CN110046910A - The method and apparatus for obtaining customer group relevant to particular customer - Google Patents
The method and apparatus for obtaining customer group relevant to particular customer Download PDFInfo
- Publication number
- CN110046910A CN110046910A CN201811522367.4A CN201811522367A CN110046910A CN 110046910 A CN110046910 A CN 110046910A CN 201811522367 A CN201811522367 A CN 201811522367A CN 110046910 A CN110046910 A CN 110046910A
- Authority
- CN
- China
- Prior art keywords
- client
- customer
- data
- relevant
- customer group
- 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.)
- Granted
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
- G06Q30/01—Customer relationship services
-
- 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
- G06Q30/0185—Product, service or business identity fraud
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
A kind of method for obtaining customer group relevant to particular customer is provided, including obtaining the particular customer in multiple clients；Client-related data relevant to each client in multiple clients is obtained, the client-related data includes at least the client characteristics data of the customer relationship data and the client that indicate the relationship between each client and other clients in the multiple client；Obtain predefined extension rule data；And one or more client relevant to the particular customer in the multiple client is determined based on the client-related data and the extension rule data, to obtain customer group relevant to the particular customer.Thereby, it is possible to further obtain relative customer group on the basis of the individual suspicious client of identification, to improve trial efficiency while expanding and trying client's overlay capacity.
Description
Technical field
The present invention relates to Internet technical fields, more particularly to identify in multiple clients and be mutually related with particular customer
One or more clients.
Background technique
It with the continuous development of Internet technology, is applied in every field, has derived as internet finance
New technical field.The electronic payment platform of such as Alipay relies on Internet technology to can be realized payment funding, transfer accounts,
Greatly facilitate people's lives.
However, provided for people's lives facilitate while, these electronic payment platforms equally exist hidden danger.For example,
Certain clients may want to realize certain transaction with illegal objective using electronic payment platform, therefore, how judge visitor
The legitimacy for the transaction that family is carried out by electronic payment platform is important subject under discussion.
Currently, screening, the suspicious client of identification individual, then artificially to this are carried out by all clients to payment platform
A little suspicious clients carry out trying the client to identify illegal transaction one by one.
Summary of the invention
Relative client can further be obtained on the basis of the individual suspicious client of identification by being intended to provide one kind
Group, to expand client's overlay capacity while improve the means of trial efficiency.
According to one embodiment, a kind of method for obtaining customer group relevant to particular customer is provided, including obtains table
Show the data of the particular customer in multiple clients；Obtain client's dependency number relevant to each client in multiple clients
According to the client-related data, which includes at least, indicates the relationship between each client and other clients in the multiple client
The client characteristics data of customer relationship data and the client；Obtain predefined extension rule data；And it is based on the visitor
Family related data and the extension rule data determine one or more relevant to the particular customer in the multiple client
Client, to obtain customer group relevant to the particular customer.
Currently to the suspicious client's identification of the individual for the client's progress for using electronic payment platform to trade and trial style
Being associated with for the identified suspicious client of individual and other clients is not accounted for, only individual suspicious client is tried, this makes
The limited sample size that must be tried.And in fact, be not between the client to trade it is isolated, often have between them
Certain correlation, especially in the transaction for carrying out certain illegal objectives, when such as money laundering, there are very strong funds between each client
And/or non-fund relationship.Inventor recognizes the strong correlation between this client, by each implementation according to the present invention
Example, according to the acquisition of both client-related data and predefined extension rule and its on the basis of the individual suspicious client of acquisition
Other clients with strong correlation, using the suspicious client of acquired individual and with its client with strong correlation as one
Entirety is pushed to trial person, so as to be tried to the customer group in this way with strong correlation in hearing process, this
Sample can not only improve trial efficiency, be also easy to obtain complete chain of evidence on the docket, furthermore, by this kind of groups trial side
Formula, additionally it is possible to expand the coverage area of trial client, increase the defence radius of illegal transaction trial.
According to further embodiments, it is determined based on the client-related data and the extension rule data the multiple
Relevant to each client in one or more of clients another or multiple clients in client；And by it is described in addition
One or more clients are included in customer group relevant to the particular customer.
Thereby, it is possible to the bases of one or more clients in the customer group relevant to particular customer obtained for the first time
On, the client that there is strong correlation with one or more client is further obtained, which is also added
In customer group relevant to the particular customer, so that further expansion will increase illegal hand over by the coverage area of trial client
The defence radius easily tried.If desired, the such extension of multilayer can be carried out.
According to further embodiments, using the Rating Model based on machine learning based on acquisition and the particular customer
The client-related data of each client in relevant customer group scores to each client in the customer group；And
And it is based on the scoring, the client in the customer group is ranked up.
Score using the Rating Model based on machine learning as a result, and then sort, the accurate of scoring can be increased
Degree as required sorts to client convenient for distinguishing in the acquired customer group for having strong correlation with particular customer, example
Such as client high a possibility that carrying out illegal transaction can be come front.
According to further embodiments, it is determined with reference to each of client with reference to the client-related data of client according to one group
Indicate the characteristic of multiple client characteristics with reference to client；Also, use the characteristic training with reference to client
The Rating Model based on machine learning.
The Rating Model based on machine learning is targetedly trained before marking and queuing thereby, it is possible to realize.
