of illustrative plates building in advance, for wider user, extract with it is above-mentioned specific
The relevant part of crowd, as the human relation collection of illustrative plates for the specific crowd.
In another embodiment, above-mentioned specific crowd is by being added to associated user in the user being pre-selected set
And it is formed.If system has been directed to the user being pre-selected, set constructs human relation collection of illustrative plates, and step 32 can be with
Including obtaining the relation map for gathering structure for the user being pre-selected first；It is advance with this to obtain above-mentioned associated user
The incidence relation of user in user's set of selection；Then, above-mentioned incidence relation is added in above-mentioned relation collection of illustrative plates, as
The human relation collection of illustrative plates of the specific crowd.
Either structure or scene are rebuild in advance, and the structure of human relation collection of illustrative plates can be based on a variety of relationships.
In one embodiment, human relation collection of illustrative plates is established based on the transaction relationship of crowd.For example, between two users
Reach commodity purchasing transaction, then establishes transaction association between the two users.It can be by obtaining and analyzing a large number of users
Transaction record and determine the transaction relationship between user, and then establish human relation collection of illustrative plates.
In one embodiment, human relation collection of illustrative plates is established based on the device relationships of crowd.For example, when two or more
When user account is logged in using same station terminal equipment, it may be determined that there are equipment between the two or multiple user accounts
Association.There are the associated two or more user accounts of equipment, it may be possible to which multiple accounts of same entity user registration also may be used
To be that there are the accounts corresponding to multiple users of tight association (such as household, colleague etc.).Device relationships can pass through acquisition
User's entity end message corresponding when logging in its account and determine.
In one embodiment, human relation collection of illustrative plates is established based on fund relationship.For example, existing when between two users
Transfer accounts, collect money etc. funds transfer operation when, then fund association is established between the two users.It can be by obtaining and analyzing use
Family carries out the record of fund operation using stored value card and determines the fund relationship between user, and then is established based on fund relationship
Human relation collection of illustrative plates.
In one embodiment, human relation collection of illustrative plates is established based on social networks.Nowadays people use more and more
Social networking application interacts, for example, the interactions such as two users can be chatted by social networking application, be given bonus, file transmission,
Social association can be so established between the two users.The a large amount of social interactions that can be captured based on social networking application determine people
Social networks between group, and then establish human relation collection of illustrative plates.
Although several examples are presented above, it will be appreciated that being also based on more kinds of crowd's incidence relations to build
Vertical human relation collection of illustrative plates.Also, human relation collection of illustrative plates can be established based on several crowd's incidence relation simultaneously.
In one embodiment, human relation collection of illustrative plates can be formed as the form of meshed network.Under the form, Ren Qunguan
Be collection of illustrative plates include multiple nodes, each node corresponds to a user, and there are can be connected to each other between the node of incidence relation.One
In a embodiment, the connection between node can have a variety of attributes, such as connection type, bonding strength etc., wherein connecting class
Type includes again, such as fund connection (connection based on fund relationship), social activity connection (connection etc. based on social interaction), connection
Intensity may include again, such as strong ties, Weak link etc..
Fig. 4 shows the example of the human relation collection of illustrative plates according to one embodiment.As shown in figure 4, in this example embodiment, Ren Qunguan
Be collection of illustrative plates include multiple nodes, each node corresponds to a user.Connection between node indicates there is association between user
Relationship.It is assumed that the human relation collection of illustrative plates of Fig. 4 is fund relationship based on crowd and social networks and establishes.Correspondingly, node it
Between connection can be fund connection or social connection.In the example in fig. 4, different connection classes is shown with different line styles
Type, i.e., the social connection between node shown in dotted line, the fund connection being shown in solid between node.Also, with connecting line
Thickness show connection intensity.For example, thick line shows that strong ties, filament show Weak link.More specifically, heavy line can show
Go out, stronger fund connection (such as fund interaction is more than an amount of money threshold value, such as 10,000 yuan), fine line is shown, weaker fund
Connection (such as fund interaction is no more than above-mentioned amount of money threshold value)；Thick dashed line is it can be shown that stronger social activity connects (for example, interaction
The frequency is more than a frequency threshold value, such as 10 times a day), fine dotted line is shown, weaker social connection is not (for example, the interaction frequency surpasses
Cross above-mentioned frequency threshold value).
It is appreciated that in more embodiments, human relation collection of illustrative plates can also be characterized as other forms, such as table, figure
The forms such as shape.
Fig. 3 is returned to, on the basis of obtaining the human relation collection of illustrative plates built for specific crowd, in step 33, is based on
The human relation collection of illustrative plates, determines the relationship characteristic of the associated user involved by current event.