According to further embodiments, being determined according to the client-related data of each client in the customer group indicates
The characteristic of multiple client characteristics of the client；Using the Rating Model based on machine learning based on the expression client's
The characteristic of multiple client characteristics scores to each client in the customer group；And based on described each
The scoring of client is ranked up the client in the customer group.This is provided is carried out using the Rating Model of above-mentioned training
The specific embodiment of scoring.
According to further embodiments, the extension rule data include by based on each visitor in the multiple client
The client-related data at family carries out the expression client relevant to the particular customer of data mining acquisition to the multiple client
The regular data of group.
According to further embodiments, the extension rule data further include predetermined indicate in the customer group
Each client and the regular data of the correlation of the particular customer；And/or the predetermined and acquisition customer group
The relevant regular data of purpose.
Using the extension rule data in above-mentioned at least two parties face, can more comprehensively, accurately obtain related to particular customer
Customer group.
According to further embodiments, by carrying out the client-related data of predefined template and the multiple client
To identify one or more customer groups in the multiple client, the predefined template definition is corresponding each for matching
The characteristic of relational structure and/or corresponding each client between client；And it will indicate one or more of clients
The data of customer group relevant to the particular customer in group are determined as indicating client relevant with the particular customer
The regular data of group.
According to further embodiments, it is determined from the multiple client based on the client-related data of the multiple client
One or more customer groups, so that each client in each customer group in one or more of customer groups is in institute
State the relative clients that at least predetermined quantity is all had in customer group；And it will indicate in one or more of customer groups
The data of customer group relevant to the particular customer are determined as indicating the rule of customer group relevant with the particular customer
Then data.
Foregoing provide obtain the two ways for indicating the regular data of customer group relevant to the particular customer.
According to another embodiment, a kind of equipment for obtaining customer group relevant to particular customer is provided, including
Memory；And processor, it is configured as when operation is from the program code of the memory, executes according to the present invention each
Method described in a embodiment.
According to another embodiment, a kind of machine readable media is provided, computer program code is stored, when the calculating
Machine program code is performed, and computer or processor is enabled to execute method described in each embodiment according to the present invention.
According to another embodiment, a kind of equipment for obtaining customer group relevant to particular customer, including first are provided
Acquiring unit is configured as obtaining the data for indicating the particular customer in multiple clients, and obtains and multiple clients
In the relevant client-related data of each client, the client-related data, which includes at least, indicates every in the multiple client
The customer relationship data of relationship between a client and other clients and the client characteristics data of the client；Second obtains list
Member is configured as obtaining predefined extension rule data；And determination unit, it is configured as based on client's dependency number
One or more client relevant to the particular customer in the multiple client is determined according to the extension rule data, to obtain
Take customer group relevant to the particular customer.
Detailed description of the invention
Fig. 1 shows the square of the equipment for obtaining customer group relevant to particular customer according to one embodiment
Figure；
Fig. 2 shows the sides according to the equipment for obtaining relevant to particular customer customer group of another embodiment
Block figure；
Fig. 3 shows the side of the equipment for obtaining customer group relevant to particular customer according to further embodiment
Block figure；
Fig. 4 shows the flow chart of the method for the acquisition customer group relevant to particular customer according to one embodiment；
Fig. 5 shows the process of the method for the acquisition customer group relevant to particular customer according to another embodiment
Figure.
Various aspects and features of the invention are described referring to above-mentioned attached drawing.Generally use the same or similar drawing reference numeral
To indicate identical component.Above-mentioned attached drawing is only schematical, and not restrictive.In the feelings for not departing from purport of the invention
Under condition, the size, shape of each element, label or appearance can change in above-mentioned attached drawing, without being limited to only
As only Figure of description is shown.
Specific embodiment
It is of the invention each to describe that the application of illegal transaction examination is carried out below with reference to the client to electronic payment platform
The application of embodiment, it should be understood that each embodiment of the invention using not limited to this, any need should be can be used in
It to be extended on the basis of particular customer under application scenarios of the customer range to obtain customer group relevant to particular customer.Cause
This, signified client is also not limited to the client to trade in electronic payment platform below.
Fig. 1 shows the square of the equipment 10 of the acquisition customer group relevant to particular customer according to one embodiment
Figure.The equipment 10 includes first acquisition unit 11, second acquisition unit 12, determination unit 13 and output unit 14.
First acquisition unit 11 obtains the data for indicating particular customer.The electronic payment platform of such as Alipay is to transaction visitor
Family provides the examination for being directed to illegal transaction, such as each transaction of current Alipay supervision platform transacting customer, and predetermined
In period, such as weekly, the list of the suspicious client supervised is exported, so that examiner examines.First acquisition unit 11
The list that the suspicious client can be obtained is stored these suspicious clients as particular customer.The particular customer can be pre-
One or more client in multiple clients to trade in section of fixing time.
In addition, the first acquisition unit 11 can also obtain multiple clients, such as in electronics branch in the predetermined amount of time
Pay all clients for trading on platform, client-related data.Above-mentioned particular customer is included in multiple client.The visitor
Family related data includes at least the customer relationship data for indicating the relationship between each client and other clients in multiple clients
With the client characteristics data of each client.Customer relationship data include the fund and non-fund pass between two clients of meaning in office
System.Fund relationship is referred in the equal trading activities of transferring accounts occurred between two clients in the predetermined amount of time, rather than fund
Relationship refers to any relationship other than the fund relationship occurred between two clients, such as in the predetermined amount of time
Same device relationships between two clients such as share the address mac, share cell phone address book contact person etc..Client characteristics data are
Characteristic relevant to an individual consumers, for example, the fund of the client within a predetermined period of time flows in and out the amount of money, hands over
Easy opponent's situation etc. and the client whether once received within a predetermined period of time system about illegal transaction alarm either
It is no that once illegal transaction was reported.
Second acquisition unit 12 obtains predefined extension rule data.The extension rule data are specified how current
Particular customer on the basis of be extended to obtain client relevant to the particular customer in the multiple clients to trade
The rule of group is specifically specified how the quantity for the client that extension needs to be investigated on the basis of particular customer.?
In preferred embodiment, which may include both sides regular data.On the one hand, the extension rule data packet
The predetermined regular data for indicating customer group relevant to particular customer is included, this can be by predetermined amount of time
Payment platform transacting customer carries out the data mining based on client-related data to predefine；Or it is determined before being able to use
The feature that can characterize customer group relevant to particular customer any regular data.On the other hand, the extension rule number
According to may include indicate customer group in each client and particular customer and/or acquisition customer group purpose correlation
Regular data.While it is preferred that customer group relevant to particular customer is determined using the regular data of these two aspects, but this
It is not limiting, the regular data of wherein one side also can be used, or introduce other regular numbers in the case of necessary
According to.
In order to predefine the regular data of expression customer group relevant to particular customer, it is able to use various methods pair
Multiple transacting customers carry out data mining.It in one embodiment, can be by by the visitor of predefined template and multiple clients
Family related data is matched to identify one or more customer groups in multiple clients, the predefined template definition pair
The characteristic for the relational structure and/or corresponding each client between each client answered.In a further embodiment, energy
It is enough to be existed first using the template of the relational structure (such as funds flow relationship) between each client for indicating such as illegal transaction
Certain doubtful customer groups are identified on the basis of relational graph between multiple clients, then again to every in the customer group of identification
A client further determines whether should belong to a member in the customer group of illegal transaction based on its characteristic.
It in another embodiment, can be based on client's phase of the multiple clients to trade in such as predetermined amount of time
It closes data and determines one or more customer group from multiple clients, so that each visitor in one or more of customer groups
Each client in the group of family all has the relative clients of at least predetermined quantity in the customer group.
Determined from multiple clients by above-mentioned different embodiment may for example be related to one of illegal transaction or
After multiple customer groups, the customer group relevant to the particular customer that will indicate in one or more of customer groups
Data be determined as indicating the regular data of customer group relevant to the particular customer.Above-mentioned determining expression and particular customer
The process of the regular data of relevant customer group also can by second acquisition unit 12 the multiple clients to trade visitor
It is executed on the basis of the related data of family.Of course it is also possible to it is previously determined above-mentioned regular data, and only according to this
It is used in the equipment of the embodiment of invention.
The regular data that determination unit 13 is based not only on expression customer group relevant to particular customer is also based on handing over
The client-related data of each client in easy multiple clients, including customer relationship data and client characteristics data, to determine
One or more client relevant to particular customer in multiple clients, to obtain customer group relevant to the particular customer.
One or more clients of the determination are included in customer group relevant to the particular customer.For example, if predetermined
Indicate that the regular data instruction of customer group relevant to particular customer uses cash flow in the same address mac and predetermined amount of time
The client that output is greater than some threshold value of the fund discharge of the particular customer is in customer group relevant to the particular customer
Member, then then can identify such client according to the client-related data of the multiple clients to trade.
In addition to the regular data of above-mentioned use expression customer group relevant to particular customer identifies relative clients group
Except, also the multiple clients to trade can further be judged based on client-related data, with identification with it is specific
The relevant client of client, for example, using each client and particular customer indicated in customer group and/or customers can be obtained
The regular data of the correlation of the purpose of body.
Such regular data for example including be related to fund magnitude, non-fund relationship, fund accounting, whether received it is alert
Report, the regular data whether being once reported.The regular data for being related to fund magnitude being capable of the regulation client customers to be determined
The relationship of the fund inflow and outflow total value of the fund inflow and outflow total value and particular customer of client in body；It is related to non-fund to close
The regular data of system can provide the client in the customer group to be determined and the non-fund relationship index between particular customer is
It is no to be greater than some threshold value；Be related to fund accounting regular data can provide client in the customer group to be determined inflow or
Whether the accounting for flowing in or out fund magnitude that outflow fund magnitude accounts for particular customer is greater than some threshold value；It relates to whether to receive
The regular data for crossing alarm can provide that the client in the customer group to be determined was received within a predetermined period of time about for example
The alarm of illegal transaction；Relate to whether just the regular data being once reported provides the client in the customer group to be determined
Such as illegal transaction was reported in the given time.It is above-mentioned to be related to fund magnitude, non-fund relationship, the rule of fund accounting
Data belong to the regular data of expression with the correlation of particular customer, and it is above-mentioned relate to whether to receive alarm, whether once by
The regular data reported belongs to the regular data of expression with the correlation for the purpose for obtaining customer group.It is also contemplated that being related to it
Regular data in terms of him.In addition, it is relevant to particular customer to identify to carry out arbitrary combination to above-mentioned regular data
Client, such as the client of the one or more met in above-mentioned rule in multiple clients can be determined as and the particular customer
Relevant client.It is equal using the client relevant to particular customer that the regular data for the use of above-mentioned two is identified from multiple clients
The member that can be confirmed as in customer group relevant to particular customer.