As previously mentioned, in human relation collection of illustrative plates, there are the users of incidence relation to be connected to each other.Correspondingly, at one
In embodiment, for some user, the feature of connection related with the user, example can be extracted from human relation collection of illustrative plates
Such as the number of connection, the type of connection, the intensity of connection, other users being connected to, etc. make such connection features
For the relationship characteristic of the user.
In another embodiment, using machine learning householder method, human relation collection of illustrative plates is analyzed and is characterized.It is practical
On, human relation collection of illustrative plates can be understood as a kind of network, wherein containing the node (corresponding to user) of certain amount, Yi Jijie
Connection relation (incidence relation between user) between point.Compared to text and image, the network information is more difficult to be structured to
The data of standard, accordingly, it is difficult to be applied to machine learning.Recently, it is proposed that several network representation (network
Representation) learning algorithm, to characterize and analyze network structure.The target of these algorithms is with low-dimensional, dense, real
The vector of value indicates the node with semantic relation in network, to conducive to storage is calculated, not have to manual extraction feature again, and
Heterogeneous information can be projected in the same lower dimensional space, facilitate and carry out downstream calculating.
According to network representation learning algorithm, by internet startup disk a to geometric space, by the space coordinate of each node
It is regarded as the feature of the node, is learnt and is trained to be put into neural network.It, can accordingly for human relation collection of illustrative plates
The collection of illustrative plates to be mapped in geometric space, the space coordinate of each user node is calculated, as its relationship characteristic vector.For
The calculating of the space coordinate of network node, may be used many algorithms.
In one embodiment, each node in the network corresponding to human relation collection of illustrative plates is determined using DeepWalk algorithms
Vector indicate.According to DeepWalk algorithms, a large amount of random walk particle is discharged on network, these particles are when given
The interior sequence that will walk out node composition.If node is regarded as word, the sequence thus generated just constitutes sentence,
Then a kind of node " language " by Sequence composition can be obtained.Then, (Word2Vec) algorithm is converted using term vector, so that it may
Vector to calculate each node " word " indicates.
In one embodiment, using node-vector (node2vec) structure feature extraction algorithm, by human relation collection of illustrative plates
Be converted to the form of the vectorial factor.Node2vec nodes-vector structure feature extraction algorithm, improve in DeepWalk and swim at random
The strategy walked, in the search (Breadth- of search (the Depth-First Search, DFS) and breadth First of depth-first
First Search, BFS) between reach a balance, while in view of the information of part and macroscopic view, to which superior vector generates
Mode.In this way, the user node in human relation collection of illustrative plates can be converted into the form of vector expression, may thereby determine that current
Vector expression of the user in the human relation collection of illustrative plates involved by event, as its relationship characteristic vector.
In other embodiments, more kinds of modes can also be used, current event is obtained from human relation collection of illustrative plates and is related to
The relationship characteristic vector of user.According to the different building modes of human relation collection of illustrative plates, different representations, the relationship characteristic of acquisition
The dimension of vector, element would also vary from.However, it will be understood that relationship characteristic vector is by characterizing the corresponding section of user
Position of the point in human relation collection of illustrative plates, and the connection relation with other nodes, to which comprehensively characterization user is closed in crowd
Incidence relation in system's net with other users.
Based on the individual subscriber feature that the affair character obtained in step 21, step 22 obtain, and as described above in step
The rapid 23 customer relationship features obtained, in step 24, in summary various features, determine the risk probability of service request event.
In a specific embodiment, it is based on affair character, determines the first assessment score of service request event；Based on use
Family personal characteristics determines the second assessment score of service request event；Based on customer relationship feature, service request event is determined
Third assesses score；Summation finally is weighted to the first, second, third assessment score, determines the risk of service request event
Probability score.The mode for wherein determining the first, second, and third assessment score, can pass through trained in advance model algorithm and mould
Shape parameter carries out.
In another specific embodiment, individual subscriber feature and customer relationship feature are represented as vectorial form.
Step 24, the feature vector of the feature vector of individual subscriber feature and customer relationship feature is spliced first, obtains user
Comprehensive characteristics.Then, user's comprehensive characteristics can be based on, the first assessment score of service request event is determined, is asked based on business
The affair character for seeking event determines the second assessment score of the event, is finally based on the first and second assessment scores, determines business
The risk probability score of request event.The mode for wherein determining the first and second assessment scores, can pass through mould trained in advance
Type algorithm and model parameter carries out.