After each member that determination unit 13 has determined customer group relevant to particular customer, 14 energy of output unit
It is enough to export the relative customer group of the particular customer and determination together, so that trial person can be basic herein
On group's trial is carried out to these clients, to improve trial efficiency, and convenient for obtaining complete chain of evidence in trial.
On the other hand, which can also export the particular customer and client relevant to the particular customer
While each client in group, the output extension rule that wherein each client is met, in order to which the person of trial is tried.
For example, if some client is related to fund magnitude and whether received the regular data of alarm and be confirmed as and spy because meeting
A member in the relevant customer group of client is determined, then also exporting above-mentioned be related to while exporting its own for the client
Fund magnitude and the regular data for whether receiving alarm, or output expression relevant to the regular data, such as the client
Fund inflow and outflow total value and particular customer fund inflow and outflow total value relationship and the client data with alert.
As described above, one or more clients relevant to particular customer in multiple clients are determined in determination unit 13,
After obtaining customer group relevant to the particular customer, the particular customer and its correlation directly are exported in output unit 14
Customer group.However, in one embodiment, determination unit 13 is also in order to further expand the coverage area of trial client
It can be further expanded out and each visitor in the one or more client on the basis of identified one or more clients
The relevant client in family, to further increase the coverage area of identified customer group.Specifically, it is determined that unit 13 can be based on
In multiple clients that the client-related data of each client and the determination of extension rule data are traded in multiple clients and before
Each client in determining one or more clients relevant another or multiple clients, another or multiple visitors by this
Family is included in customer group relevant to particular customer, to export for output unit 14.In the implementation further expanded
It, being capable of extension rule data that directly second acquisition unit 12 obtains before use in example.It is also contemplated that by above-mentioned further
Extend scoring and sequence that each client in the customer group obtained is discussed below.
Fig. 2 shows the squares according to the equipment 20 of the acquisition customer group relevant to particular customer of another embodiment
Figure.The difference of equipment 20 shown in Fig. 2 and equipment 10 shown in FIG. 1 essentially consists in, and equipment 10 shown in Fig. 2 is further wrapped
Include scoring unit 15 and sequencing unit 16.The scoring unit 15 from determination unit 13 receive particular customer and it is identified with should
The relevant customer group of particular customer in one embodiment, can receive each in the case where the particular customer is multiple
The customer group relevant to the particular customer of particular customer and determination receives the customers of corresponding each particular customer
Body.Scoring unit 15 is able to use the Rating Model M based on machine learning and is based on each client relevant to the particular customer
The client-related data of each client in group scores to each client in the customer group.The Rating Model can
It is trained using the means of machine learning in advance.The example of one available Rating Model is to promote decision tree based on gradient
The Rating Model of algorithm.Sequencing unit 16 can the scoring based on each client in each customer group in the customer group
Client be ranked up.In this case, output unit 14 is based on the sequence and exports each customer group.It also it is contemplated that will
The particular customer, which is included in relative customer group, to score and sorts.
Fig. 3 shows the square of the equipment 30 of the acquisition customer group relevant to particular customer according to further embodiment
Figure.The difference of equipment 30 shown in Fig. 3 and equipment 20 shown in Fig. 2 essentially consists in, in equipment shown in Fig. 3 further
Including training unit 17.The training unit 17 can refer to each of client with reference to the client-related data of client according to one group
Determining indicates that this refers to the characteristic of multiple client characteristics of client；And it is instructed using the characteristic with reference to client
Practice the Rating Model M based on machine learning.This is referred to known to the scoring of client.Features described above data include but is not limited to table
Example such as " the inflow amount of money in the predetermined time ", the characteristic of " with mac relational index " and/or " whether reporting " etc..It can
With the expected characteristic for using any amount and any kind.It can be before the equipment using embodiment according to the present invention
Above-mentioned training process is realized first.
Using the Rating Model M trained as described above, determination unit 13 is according in determining customer group
Each client client-related data determine indicate the client multiple client characteristics characteristic.The unit 15 that scores uses
Based on the Rating Model of machine learning based on the characteristic for the multiple client characteristics for indicating client in determining customer group
Each client score.Scoring of the sequencing unit 16 based on each client carries out the client in determining customer group
Sequence.
Above-mentioned embodiment shown in -3 referring to Fig.1 describes each embodiment of the invention, and those skilled in the art should
It is understood that above-mentioned each embodiment is not limiting ,/modification/can be changed on the basis of each embodiment and is deleted
Except certain features therein, to obtain new technical solution.For example, training method defined by above-mentioned training unit 17 can
It is substituted by other training methods well known in the prior art.
Describe acquisition according to an embodiment of the invention customer group's relevant to particular customer below with reference to Fig. 4
The flow chart of method 400.
401, obtaining indicates that the data of the particular customer in multiple clients can connect in one embodiment from outside
The data for indicating the particular customer are received, which can determine from the multiple clients to trade in advance.
402, client-related data relevant to each client in multiple clients is obtained, the client-related data is extremely
Few includes the customer relationship data of the relationship between each client and other clients indicated in multiple clients and the visitor of the client
Family characteristic.
403, predefined extension rule data are obtained.As described above, in a preferred embodiment, the extension rule data
Regular data of both may include.On the one hand, which includes predetermined expression and particular customer phase
The regular data of the customer group of pass.On the other hand, which may include each visitor indicated in customer group
The regular data at family and particular customer and/or the correlation for the purpose for obtaining customer group.
404, determined relevant to particular customer one in multiple clients based on client-related data and extension rule data
A or multiple clients, to obtain customer group relevant to the particular customer.In a preferred embodiment, it 404, is also based on
Client-related data and extension rule data further determine that in the multiple client with it is every in one or more of clients
A client relevant another or multiple clients, another or multiple clients are included in related to the particular customer
Customer group in, to further expand customer group relevant to particular customer.
405, the particular customer and its relevant customer group are exported.
Fig. 5 shows the method according to an embodiment of the invention for obtaining customer group relevant to particular customer
Flow chart 500, wherein the processing in 501-503 is identical as the processing of 401-403 in flow chart 400 shown in Fig. 4.
504, in addition to processing identical with above-mentioned 404, in one embodiment, also according in determining customer group
The client-related data of each client determine the characteristic for indicating multiple client characteristics of the client.
Characteristic 505, using the Rating Model based on machine learning based on the multiple client characteristics for indicating client
It scores each client in the customer group of each determination.
506, based on the scoring of each client, the client in the customer group is ranked up.
507, which is inputted based on the sequence.
Although the flow chart referring to shown in Figure 4 and 5 describes each embodiment according to the method for the present invention.It can manage
Solution, can be handled accordingly, to constitute new technology in addition/modification/deletion on the basis of the flow chart of above-described embodiment
Scheme, to realize different effects.
In one embodiment, predefined extension rule data can be obtained in the following way: will be made a reservation for 403
The template of justice is matched with the client-related data of multiple clients to identify one or more customers in multiple clients
Body, the feature of relational structure and/or corresponding each client between the corresponding each client of the predefined template definition
Data；The data of the customer group relevant to particular customer indicated in one or more customer groups are determined as expression and institute
State the regular data of the relevant customer group of particular customer.
In another embodiment, predefined extension rule data can be obtained in the following way: being based on 403
The client-related data of multiple clients determines one or more customer group from multiple clients, so that one or more customers
Each client in each customer group in body all has the relative clients of at least predetermined quantity in the customer group；By table
Show that the data of the customer group relevant to particular customer in one or more customer groups are determined as indicating and the particular customer
The regular data of relevant customer group.
In another embodiment, it can be based on and the specific visitor 505 using the Rating Model based on machine learning
The client-related data of each client in the relevant customer group in family scores to each client in the customer group；And
And 506, it is based on the scoring, the client in the customer group is ranked up.
In a further embodiment, Rating Model can be instructed before executing the step 505 according to above-mentioned process
Practice, specifically, being determined with reference to each of client with reference to the client-related data of client according to one group indicates that this refers to client
Multiple client characteristics characteristic；And the Rating Model is trained using the characteristic with reference to client.
It is appreciated that in the equipment for obtaining customer group relevant to particular customer of each embodiment of the invention
The function of each unit and the process of method can be realized by computer program/software.These softwares can be loaded into
In the working storage of data processor, when running for the method that executes each embodiment according to the present invention.
Following the two of exemplary embodiment covering of the invention: computer journey of the invention is created that/used from the beginning
Sequence/software, and switch to existing program/software using computer program/software of the invention by means of updating.
Other embodiment according to the present invention provides a kind of machine (such as computer) readable medium, such as CD-ROM, wherein
The readable medium has the computer program code being stored in thereon, which enables calculating upon being performed
The method that machine or processor execute each embodiment according to the present invention.The machine readable media be, for example, together with other hardware or
The optical storage medium or solid state medium that part as other hardware is supplied.
The computer program for being used to execute the method for each embodiment according to the present invention can also be distributed otherwise,
Such as via internet or other wired or wireless telecommunication systems.Computer program also may be provided in such as WWW
On network, and can be from the working computer that such network is downloaded to data processor.
It is also to be understood that the equipment for obtaining customer group relevant to particular customer of each embodiment of the invention
In each unit and the process of method can also be realized by the combination of hardware or hardware and software.
In one embodiment, a kind of system for obtaining customer group relevant to particular customer can be by memory
It is realized with processor.Memory can store the computer of the method flow for running each embodiment according to the present invention
Program code；When running the program code from memory, processor executes the process of each embodiment according to the present invention.
It must be noted that the embodiment of the present invention is described with reference to different themes.In particular, some embodiments are references
Method type claim describes, and other embodiments are reference device type claims to describe.However, this field skill
Art personnel will learn from described above and below, unless otherwise specified, in addition to belong to a type of theme feature it is any
Other than combination, be related to any combination between the feature of different themes be also regarded as by this application discloses.Further, it is possible to combine
Whole features provide the synergistic effect simply summed it up greater than feature.
The present invention is described above by reference to specific embodiment, it will be appreciated by those skilled in the art that without departing substantially from the present invention
Spirit and essential characteristics in the case where, can realize technical solution of the present invention in various ways.Specific embodiment is only
It is only illustrative, and not restrictive.In addition, between these embodiments can any combination, to achieve the object of the present invention.
Protection scope of the present invention is defined by appended claims.