In another embodiment, one assessment models of training, the assessment models are directly based upon affair character, user in advance
People's feature and customer relationship feature, assess the risk probability of service request event.It is appreciated that the assessment models base
It is trained in proven training dataset.In practice, for the service request event of its known risk probability, such as people
Work audit is determined as the negative sample event of Claims Resolution insurance fraud or manual examination and verification are determined as the positive sample event normally settled a claim, and obtains
The affair character of event, the individual subscriber feature of user involved by event.It is put into crowd, obtains in addition, will also be related to user
Relationship characteristic of the user in human relation collection of illustrative plates, especially relationship characteristic vector.Training dataset is added in data above.Such as
This, may be used certain model algorithm and model parameter, based on training data concentrate affair character, individual subscriber feature and
Customer relationship feature determines the risk probability of event, obtains the risk probability of some event.Then, based on obtained risk probability
With the comparison (i.e. loss function) of the actual known risk probability of the event, model algorithm and model parameter are continued to optimize, to
Training obtains above-mentioned assessment models.
A variety of specific model algorithms may be used in above-mentioned assessment models.In one embodiment, it is determined using gradient promotion
Plan tree GBDT (Gradient Boosting Decision Tree) method trains to obtain above-mentioned assessment models.
As it is known by the man skilled in the art, it is a kind of side for the integrated study having supervision that gradient, which promotes decision tree GBDT methods,
Method.In integrated learning approach, training sample set is learnt respectively using multiple learners, final model is to above-mentioned
The synthesis of multiple learners.The most important two methods of integrated study are Bagging and Boosting, wherein according to Boosting
Algorithm, there are sequencings between learner, and have different weights, while being also that each sample distributes weight.Initially
The weight on ground, each sample is equal, after learning to training sample using some learner, increases error sample
Weight reduces the weight of correct sample, and subsequent learner is recycled to learn it.In this way, final prediction result is
The merging of multiple learner results.On this basis, the mode that gradient transmission may be used is based on prediction result Optimized model letter
Number, such method are known as gradient and promote Gradient Boost methods.
In the case where gradient promotes Gradient Boost frames, each base learner uses post-class processing algorithm, just constitutes
Gradient promotes decision tree GBDT models.Post-class processing algorithm is a kind of machine learning algorithm based on binary tree.In gradient
Promoted in decision tree GBDT algorithms, due to being integrated with multiple such post-class processings as learner so that model it is accurate
Property and coverage rate are more efficient.
More specifically, according to GBDT algorithms, various features, including affair character, individual subscriber feature and use can be directed to
Family relationship characteristic, the multiple learners using post-class processing of training, to form above-mentioned assessment models.
In other embodiments, above-mentioned assessment models can also use the training of other algorithms to realize, such as above-mentioned integrated
Bagging algorithms in study, and the learner etc. using other algorithms.
After assessment models training is completed, in step 24, assessment models can be directly used, determine that current business is asked
Seek the risk probability of event.
In this way, affair character, individual subscriber feature and the customer relationship feature of a comprehensive service request event, Ke Yiquan
Face the risk probability of the service request event is assessed, to which more efficiently and accurately control business executes risk.
According to the embodiment of another aspect, a kind of device of the risk probability of determining service request event is also provided.Fig. 5 shows
Go out the schematic block diagram of the risk determining device according to one embodiment.As shown in figure 5, the risk determining device 500 includes：Thing
Part feature acquiring unit 510 is configured to obtain the affair character of service request event；Personal characteristics acquiring unit 520, is configured to
Obtain the individual subscriber feature of at least one user involved by the service request event；Relationship characteristic acquiring unit 530, matches
It is set to the human relation collection of illustrative plates based on specific crowd, determines the relationship characteristic of at least one user, wherein the particular person
Group includes at least one user；Risk determination unit 540 is configured to according to the affair character, at least one use
The relationship characteristic of the individual subscriber feature at family and at least one user determine that the risk of the service request event is general
Rate.
In one embodiment, the affair character that above-mentioned affair character acquiring unit 510 obtains include it is following at least
One：The requested service amount of money, service log-on time, Time To Event, the time of service log-on time and Time To Event
Difference, venue location point.
According to one embodiment, at least one user involved by service request event includes that service request event is asked
It asks for help and the beneficiary of service request.
In one embodiment, during the individual subscriber feature acquired in above-mentioned personal characteristics acquiring unit 520 includes following
One or more, user's essential attribute feature, user behavior characteristics, user location feature.
According to a kind of embodiment, above-mentioned relation feature acquiring unit 530 includes：Crowd's acquisition module 531 is configured to obtain
Take the specific crowd for including at least one user；Collection of illustrative plates acquisition module 532 is configured to obtain the crowd of the specific crowd
Relation map；Feature acquisition module 533 is configured to the human relation collection of illustrative plates, determines the pass of at least one user
It is feature.