One word of " comprising " in description and claims is not excluded for the presence of other element or steps.It says in the description
The function of each element that is bright or recording in the claims can also be split or combine, by corresponding multiple element or list
One element is realized.
Claims (20)
1. a kind of method for obtaining customer group relevant to particular customer, including
Obtain the data for indicating the particular customer in multiple clients；
Client-related data relevant to each client in multiple clients is obtained, the client-related data, which includes at least, to be indicated
The customer relationship data of the relationship between each client and other clients in the multiple client and the client of the client are special
Levy data；
Obtain predefined extension rule data；With
Based on the client-related data and the extension rule data determine in the multiple client with the particular customer phase
One or more clients of pass, to obtain customer group relevant to the particular customer.
2. the method as described in claim 1 further includes
Based on the client-related data and the extension rule data determine in the multiple client with it is one or more of
Each client in client relevant another or multiple clients；With
It include in customer group relevant to the particular customer by another or multiple clients.
3. the method as described in claim 1 further includes
Each of customer group relevant to the particular customer using the Rating Model based on machine learning based on acquisition
The client-related data of client scores to each client in the customer group；With
Based on the scoring, the client in the customer group is ranked up.
4. method as claimed in claim 3 further includes
Being determined with reference to each of client with reference to the client-related data of client according to one group indicates described with reference to the multiple of client
The characteristic of client characteristics；With
Use the Rating Model based on machine learning described in the characteristic training with reference to client.
5. method as claimed in claim 4 further includes
The multiple client characteristics for indicating the client are determined according to the client-related data of each client in the customer group
Characteristic；
The characteristic pair using the Rating Model based on machine learning based on the multiple client characteristics for indicating the client
Each client in the customer group scores；With
Based on the scoring of each client, the client in the customer group is ranked up.
6. method according to any one of claims 1 to 5, wherein the extension rule data include by based on described more
Expression and the spy of the client-related data of each client in a client to the multiple client progress data mining acquisition
Determine the regular data of the relevant customer group of client.
7. method as claimed in claim 6, wherein the extension rule data further include
The predetermined each client and the regular data of the correlation of the particular customer indicated in the customer group；
And/or
Predetermined regular data relevant to purpose that is obtaining the customer group.
8. method as claimed in claim 6, further including
It is identified by matching predefined template with the client-related data of the multiple client in the multiple visitor
One or more customer groups in family, relational structure between the corresponding each client of the predefined template definition and/
Or the characteristic of corresponding each client；With
The data of the customer group relevant to the particular customer indicated in one or more of customer groups are determined as
Indicate the regular data of customer group relevant to the particular customer.
9. method as claimed in claim 6, further including
One or more customer group is determined from the multiple client based on the client-related data of the multiple client, so that
Each client in each customer group in one or more of customer groups all has at least in the customer group
The relative clients of predetermined quantity；
The data of the customer group relevant to the particular customer indicated in one or more of customer groups are determined as
Indicate the regular data of customer group relevant to the particular customer.
10. a kind of equipment for obtaining customer group relevant to particular customer, including
Memory；With
Processor is configured as executing and appointing in -9 according to claim 1 when operation is from the program code of the memory
Method described in one.
11. a kind of machine readable media, computer program code is stored, when the computer program code is performed, is enabled
Computer or processor execute method according to claim 1 to 9.
12. a kind of equipment for obtaining customer group relevant to particular customer, including
First acquisition unit is configured as obtaining the data for indicating the particular customer in multiple clients, and obtain with
The relevant client-related data of each client in multiple clients, the client-related data, which includes at least, indicates the multiple visitor
The customer relationship data of the relationship between each client and other clients in family and the client characteristics data of the client；
Second acquisition unit is configured as obtaining predefined extension rule data；With
Determination unit is configured as determining the multiple client based on the client-related data and the extension rule data
In one or more clients relevant to the particular customer, to obtain customer group relevant to the particular customer.
13. equipment as claimed in claim 12, wherein the determination unit is additionally configured to
Based on the client-related data and the extension rule data determine in the multiple client with it is one or more of
Each client in client relevant another or multiple clients；And
It include in customer group relevant to the particular customer by another or multiple clients.
14. equipment as claimed in claim 12, further includes
Score unit, is configured with the Rating Model based on machine learning based on the related to the particular customer of acquisition
Customer group in the client-related data of each client score each client in the customer group；With
Sequencing unit is configured as being ranked up the client in the customer group based on the scoring.
15. equipment as claimed in claim 14, further includes
Training unit, being configured as being determined with reference to each of client with reference to the client-related data of client according to one group indicates
The characteristic of multiple client characteristics with reference to client；And use the characteristic training base with reference to client
In the Rating Model of machine learning.
16. equipment as claimed in claim 15, wherein
The determination unit is additionally configured to be determined according to the client-related data of each client in the customer group and indicate
The characteristic of multiple client characteristics of the client；
The scoring unit is also configured to use the Rating Model based on machine learning based on the multiple visitors for indicating the client
The characteristic of family feature scores to each client in the customer group；And
The sequencing unit is additionally configured to arrange the client in the customer group based on the scoring of each client
Sequence.
17. the equipment as described in any one of claim 12-16, wherein the extension rule data include by based on institute
The client-related data for stating each client in multiple clients carries out expression and the institute of data mining acquisition to the multiple client
State the regular data of the relevant customer group of particular customer.
18. equipment as claimed in claim 17, wherein the extension rule data further include
The predetermined each client and the regular data of the correlation of the particular customer indicated in the customer group；
And/or
Predetermined regular data relevant to purpose that is obtaining the customer group.
19. equipment as claimed in claim 17, wherein the second acquisition unit is additionally configured to
It is identified by matching predefined template with the client-related data of the multiple client in the multiple visitor
One or more customer groups in family, relational structure between the corresponding each client of the predefined template definition and/
Or the characteristic of corresponding each client；And
The data of the customer group relevant to the particular customer indicated in one or more of customer groups are determined as
Indicate the regular data of customer group relevant to the particular customer.
20. equipment as claimed in claim 17, wherein the second acquisition unit is additionally configured to
One or more customer group is determined from the multiple client based on the client-related data of the multiple client, so that
Each client in each customer group in one or more of customer groups all has at least in the customer group
The relative clients of predetermined quantity；
The data of the customer group relevant to the particular customer indicated in one or more of customer groups are determined as
Indicate the regular data of customer group relevant to the particular customer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811522367.4A CN110046910B (en) | 2018-12-13 | 2018-12-13 | Method and equipment for judging validity of transaction performed by customer through electronic payment platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811522367.4A CN110046910B (en) | 2018-12-13 | 2018-12-13 | Method and equipment for judging validity of transaction performed by customer through electronic payment platform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110046910A true CN110046910A (en) | 2019-07-23 |
| CN110046910B CN110046910B (en) | 2023-04-14 |
Family
ID=67273661
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811522367.4A Active CN110046910B (en) | 2018-12-13 | 2018-12-13 | Method and equipment for judging validity of transaction performed by customer through electronic payment platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110046910B (en) |
Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030065520A1 (en) * | 2001-09-28 | 2003-04-03 | Jutzi Curtis E. | System and method for selecting relevant products to be transparently acquired for a consumer |
| US20040181551A1 (en) * | 2003-03-14 | 2004-09-16 | Goodman John F. | Method of providing online database containing proprietary information |
| CN101147169A (en) * | 2004-12-21 | 2008-03-19 | 西特尔私人有限公司 | Change management |
| CN102004889A (en) * | 2010-11-30 | 2011-04-06 | 石家庄开发区冀科双实科技有限公司 | Encryption method of hierarchical data in hierarchical tree system |
| CN102083010A (en) * | 2009-11-26 | 2011-06-01 | 中国移动通信集团公司 | Method and equipment for screening user information |
| US20140082568A1 (en) * | 2012-09-14 | 2014-03-20 | Ancestry.com Operations, Inc. | System and method for creating a family tree data structure |
| CN104765874A (en) * | 2015-04-24 | 2015-07-08 | 百度在线网络技术（北京）有限公司 | Method and device for detecting click-cheating |
| CN105488211A (en) * | 2015-12-11 | 2016-04-13 | 成都陌云科技有限公司 | Method for determining user group based on feature analysis |
| CN105653518A (en) * | 2015-12-25 | 2016-06-08 | 北京理工大学 | Specific group discovery and expansion method based on microblog data |
| CN106504029A (en) * | 2016-11-08 | 2017-03-15 | 山东大学 | A kind of gas station's Method for Sales Forecast method based on customer group's behavior analysiss |
| CN106651564A (en) * | 2016-11-18 | 2017-05-10 | 中国农业银行股份有限公司 | Target customer obtaining method and system |
| CN107517207A (en) * | 2017-03-13 | 2017-12-26 | 平安科技（深圳）有限公司 | Server, auth method and computer-readable recording medium |
| CN108141714A (en) * | 2015-09-25 | 2018-06-08 | 三星电子株式会社 | The personalization of mobile health application, companion derive the automatic structure of message |
| CN108174359A (en) * | 2017-12-11 | 2018-06-15 | 成都欧督系统科技有限公司 | For the message push processing method of mobile terminal |
| CN108428138A (en) * | 2017-02-13 | 2018-08-21 | 北京京东尚科信息技术有限公司 | A kind of client's survival Analysis device and analysis method based on customer grouping |
| CN108573432A (en) * | 2018-04-23 | 2018-09-25 | 重庆南米电子商务有限公司 | Transaction supervisory systems and method for e-commerce |
| CN108664483A (en) * | 2017-03-28 | 2018-10-16 | 北大方正集团有限公司 | The management method and management system of specific user group |
| CN108776857A (en) * | 2018-04-22 | 2018-11-09 | 平安科技（深圳）有限公司 | NPS short messages method of investigation and study, system, computer equipment and storage medium |
| CN108846700A (en) * | 2018-06-15 | 2018-11-20 | 山东外贸职业学院 | A kind of big data intelligent marketing system and marketing method |
-
2018
- 2018-12-13 CN CN201811522367.4A patent/CN110046910B/en active Active
Patent Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030065520A1 (en) * | 2001-09-28 | 2003-04-03 | Jutzi Curtis E. | System and method for selecting relevant products to be transparently acquired for a consumer |
| US20040181551A1 (en) * | 2003-03-14 | 2004-09-16 | Goodman John F. | Method of providing online database containing proprietary information |
| CN101147169A (en) * | 2004-12-21 | 2008-03-19 | 西特尔私人有限公司 | Change management |