In one embodiment, crowd's acquisition module 531 is configured to, in the multiple user's subsets divided in advance,
User's subset belonging at least one user is determined, using user's subset as above-mentioned specific crowd.
In another embodiment, crowd's acquisition module 531 is configured to, and at least one user is added in advance
In user's set of selection, the user is gathered and is used as the specific crowd.
Further, in one embodiment, collection of illustrative plates acquisition module 532 is configured to：What acquisition was pre-selected for described in
User gathers the first relation map of structure；Obtain at least one user and the use in the user's set being pre-selected
The incidence relation at family；The incidence relation is added to first relation map, the human relation as the specific crowd
Collection of illustrative plates.
According to a kind of embodiment, the human relation collection of illustrative plates of specific crowd is established based on one or more of relationship：
Transaction relationship, device relationships, fund relationship, social networks.
In one embodiment, above-mentioned relation feature acquiring unit 530 is configured to, special using node-vector network structure
Extraction algorithm is levied, the relation map is converted into the vectorial factor, at least one user is determined based on the vectorial factor
Relationship characteristic vector.
In one embodiment, risk determination unit 540 is configured to, and the industry is determined using assessment models trained in advance
The risk probability of business request event, the assessment models promote decision Tree algorithms based on gradient and train.
By above-mentioned apparatus, affair character, individual subscriber feature and the customer relationship of a comprehensive service request event are special
Sign, comprehensively assesses the risk probability of the service request event, to which more efficiently and accurately control business executes wind
Danger
According to the embodiment of another aspect, a kind of computer readable storage medium is also provided, is stored thereon with computer journey
Sequence enables computer execute method described in conjunction with Figure 2 when the computer program executes in a computer.
According to the embodiment of another further aspect, a kind of computing device, including memory and processor, the memory are also provided
In be stored with executable code, when the processor executes the executable code, realize the method in conjunction with described in Fig. 2.
Those skilled in the art are it will be appreciated that in said one or multiple examples, work(described in the invention
It can be realized with hardware, software, firmware or their arbitrary combination.It when implemented in software, can be by these functions
Storage in computer-readable medium or as on computer-readable medium one or more instructions or code be transmitted.
Above-described specific implementation mode has carried out further the purpose of the present invention, technical solution and advantageous effect
It is described in detail, it should be understood that the foregoing is merely the specific implementation mode of the present invention, is not intended to limit the present invention
Protection domain, all any modification, equivalent substitution, improvement and etc. on the basis of technical scheme of the present invention, done should all
Including within protection scope of the present invention.
Claims (24)
1. a kind of method of the risk probability of determining service request event, including：
Obtain the affair character of service request event；
Obtain the individual subscriber feature of at least one user involved by the service request event；
Human relation collection of illustrative plates based on specific crowd determines the relationship characteristic of at least one user, wherein the particular person
Group includes at least one user；
According to the affair character, the pass of the individual subscriber feature of at least one user and at least one user
It is feature, determines the risk probability of the service request event.
2. according to the method described in claim 1, the wherein described affair character includes at least one of the following：Requested service gold
Volume, service log-on time, Time To Event, the time difference of service log-on time and Time To Event, venue location point.
3. according to the method described in claim 1, wherein described at least one user includes the request of the service request event
People and the beneficiary of service request.
4. according to the method described in claim 1, the wherein described individual subscriber feature include it is following in one or more, user
Essential attribute feature, user behavior characteristics, user location feature.
5. according to the method described in claim 1, the wherein human relation collection of illustrative plates based on specific crowd, determines described at least one
The relationship characteristic vector of user, including：
Obtain the specific crowd for including at least one user；
Obtain the human relation collection of illustrative plates of the specific crowd；
Based on the human relation collection of illustrative plates, the relationship characteristic of at least one user is determined.
6. according to the method described in claim 5, wherein obtain and include comprising the specific crowd of at least one user,
In the multiple user's subsets divided in advance, user's subset belonging at least one user is determined, by user's subset
As above-mentioned specific crowd.
7. according to the method described in claim 5, wherein obtain and include comprising the specific crowd of at least one user,
At least one user is added in the user's set being pre-selected, the user is gathered and is used as the specific crowd.
8. according to the method described in claim 7, the human relation collection of illustrative plates for wherein obtaining the specific crowd includes：
Obtain the first relation map for gathering structure for the user being pre-selected；
Obtain the incidence relation of at least one user and the user in the user's set being pre-selected；
The incidence relation is added to first relation map, the human relation collection of illustrative plates as the specific crowd.
9. according to the method described in claim 1, the human relation collection of illustrative plates of the wherein described specific crowd is based on following one kind or more
Kind of relationship and establish：Transaction relationship, device relationships, fund relationship, social networks.
10. according to the method described in claim 1, wherein determining that the relationship characteristic of at least one user includes, using section
Point-vector network structure feature extraction algorithm, the vectorial factor is converted to by the relation map, is determined based on the vectorial factor
The relationship characteristic vector of at least one user.
11. according to the method described in claim 1, wherein determining that the risk probability of the service request event includes, using pre-
First trained assessment models determine that the risk probability of the service request event, the assessment models are based on gradient and promote decision tree
Algorithm and train.
12. a kind of device of the risk probability of determining service request event, including：
Affair character acquiring unit is configured to obtain the affair character of service request event；
Personal characteristics acquiring unit is configured to obtain the individual subscriber of at least one user involved by the service request event
Feature；
Relationship characteristic acquiring unit is configured to the human relation collection of illustrative plates of specific crowd, determines at least one user's
Relationship characteristic, wherein the specific crowd includes at least one user；
Risk determination unit is configured to according to the affair character, the individual subscriber feature of at least one user, Yi Jisuo
The relationship characteristic for stating at least one user determines the risk probability of the service request event.
13. device according to claim 12, wherein the affair character includes at least one of the following：Requested service
The amount of money, service log-on time, Time To Event, the time difference of service log-on time and Time To Event, venue location
Point.
14. device according to claim 12, wherein at least one user includes that the service request event is asked
It asks for help and the beneficiary of service request.
15. device according to claim 12, wherein the individual subscriber feature include it is following in it is one or more, use
Family essential attribute feature, user behavior characteristics, user location feature.
16. device according to claim 12, wherein the relationship characteristic acquiring unit includes：
Crowd's acquisition module is configured to obtain the specific crowd for including at least one user；
Collection of illustrative plates acquisition module is configured to obtain the human relation collection of illustrative plates of the specific crowd；
Feature acquisition module is configured to the human relation collection of illustrative plates, determines the relationship characteristic of at least one user.
17. device according to claim 16, wherein crowd's acquisition module is configured to, in the multiple use divided in advance
In the subset of family, user's subset belonging at least one user is determined, using user's subset as above-mentioned specific crowd.
18. device according to claim 16, wherein crowd's acquisition module is configured to, by least one user
It is added in the user's set being pre-selected, the user is gathered and is used as the specific crowd.
19. device according to claim 18, wherein the collection of illustrative plates acquisition module is configured to：
Obtain the first relation map for gathering structure for the user being pre-selected；
Obtain the incidence relation of at least one user and the user in the user's set being pre-selected；
The incidence relation is added to first relation map, the human relation collection of illustrative plates as the specific crowd.
20. device according to claim 12, wherein the human relation collection of illustrative plates of the specific crowd be based on it is following a kind of or
A variety of relationships and establish：Transaction relationship, device relationships, fund relationship, social networks.
21. device according to claim 12, wherein the relationship characteristic acquiring unit is configured to, using node-vector
The relation map is converted to the vectorial factor by network structure feature extraction algorithm, based on the vectorial factor determine described in extremely
The relationship characteristic vector of a few user.
22. device according to claim 12, wherein the risk determination unit is configured to, using assessment trained in advance
Model determines that the risk probability of the service request event, the assessment models promote decision Tree algorithms based on gradient and train.
23. a kind of computer readable storage medium, is stored thereon with computer program, when the computer program in a computer
When execution, computer perform claim is enabled to require the method for any one of 1-11.
24. a kind of computing device, including memory and processor, which is characterized in that be stored with executable generation in the memory
Code when the processor executes the executable code, realizes the method described in any one of claim 1-11.
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810327337.1A CN108399509A (en) | 2018-04-12 | 2018-04-12 | Determine the method and device of the risk probability of service request event |
| PCT/CN2019/073869 WO2019196546A1 (en) | 2018-04-12 | 2019-01-30 | Method and apparatus for determining risk probability of service request event |
| TW108104899A TW201944305A (en) | 2018-04-12 | 2019-02-14 | Method and apparatus for determining risk probability of service request event |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810327337.1A CN108399509A (en) | 2018-04-12 | 2018-04-12 | Determine the method and device of the risk probability of service request event |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108399509A true CN108399509A (en) | 2018-08-14 |
Family
ID=63100004
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810327337.1A Pending CN108399509A (en) | 2018-04-12 | 2018-04-12 | Determine the method and device of the risk probability of service request event |
Country Status (3)
| Country | Link |
|---|---|
| CN (1) | CN108399509A (en) |
| TW (1) | TW201944305A (en) |
| WO (1) | WO2019196546A1 (en) |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109377347A (en) * | 2018-09-27 | 2019-02-22 | 深圳先进技术研究院 | Network credit method for early warning, system and electronic equipment based on feature selecting |
| CN109559192A (en) * | 2018-10-25 | 2019-04-02 | 深圳壹账通智能科技有限公司 | Risk checking method, device, equipment and storage medium based on association map |
| CN109598513A (en) * | 2018-11-22 | 2019-04-09 | 阿里巴巴集团控股有限公司 | A kind of Risk Identification Method and device |
| CN109636083A (en) * | 2018-10-16 | 2019-04-16 | 深圳壹账通智能科技有限公司 | Blacklist analysis method, device, equipment and computer readable storage medium |
| CN109636564A (en) * | 2018-10-16 | 2019-04-16 | 平安科技（深圳）有限公司 | Information verification mechanism, device, equipment and storage medium for air control |
| CN109636565A (en) * | 2018-10-16 | 2019-04-16 | 深圳壹账通智能科技有限公司 | Processing method, device, equipment and the computer readable storage medium of risk data |
| CN109657917A (en) * | 2018-11-19 | 2019-04-19 | 平安科技（深圳）有限公司 | Assess method for prewarning risk, device, computer equipment and the storage medium of object |
| CN109685647A (en) * | 2018-12-27 | 2019-04-26 | 阳光财产保险股份有限公司 | The training method of credit fraud detection method and its model, device and server |
| CN109801077A (en) * | 2019-01-21 | 2019-05-24 | 北京邮电大学 | A kind of arbitrage user detection method, device and equipment |
| CN109919782A (en) * | 2019-01-24 | 2019-06-21 | 平安科技（深圳）有限公司 | It is associated with case recognition methods, electronic device and computer readable storage medium |
| CN110008349A (en) * | 2019-02-01 | 2019-07-12 | 阿里巴巴集团控股有限公司 | The method and device for the event risk assessment that computer executes |
| CN110009511A (en) * | 2019-01-29 | 2019-07-12 | 阿里巴巴集团控股有限公司 | Arbitrage behavior recognition methods, arbitrage behavior identification model training method and system |
| CN110033151A (en) * | 2018-11-09 | 2019-07-19 | 阿里巴巴集团控股有限公司 | Relational risk evaluation method, device, electronic equipment and computer storage medium |
| CN110046784A (en) * | 2018-12-14 | 2019-07-23 | 阿里巴巴集团控股有限公司 | A kind of risk of user's access determines method and device |
| CN110084468A (en) * | 2019-03-14 | 2019-08-02 | 阿里巴巴集团控股有限公司 | A kind of Risk Identification Method and device |
| CN110097450A (en) * | 2019-03-26 | 2019-08-06 | 中国人民财产保险股份有限公司 | Vehicle borrows methods of risk assessment, device, equipment and storage medium |
| WO2019196546A1 (en) * | 2018-04-12 | 2019-10-17 | 阿里巴巴集团控股有限公司 | Method and apparatus for determining risk probability of service request event |
| CN110544100A (en) * | 2019-09-10 | 2019-12-06 | 北京三快在线科技有限公司 | Business identification method, device and medium based on machine learning |
| CN110599329A (en) * | 2019-09-09 | 2019-12-20 | 腾讯科技（深圳）有限公司 | Credit evaluation method, credit evaluation device and electronic equipment |
| CN111198967A (en) * | 2019-12-20 | 2020-05-26 | 北京淇瑀信息科技有限公司 | User grouping method and device based on relational graph and electronic equipment |
| CN111798092A (en) * | 2020-05-27 | 2020-10-20 | 深圳奇迹智慧网络有限公司 | Customs inspection monitoring method, customs inspection monitoring device, computer equipment and storage medium |
| WO2021174966A1 (en) * | 2020-03-05 | 2021-09-10 | 支付宝(杭州)信息技术有限公司 | Risk identification model training method and apparatus |
| CN114912717A (en) * | 2022-07-13 | 2022-08-16 | 成都秦川物联网科技股份有限公司 | Smart city guarantee housing application risk assessment method and system based on Internet of things |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW202147207A (en) * | 2020-06-08 | 2021-12-16 | 財團法人資訊工業策進會 | Risk detection system and risk detection method |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106469376B (en) * | 2015-08-20 | 2021-12-03 | 创新先进技术有限公司 | Risk control method and equipment |
| CN107679686B (en) * | 2017-08-28 | 2024-03-29 | 创新先进技术有限公司 | Service execution method and device |
| CN107741953B (en) * | 2017-09-14 | 2020-01-21 | 平安科技（深圳）有限公司 | Method and device for matching realistic relationship of social platform user and readable storage medium |
| CN107818513A (en) * | 2017-11-24 | 2018-03-20 | 泰康保险集团股份有限公司 | Methods of risk assessment and device, storage medium, electronic equipment |
| CN108399509A (en) * | 2018-04-12 | 2018-08-14 | 阿里巴巴集团控股有限公司 | Determine the method and device of the risk probability of service request event |
-
2018
- 2018-04-12 CN CN201810327337.1A patent/CN108399509A/en active Pending
-
2019
- 2019-01-30 WO PCT/CN2019/073869 patent/WO2019196546A1/en active Application Filing
- 2019-02-14 TW TW108104899A patent/TW201944305A/en unknown
Cited By (32)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019196546A1 (en) * | 2018-04-12 | 2019-10-17 | 阿里巴巴集团控股有限公司 | Method and apparatus for determining risk probability of service request event |
| CN109377347A (en) * | 2018-09-27 | 2019-02-22 | 深圳先进技术研究院 | Network credit method for early warning, system and electronic equipment based on feature selecting |
| CN109636083A (en) * | 2018-10-16 | 2019-04-16 | 深圳壹账通智能科技有限公司 | Blacklist analysis method, device, equipment and computer readable storage medium |
| CN109636564A (en) * | 2018-10-16 | 2019-04-16 | 平安科技（深圳）有限公司 | Information verification mechanism, device, equipment and storage medium for air control |
| CN109636565A (en) * | 2018-10-16 | 2019-04-16 | 深圳壹账通智能科技有限公司 | Processing method, device, equipment and the computer readable storage medium of risk data |
| CN109559192A (en) * | 2018-10-25 | 2019-04-02 | 深圳壹账通智能科技有限公司 | Risk checking method, device, equipment and storage medium based on association map |
| WO2020082559A1 (en) * | 2018-10-25 | 2020-04-30 | 深圳壹账通智能科技有限公司 | Risk detection method, apparatus and device based on association graph, and storage medium |
| CN110033151B (en) * | 2018-11-09 | 2024-01-19 | 创新先进技术有限公司 | Relation risk evaluation method and device, electronic equipment and computer storage medium |
| CN110033151A (en) * | 2018-11-09 | 2019-07-19 | 阿里巴巴集团控股有限公司 | Relational risk evaluation method, device, electronic equipment and computer storage medium |
| CN109657917A (en) * | 2018-11-19 | 2019-04-19 | 平安科技（深圳）有限公司 | Assess method for prewarning risk, device, computer equipment and the storage medium of object |
| CN109657917B (en) * | 2018-11-19 | 2022-04-29 | 平安科技（深圳）有限公司 | Risk early warning method and device for evaluation object, computer equipment and storage medium |
| CN109598513A (en) * | 2018-11-22 | 2019-04-09 | 阿里巴巴集团控股有限公司 | A kind of Risk Identification Method and device |
| CN110046784A (en) * | 2018-12-14 | 2019-07-23 | 阿里巴巴集团控股有限公司 | A kind of risk of user's access determines method and device |
| CN109685647A (en) * | 2018-12-27 | 2019-04-26 | 阳光财产保险股份有限公司 | The training method of credit fraud detection method and its model, device and server |
| CN109685647B (en) * | 2018-12-27 | 2021-08-10 | 阳光财产保险股份有限公司 | Credit fraud detection method and training method and device of model thereof, and server |
| CN109801077A (en) * | 2019-01-21 | 2019-05-24 | 北京邮电大学 | A kind of arbitrage user detection method, device and equipment |
| CN109919782A (en) * | 2019-01-24 | 2019-06-21 | 平安科技（深圳）有限公司 | It is associated with case recognition methods, electronic device and computer readable storage medium |
| CN110009511A (en) * | 2019-01-29 | 2019-07-12 | 阿里巴巴集团控股有限公司 | Arbitrage behavior recognition methods, arbitrage behavior identification model training method and system |
| WO2020156000A1 (en) * | 2019-02-01 | 2020-08-06 | 阿里巴巴集团控股有限公司 | Computer implemented event risk assessment method and device |
| CN110008349A (en) * | 2019-02-01 | 2019-07-12 | 阿里巴巴集团控股有限公司 | The method and device for the event risk assessment that computer executes |
| CN110084468A (en) * | 2019-03-14 | 2019-08-02 | 阿里巴巴集团控股有限公司 | A kind of Risk Identification Method and device |
| CN110097450A (en) * | 2019-03-26 | 2019-08-06 | 中国人民财产保险股份有限公司 | Vehicle borrows methods of risk assessment, device, equipment and storage medium |
| CN110599329A (en) * | 2019-09-09 | 2019-12-20 | 腾讯科技（深圳）有限公司 | Credit evaluation method, credit evaluation device and electronic equipment |
| CN110544100A (en) * | 2019-09-10 | 2019-12-06 | 北京三快在线科技有限公司 | Business identification method, device and medium based on machine learning |
| CN111198967A (en) * | 2019-12-20 | 2020-05-26 | 北京淇瑀信息科技有限公司 | User grouping method and device based on relational graph and electronic equipment |
| CN111198967B (en) * | 2019-12-20 | 2024-03-08 | 北京淇瑀信息科技有限公司 | User grouping method and device based on relationship graph and electronic equipment |
| WO2021174966A1 (en) * | 2020-03-05 | 2021-09-10 | 支付宝(杭州)信息技术有限公司 | Risk identification model training method and apparatus |
| CN111798092A (en) * | 2020-05-27 | 2020-10-20 | 深圳奇迹智慧网络有限公司 | Customs inspection monitoring method, customs inspection monitoring device, computer equipment and storage medium |
| CN111798092B (en) * | 2020-05-27 | 2024-03-12 | 深圳奇迹智慧网络有限公司 | Customs inspection monitoring method, customs inspection monitoring device, computer equipment and storage medium |
| CN114912717B (en) * | 2022-07-13 | 2022-10-25 | 成都秦川物联网科技股份有限公司 | Smart city guarantee housing application risk assessment method and system based on Internet of things |
| US11715156B1 (en) | 2022-07-13 | 2023-08-01 | Chengdu Qinchuan Iot Technology Co., Ltd. | Risk assessment methods and systems for affordable housing application in a smart city based on internet of things |
| CN114912717A (en) * | 2022-07-13 | 2022-08-16 | 成都秦川物联网科技股份有限公司 | Smart city guarantee housing application risk assessment method and system based on Internet of things |
Also Published As
| Publication number | Publication date |
|---|---|
| TW201944305A (en) | 2019-11-16 |
| WO2019196546A1 (en) | 2019-10-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108399509A (en) | Determine the method and device of the risk probability of service request event | |
| CN109564669A (en) | Based on trust score and geographic range searching entities | |
| CN108648074A (en) | Loan valuation method, apparatus based on support vector machines and equipment | |
| CN109410036A (en) | A kind of fraud detection model training method and device and fraud detection method and device | |
| CN108596434A (en) | Fraud detection and methods of risk assessment, system, equipment and storage medium | |
| US20040177030A1 (en) | Psychometric Creditworthiness Scoring for Business Loans | |
| CN110349000A (en) | Method, apparatus and electronic equipment are determined based on the volume strategy that mentions of tenant group | |
| CN110796542A (en) | Financial risk control method, financial risk control device and electronic equipment | |
| KR101996797B1 (en) | Method for providing transaction security analysis service for risk management in real estate transaction based on disclosure data and statistics | |
| CN108805391A (en) | Determine the method and device of high risk user | |
| CN111192131A (en) | Financial risk prediction method and device and electronic equipment | |
| CN112150298B (en) | Data processing method, system, device and readable medium | |
| CN111127178A (en) | Data processing method and device, storage medium and electronic equipment | |
| US10929859B2 (en) | Systems and methods for determining economic impact of an event within a geographic area | |
| CN110659961A (en) | Method and device for identifying off-line commercial tenant | |
| US20240013919A1 (en) | Supervised machine learning-based modeling of sensitivities to potential disruptions | |
| CN110415103A (en) | The method, apparatus and electronic equipment that tenant group mentions volume are carried out based on variable disturbance degree index | |
| CN107220867A (en) | object control method and device | |
| CN109800363A (en) | Construct method, apparatus, equipment and the storage medium of standing relational network | |
| CN109377339A (en) | A kind of the auxiliary discriminating method and system of suspicious transaction case | |
| CN111192133A (en) | Method and device for generating risk model after user loan and electronic equipment | |
| CN110349007A (en) | The method, apparatus and electronic equipment that tenant group mentions volume are carried out based on variable discrimination index | |
| CN110349013A (en) | Risk control method and device | |
| CN110119980A (en) | A kind of anti-fraud method, apparatus, system and recording medium for credit | |
| CN113344700A (en) | Wind control model construction method and device based on multi-objective optimization and electronic equipment |
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 | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 1258566 Country of ref document: HK |
|
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180814 |
|
| RJ01 | Rejection of invention patent application after publication |