| CN102083010A (en) * | 2009-11-26 | 2011-06-01 | 中国移动通信集团公司 | Method and equipment for screening user information |
| CN102004889A (en) * | 2010-11-30 | 2011-04-06 | 石家庄开发区冀科双实科技有限公司 | Encryption method of hierarchical data in hierarchical tree system |
| US20140082568A1 (en) * | 2012-09-14 | 2014-03-20 | Ancestry.com Operations, Inc. | System and method for creating a family tree data structure |
| CN104765874A (en) * | 2015-04-24 | 2015-07-08 | 百度在线网络技术（北京）有限公司 | Method and device for detecting click-cheating |
| CN108141714A (en) * | 2015-09-25 | 2018-06-08 | 三星电子株式会社 | The personalization of mobile health application, companion derive the automatic structure of message |
| CN105488211A (en) * | 2015-12-11 | 2016-04-13 | 成都陌云科技有限公司 | Method for determining user group based on feature analysis |
| CN105653518A (en) * | 2015-12-25 | 2016-06-08 | 北京理工大学 | Specific group discovery and expansion method based on microblog data |
| CN106504029A (en) * | 2016-11-08 | 2017-03-15 | 山东大学 | A kind of gas station's Method for Sales Forecast method based on customer group's behavior analysiss |
| CN106651564A (en) * | 2016-11-18 | 2017-05-10 | 中国农业银行股份有限公司 | Target customer obtaining method and system |
| CN108428138A (en) * | 2017-02-13 | 2018-08-21 | 北京京东尚科信息技术有限公司 | A kind of client's survival Analysis device and analysis method based on customer grouping |
| CN107517207A (en) * | 2017-03-13 | 2017-12-26 | 平安科技（深圳）有限公司 | Server, auth method and computer-readable recording medium |
| CN108664483A (en) * | 2017-03-28 | 2018-10-16 | 北大方正集团有限公司 | The management method and management system of specific user group |
| CN108174359A (en) * | 2017-12-11 | 2018-06-15 | 成都欧督系统科技有限公司 | For the message push processing method of mobile terminal |
| CN108776857A (en) * | 2018-04-22 | 2018-11-09 | 平安科技（深圳）有限公司 | NPS short messages method of investigation and study, system, computer equipment and storage medium |
| CN108573432A (en) * | 2018-04-23 | 2018-09-25 | 重庆南米电子商务有限公司 | Transaction supervisory systems and method for e-commerce |
| CN108846700A (en) * | 2018-06-15 | 2018-11-20 | 山东外贸职业学院 | A kind of big data intelligent marketing system and marketing method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110046910B (en) | 2023-04-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110991552B (en) | Isolated forest model construction and prediction method and device based on federal learning | |
| CN108228706A (en) | For identifying the method and apparatus of abnormal transaction corporations | |
| CN109922032B (en) | Method, device, equipment and storage medium for determining risk of logging in account | |
| CN110992167A (en) | Bank client business intention identification method and device | |
| CN109376766B (en) | Portrait prediction classification method, device and equipment | |
| Ray | Fraud detection in e-Commerce using machine learning | |
| CN111078880B (en) | Sub-application risk identification method and device | |
| CN111629010B (en) | Malicious user identification method and device | |
| CN109949154A (en) | Customer information classification method, device, computer equipment and storage medium | |
| CN111340565B (en) | Information recommendation method, device, equipment and storage medium | |
| CN108038687A (en) | Method of commerce, server and computer-readable recording medium based on speech recognition | |
| CN107491674A (en) | Feature based information carries out the method and device of user's checking | |
| CN108109066A (en) | A kind of credit scoring model update method and system | |
| CN111428217A (en) | Method and device for identifying cheat group, electronic equipment and computer readable storage medium | |
| CN109685643A (en) | Loan audit risk grade determines method, apparatus, equipment and storage medium | |
| CN109472586A (en) | Strategy determines method and device, storage medium, electronic device | |
| CN110046889A (en) | A kind of detection method, device and the server of abnormal behaviour main body | |
| CN107274042A (en) | A kind of business participates in the Risk Identification Method and device of object | |
| CN116302640A (en) | Abnormality analysis method, abnormality analysis device, abnormality analysis computer device, and abnormality analysis program | |
| CN106897880B (en) | Account risk assessment method and equipment | |
| CN106604281A (en) | Method and equipment for determining stability of mobile phone number | |
| CN117376228A (en) | Network security testing tool determining method and device | |
| CN110046910A (en) | The method and apparatus for obtaining customer group relevant to particular customer | |
| CN116976664A (en) | Risk merchant prediction method, system, computer and readable storage medium | |
| CN110705994A (en) | Risk user detection method and device |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20201013 Address after: English genus Applicant after: Innovative advanced technology Co.,Ltd. Address before: English genus Applicant before: Advanced innovation technology Co.,Ltd. Effective date of registration: 20201013 Address after: English genus Applicant after: Advanced innovation technology Co.,Ltd. Address before: A four-storey 847 mailbox in Grand Cayman Capital Building, British Cayman Islands Applicant before: Alibaba Group Holding Ltd. |
|
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20230307 Address after: 801-10, Section B, 8th floor, 556 Xixi Road, Xihu District, Hangzhou City, Zhejiang Province Applicant after: Ant financial (Hangzhou) Network Technology Co.,Ltd. Address before: Cayman Islands Applicant before: Innovative advanced technology Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |