CN108564467A - A kind of determination method and apparatus of consumer's risk grade - Google Patents
A kind of determination method and apparatus of consumer's risk grade Download PDFInfo
- Publication number
- CN108564467A CN108564467A CN201810436177.4A CN201810436177A CN108564467A CN 108564467 A CN108564467 A CN 108564467A CN 201810436177 A CN201810436177 A CN 201810436177A CN 108564467 A CN108564467 A CN 108564467A
- Authority
- CN
- China
- Prior art keywords
- user
- identified
- keyword
- risk
- information
- 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
- 238000000034 method Methods 0.000 title claims abstract description 44
- 230000002159 abnormal effect Effects 0.000 claims abstract description 58
- 230000000694 effects Effects 0.000 claims description 47
- 230000004044 response Effects 0.000 claims description 16
- 238000004590 computer program Methods 0.000 claims description 14
- 230000009466 transformation Effects 0.000 claims description 9
- 230000001360 synchronised effect Effects 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 5
- 238000000151 deposition Methods 0.000 claims 1
- 230000010365 information processing Effects 0.000 abstract description 2
- 230000006399 behavior Effects 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 230000003542 behavioural effect Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 239000000284 extract Substances 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000000977 initiatory effect Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 210000001367 artery Anatomy 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000007667 floating Methods 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 239000010931 gold Substances 0.000 description 1
- 229910052737 gold Inorganic materials 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 210000003462 vein Anatomy 0.000 description 1
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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
Abstract
The present invention is suitable for technical field of information processing, provides a kind of determination method and apparatus of consumer's risk grade, including:Receive the user information of user to be identified;User information is matched with each keyword in keyword dictionary, determines the keyword sequence of user to be identified;According to each keyword sequence for having identified user in keyword sequence and customer data base, determine user to be identified and each matching degree identified between user, and choose the identification user that matching degree is more than preset matching degree threshold value, the association user as user to be identified;Based on association user, create with the customer relationship network of user-center to be identified, and mark the abnormal user for including in customer relationship network;According to the number of abnormal user in customer relationship network, the risk class of user to be identified is determined.Risk class also can be quickly determined for newly added user, effective management and control is carried out to transaction request, reduces the risk of capital loss in the present invention.
Description
Technical field
The invention belongs to technical field of information processing more particularly to a kind of determination method and apparatus of consumer's risk grade.
Background technology
With economic continuous development, the frequency that user initiates transaction request to each financial institution is also higher and higher, gold
Melt the quantity that mechanism Adds User daily to be also continuously increased.For the transaction request for the initiation that Adds User, simultaneously due to financial institution
The historical transaction record to Add User is not recorded, can not determine that the credit grade of such user, financial institution can generally be agreed to
The transaction request to Add User is initiated in response.However the illegal user in part can be noted by the data of friend or relatives at one's side
The new user of volume can not be caused using financial institution to loophole that the transaction request of new registration user is managed come the fund of defrauding of
Financial institution assumes responsibility for larger investment risk.It can be seen that the determination technology of existing consumer's risk grade, primarily directed to
There are the users of historical transaction record can accurately carry out risk class assessment, and newly added user can not be carried out
It determines, improves the risk of capital loss.
Invention content
In view of this, an embodiment of the present invention provides a kind of determination method of consumer's risk grade and its equipment, to solve
The determination method of existing consumer's risk grade, primarily directed to there are the users of historical transaction record can be accurately into sector-style
The assessment of dangerous grade, and the problem of can not be determined for newly added user, improve the risk of capital loss.
The first aspect of the embodiment of the present invention provides a kind of determination method of consumer's risk grade, including:
Receive the user information of user to be identified;
The user information is matched with each keyword in preset keyword dictionary, is determined described to be identified
The keyword sequence of user;
According to each keyword sequence for having identified user in the keyword sequence and customer data base, determine described in
User to be identified and each matching degree identified between user, and choose the matching degree more than preset matching degree threshold value
Identify user, the association user as the user to be identified;
Based on the association user, create with the customer relationship network of the user-center to be identified, and described in label
The abnormal user for including in customer relationship network;The abnormal user is specially the use that risk class is more than preset risk threshold value
Family;
According to the number of abnormal user in the customer relationship network, the risk class of the user to be identified is determined.
The second aspect of the embodiment of the present invention provides a kind of consumer's risk grade locking equipment, including memory, place really
It manages device and is stored in the computer program that can be run in the memory and on the processor, the processor executes institute
Each step of first aspect is realized when stating computer program.
The third aspect of the embodiment of the present invention provides a kind of computer readable storage medium, the computer-readable storage
Media storage has computer program, and each step of first aspect is realized when the computer program is executed by processor.
The determination method and terminal device for implementing a kind of consumer's risk grade provided in an embodiment of the present invention have with following
Beneficial effect:
The embodiment of the present invention is by receiving user to be identified, that is, when the user information to Add User, it is determined that the user
The corresponding keyword sequence of information, and identified that user carries out based on each in the keyword sequence and customer data base
Match, determine the association user of the user to be identified, then create the customer relationship net of the user to be identified, and marks the user
The abnormal user for including in network of personal connections;The risk class that the user to be identified is determined according to the number of abnormal user, to be based on
The risk class can be made whether response Client-initiated transaction request.With the determination method phase of existing consumer's risk grade
Than the embodiment of the present invention does not depend on historical transaction record to determine the risk class of user, and can be based on the pass of the user
The number for the abnormal user for including in combination family judges the member whether user belongs in abnormal user group, and is based on number
Difference obtain the risk class of the user, also can quickly determine risk class hence for newly added user, to transaction ask
It asks and carries out effective management and control, reduce the risk of capital loss.
Description of the drawings
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description be only the present invention some
Embodiment for those of ordinary skill in the art without having to pay creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is a kind of implementation flow chart of the determination method for consumer's risk grade that first embodiment of the invention provides;
Fig. 2 is the specific implementation stream of the determination method S104 for consumer's risk grade that second embodiment of the invention provides a kind of
Cheng Tu;
Fig. 3 is the specific implementation stream of the determination method S103 for consumer's risk grade that third embodiment of the invention provides a kind of
Cheng Tu;
Fig. 4 is a kind of determination method specific implementation flow chart for consumer's risk grade that fourth embodiment of the invention provides;
Fig. 5 is a kind of specific implementation flow of the determination method for consumer's risk grade that fourth embodiment of the invention provides
Figure;
Fig. 6 is a kind of structure diagram for consumer's risk grade locking equipment really that one embodiment of the invention provides;
Fig. 7 is a kind of schematic diagram for consumer's risk grade locking equipment really that another embodiment of the present invention provides.
Specific implementation mode
In order to make the purpose , technical scheme and advantage of the present invention be clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
The embodiment of the present invention is by receiving user to be identified, that is, when the user information to Add User, it is determined that the user
The corresponding keyword sequence of information, and identified that user carries out based on each in the keyword sequence and customer data base
Match, determine the association user of the user to be identified, then create the customer relationship net of the user to be identified, and marks the user
The abnormal user for including in network of personal connections;The risk class that the user to be identified is determined according to the number of abnormal user, to be based on
The risk class can be made whether response Client-initiated transaction request, solve the determination side of existing consumer's risk grade
Method, primarily directed to there are the users of historical transaction record can accurately carry out risk class assessment, and for newly added
User can not but be determined, the problem of improving the risk of capital loss.
In embodiments of the present invention, the executive agent of flow is consumer's risk grade locking equipment really.The consumer's risk etc.
Really locking equipment includes but not limited to grade:The users such as laptop, computer, server, tablet computer and smart mobile phone
Risk class locking equipment really.Fig. 1 shows the reality of the determination method for the consumer's risk grade that first embodiment of the invention provides
Existing flow chart, details are as follows:
In S101, the user information of user to be identified is received.
In the present embodiment, it before user needs to initiate relevant transactional operation, needs to register one in the financial institution
User account, and fill in relevant user information.Therefore, locking equipment can be that user's initiation registration is asked to consumer's risk grade really
The server asked, or independently of the locking equipment really of the risk class outside the server, for the institute in the server
There is registration user to carry out determining for risk class to operate.Wherein, which includes but not limited to:Name, gender, the age,
The information such as inhabitation address, Business Name, CompanyAddress, social relationships.
In the present embodiment, it if locking equipment is the server registered needed for user to consumer's risk grade really, receives
To after the registration request of user, determine that equipment can acquire the every terms of information that the user fills in enrollment page, and generate the use
The corresponding user information in family, and execute the relevant operation of S102;If the determination equipment is the independent service with registration needed for user
Device, then server can identify use by each in the user information and customer data base after receiving the user information of user
The user information at family is matched, and determines whether the user is registered user, if in the presence of identified the user information of user with
The user information of the user to be identified is consistent, then returns to a registered information to the terminal of the user to be identified;If conversely, appointing
One has identified that the user information of user is inconsistent with the user information of the user to be identified, then by the user of the user to be identified
Information is sent to determining equipment, determines that equipment after receiving user information to be identified, then executes the relevant operation of S102.
Optionally, other than receiving registration information this event triggering pattern, determination is set the triggering mode of S101
Standby to carry out risk identification to each user information in customer data base with preset time interval, i.e. customer data base exists
After the user information for receiving user to be identified, then a mark to be identified can be added to the user to be identified, it is default when reaching
Detection cycle when, determine equipment then and can be extracted from customer data base that the user information with mark to be identified executes user's wind
Dangerous grade constant current journey really.
In S102, the user information is matched with each keyword in preset keyword dictionary, is determined
The keyword sequence of the user to be identified.
In the present embodiment, determine that equipment is stored with a keyword dictionary, which includes multiple keywords, often
A keyword is arranged in order with fixed sequence, i.e., each keyword has a fixed dictionary number in keyword dictionary.Cause
This, different user informations is matched with keyword dictionary, and the element institute of same position is right in obtained keyword sequence
The keyword answered is identical.
In the present embodiment, determine that equipment can be by each keyword progress in user information and the keyword dictionary
Match, judges in the user information whether to include the keyword.If including the keyword, the corresponding position in keyword sequence
The keyword is inserted in element, to which the keyword sequence of output then indicates to contain those keywords in the user information.By
In Keywords matching for user information matching, accuracy is higher, can determine the key for including in the user information
Which information has, and then the number for the key message for including in two user informations is compared, and determines two user informations
Whether it is there are association, so that it is determined that whether two users are association user.Illustratively, the user information of a certain user is
" 16 building, Nanshan District, Shenzhen City seashore group mansion " is by the keyword sequence identified after keyword dictionary:Shenzhen,
South Mountain, seashore, group, 16 };And the user information of another user is " 5 building, Enterprises of Futian District seashore group building ", passes through pass
Obtained keyword sequence is after keyword matching:{ Shenzhen, Feitian, seashore, group, 5 }, it can be seen that, two user information phases
The number of same keyword is 3, can identify that two user informations are associated user information.
Optionally, which can acquire from upper server, and upper server can periodically update should
Keyword dictionary.With the continuous development of society, it usually will appear emerging word, therefore keyword dictionary needs timing to carry out more
Newly, to improve the accuracy identified to keyword in user information.
Optionally, the method for determining keyword sequences with keyword dictionary based on user information can be:Initialization is crucial
Word sequence, each element in keyword sequence sequentially correspond to a keyword in the keyword dictionary respectively;According to each
Whether order of a keyword in the keyword dictionary judges in the user information to include each keyword successively;If
Keyword in the keyword dictionary is contained in the user information, then the keyword is arranged in the keyword sequence
Element be the first place value, such as with " 1 " mark include the keyword;If keyword in the keyword dictionary not by comprising
In the user information, then it is the second place value that element of the keyword in the keyword sequence, which is arranged, such as is identified with " 2 "
Not comprising the keyword;Place value based on each element generates the keyword sequence of the user.That is the keyword sequence
Specially there is what the first place value and the second place value constituted to have the sequence centainly to put in order.
In S103, according to each crucial word order for having identified user in the keyword sequence and customer data base
Row, determine the user to be identified and each matching degree identified between user, and choose the matching degree and be more than default
Identification user with degree threshold value, the association user as the user to be identified.
In the present embodiment, user can be stored in by having recorded the customer data base of each user information for having identified user
Risk class in locking equipment, in this case, determines that equipment can directly read the user data being locally stored in module really
Library can acquire the user information of each user, and particularly, which can be a user database server.It should
Customer data base is also used as an independent database server in outside, in this case, determines that equipment takes with the database
Business device establishes communication connection, determines that equipment can send an instruction for authenticating, database service to the database server
Device judges that the customer data base corresponding to the identification equipment opens this really if the instruction authenticates successfully after receiving the instruction
The acquisition permission of the user data of locking equipment, then identification equipment the use of each user can be obtained by the database server
Family information.Wherein, it has each identified and has been also recorded for its corresponding keyword sequence in the user information of user.
In the present embodiment, it is determined that user to be identified and after having identified keyword sequence corresponding to user, it can be with
User to be identified and each matching degree identified between user are calculated based on keyword sequence, it is then whether big according to matching degree
Whether it is association user between preset matching degree threshold value determines two users.
In the present embodiment, which can be the number of same keyword in keyword sequence, in this case,
Determine the keyword sequence and same keyword in any keyword sequence for having identified user that equipment can count user to be identified
Number, using the number as the matching degree between two users, if the number of the same keyword between two users is big
In preset matching threshold, then two users association user each other is identified;If being preset conversely, the number of the same keyword is less than
Correlation threshold, then identify that two users are unrelated user.The matching threshold can also be a similarity threshold, in the situation
Under, the similarities of two keyword sequences can be calculated by creating equipment, i.e. whether the element of same position is identical, and calculation formula can be with
ForWherein, S is the number of identical element, and Q is the total number of element, similar between two keyword sequences to obtain
Degree, and the similarity is compared with similarity threshold, determines between two users whether be association user.Preferably, exist
When determining whether two keyword sequences are similar, other than whether the element of judgement same position is identical, it can also judge identical
Whether the element of position similar, create equipment have recorded with the associated fuzzy keyword of each keyword, such as " sun " with
" day " two keywords can identify between two keywords there is relevance, and know due to all same in kind of reference
It is similar key that other same position, which has the keyword of key relationship, when calculating similarity, in addition to considering same keyword
Outside number, the number of similar key is further accounted for, in this case, calculating the formula of similarity can be:
Wherein, a1、a2For weighting coefficient, Y is the similarity between two keyword sequences;S is the element of same position
Identical number;Sl is the similar number of element of same position;Q is the total number of element in keyword sequence.
In S104, it is based on the association user, is created with the customer relationship network of the user-center to be identified, and
Mark the abnormal user for including in the customer relationship network;The abnormal user is specially that risk class is more than preset risk
The user of threshold value.
In the present embodiment, equipment is determined after all association users of user to be identified are determined, can be based on the pass
Family structure is combined with the customer relationship network of user-center to be identified, to administrator can by the customer relationship network,
Quickly determine user to be identified and each incidence relation identified between user.It should be noted that each having identified user
There are corresponding user property, the value of user property that can be:Normal users and abnormal user.Wherein, normal users have
Body is the user that risk class is less than or equal to preset risk threshold value;And it is more than preset that abnormal user, which is specially risk class,
The user of risk threshold value.Certainly, it except through having identified that the risk class of user judges outside its user property, can also obtain
The historical transaction record for identifying user, extracts the transaction feature value of historical transaction record, by the transaction feature value and preset standard
Range of characteristic values is compared, if within the scope of the Standard Eigenvalue, identifying that the user is normal users, conversely, then identification should
User is abnormal user.
In the present embodiment, which not only has recorded the association that user to be identified is associated between user
Relationship further comprises the incidence relation between each association user of the user to be identified.For example, user A is to wait for user B
Identify the association user of user, and there is also incidence relations between user A and user B, then can exist in customer relationship network
One association access, to indicate there is association between above-mentioned two user.
In S105, according to the number of abnormal user in the customer relationship network, the wind of the user to be identified is determined
Dangerous grade.
In the present embodiment, determine that equipment after all abnormal users, can count the use in customer relationship network is marked
The number for the abnormal user for including in the relational network of family, using the number as the risk class for calculating user to be identified, to real
The purpose of risk class assessment can be also realized referring now to the user just registered.Optionally it is determined that equipment is preset with risk class
Conversion hash function, the number of the abnormal user in customer relationship network is imported into the conversion hash function, you can defeated
Go out the risk class corresponding to the number, the risk class as the user.
Optionally, when calculating the risk class of the user to be identified, the risk class of each abnormal user is further accounted for.
The risk initial value of the user to be identified can be determined based on the number of abnormal user by determining equipment first, and by each abnormal user
Risk class and risk initial value imported into risk class computation model, determine the risk class of the user to be identified.Its
In, the risk class computation model is as follows:
Wherein, RiskDegree is the risk class of user to be identified;InitalRisk (N) is the risk of user to be identified
Initial value;Riskdegree0(Abnmluseri) be each abnormal user risk class, CofftiFor i-th abnormal user
Default weight;N is the number of abnormal user in user to be identified.
Above as can be seen that a kind of determination method of consumer's risk grade provided in an embodiment of the present invention is waited for by receiving
Identify user, that is, when the user information to Add User, it is determined that the corresponding keyword sequence of the user information, and it is based on the key
Word sequence has identified that user matches with each in customer data base, determines the association user of the user to be identified, then
The customer relationship net of the user to be identified is created, and marks the abnormal user for including in the customer relationship net;According to abnormal use
The number at family determines the risk class of the user to be identified, to can be made whether that response user initiates based on the risk class
Transaction request.Compared with the determination method of existing consumer's risk grade, the embodiment of the present invention does not depend on historical transaction record
It determines the risk class of user, and can be based on the number for the abnormal user for including in the association user of the user, judge
Whether the user belongs to the member in abnormal user group, and the difference based on number obtains the risk class of the user, to right
Risk class also can be quickly determined in newly added user, and effective management and control is carried out to transaction request, reduces the wind of capital loss
Danger.
Fig. 2 shows the specific realities of the determination method S104 of consumer's risk grade of second embodiment of the invention offer a kind of
Existing flow chart.It is shown in Figure 2, relative to embodiment described in Fig. 1, a kind of determination of consumer's risk grade provided in this embodiment
S104 includes S1041 and S1042 in method, and specific details are as follows:
In S1041, each association user for having identified user is obtained, with the M degree of the determination user to be identified
Association user;Wherein, there are M-1 users to save in the associated path between the M degree association user and the user to be identified
Point;M is the positive integer more than or equal to 1.
In the present embodiment, consumer's risk grade really locking equipment when building the customer relationship network of user to be identified,
The association user of the user to be identified is not only obtained, the association user for having identified user can be also obtained, determines the M of user to be identified
Spend association user.Wherein, there are M-1 user nodes in the associated path between the M degree association user and user to be identified, i.e.,
It indicates that M degree association users needs are linked up with user to be identified, interacted, needs M-1 user collaboration ability on the path
It completes.
It should be noted that there may be a plurality of associated path between user to be identified and M degree association users, and this implementation
Example be to be identified user and M degree association user between the shortest associated path, i.e. institute selected when determining M degree association users
The minimum associated path of the user node number of process.
In the present embodiment, the numerical value of M can be configured by user based on demand, particularly, when the value of M is 1, then
Indicate the association user for only including user to be identified in the customer relationship network.Preferably, the value of M is not more than 6.Based on six degree of people
The relation principle of arteries and veins need to only pass through 5 users and association may be present arbitrarily between all natural persons, therefore in order to reduce network
The value of redundancy, the M should be not more than 6.
It should be noted that in S1041 during obtaining M degree association users, determine equipment also and can obtain M degree with
Interior association user, i.e. M-1 degree association user, M-2 degree association user ..., 1 degree of association user (association of user i.e. to be identified
User).
In S1042, according to the M degree association user, create with the M degree customer relationships of the user-center to be identified
Net, and mark the abnormal user for including in the M degree customer relationship net.
In the present embodiment, after the M degree association users of user to be identified are determined, in creating and being with user to be identified
The M degree customer relationship nets of the heart in the M degree customer relationship nets, contain the association user within user M degree to be identified, and M degree is used
Each User Status for having identified user in the network of personal connections of family, label respectively go out the quantity for the abnormal user for including.
Preferably, during the number based on abnormal user determines the risk class of user to be identified, each degree association
Weighted value corresponding to user is different, wherein the higher association user of the number of degrees, corresponding weighted value is lower, due to degree
The higher association user of number, then it represents that this has identified that influence of the user to user to be identified is smaller, and the number of degrees are smaller, then it represents that should
It has identified that the relationship between user and user to be identified is closer, has been affected, to which its corresponding weighted value is also larger.It is optional
Ground, calculating the computation model of the risk class of user to be identified can be:
In the present embodiment, NjFor the number of i-th degree of association user.
In embodiments of the present invention, the association user of user to be identified is not only obtained, the association within M degree is also obtained and uses
Family, so as to improve the association range of customer relationship network, to improve the accuracy of risk class.
Fig. 3 shows the specific reality of the determination method S103 for consumer's risk grade that third embodiment of the invention provides a kind of
Existing flow chart.It is shown in Figure 3, relative to embodiment described in Fig. 1, a kind of determination of consumer's risk grade provided in this embodiment
Method S103 includes S1031 and S1032, and specific details are as follows:
Further, the user information includes multiple information projects, and the keyword sequence of the user includes multiple passes
Keyword subsequence, each keyword subsequence correspond to a described information project;
It is described according to each keyword sequence for having identified user in the keyword sequence and customer data base, determine
The user to be identified and each matching degree identified between user, including:
In S1031, it is based on the keyword subsequence, the user to be identified is calculated separately and has identified user with each
Between identical information project similarity.
In the present embodiment, user information includes multiple information projects, and different information projects is for recording the user not
The user information of same type.Illustratively, user information includes:Gender, telephone number, residence, job site, unit one belongs to,
6 information projects of pair bond, record respectively the user each dimension corresponding subscriber data, it is thus determined that equipment can be
Each information project determines corresponding keyword subsequence, to be built corresponding to the user to be identified by keyword subsequence
Keyword sequence.
Optionally, in the present embodiment, in the keyword subsequence for determining different information projects, determine that equipment can inquire
The keyword dictionary of the information project determines that equipment can be that each keyword sequence generates corresponding lists of keywords.
Since certain keywords only appear in certain information projects, and partial information project is not appeared in necessarily, such as
Keyword " Shenzhen " and " Guangzhou ", above-mentioned two keyword belong to the keyword in geographical location, then do not appear in " property necessarily
Not ", in " pair bond " and " telephone number " this three information projects.Therefore, in order to avoid the number of single keyword dictionary
According to measuring excessive and carrying out excessive invalid matching operation when determining keyword sequence, it can be different items of information to create equipment
Mesh determines corresponding keyword dictionary, to improve the formation efficiency of keyword sequence.
In the present embodiment, equipment is determined after the keyword subsequence of each information project is determined, it is each calculating
Before matching degree between user, the similarity of keyword subsequence between each user's corresponding informance project can be calculated.For example,
In the matching degree between calculating user A and user B, the residence of the keyword subsequence and user B of the residence of user A can be calculated
Similarity between the keyword subsequence of residence, then in the keyword subsequence and user's B works for calculating user A work units
Similarity between the keyword subsequence of office, it is similar between all information projects between calculating two users
Degree, then execute the relevant operation of S1032.
In the present embodiment, calculating the concrete mode of the similarity of two keyword subsequences can be:Two passes of statistics
First number of keyword subsequence same keyword, and calculate the ratio between first number and the total element number of keyword sequence
Value, as the similarity between two keyword subsequences.If the certain information project missings of certain customers, identify about the letter
The similarity of breath project is 0.
In S1032, the similarity of each described information project is imported into matching degree transformation model, calculates the matching
Degree;The matching degree transformation model is:
Wherein, Q is the matching degree;BkIt is similar between k-th of described information project of user described in any two
Degree;αkFor the matching weight of k-th of described information project;N is the number of information project.
In the present embodiment, determine that the similarity corresponding to each information project is imported into preset matching degree and turned by equipment
In mold changing type, which defines the different items of information matching weight shared when calculating matching degree, creates equipment
The matching degree between two users can be calculated according to each matching weight and corresponding informance item purpose similarity.
In the present embodiment, matching weight can be downloaded from host computer server obtains, and can also have user according to reality
The demand of border scene adjusts the matching weight of different information projects.Certainly, if user need to ignore partial information project for
With the influence that degree calculates, the matching weight that such information project can be arranged is 0, to not consider the partial information project pair
In the influence of user-association judgement.
In embodiments of the present invention, by the similarity between classified calculating difference information project, and multiple information are based on
Project determines the matching degree between two users, to improve the accuracy of association user identification, further improves risk
The accuracy of grade.
Fig. 4 shows a kind of specific implementation stream of the determination method for consumer's risk grade that fourth embodiment of the invention provides
Cheng Tu.Shown in Figure 4, relative to embodiment described in Fig. 1~Fig. 3, a kind of consumer's risk grade provided in this embodiment is really
Determine further include in method:S401~S404, specific details are as follows:
Further, it is based on the association user described, created with the customer relationship of the user-center to be identified
Network, and before marking the abnormal user for including in the customer relationship network, further include:
In S401, each trading activity record for having identified user is obtained.
In the present embodiment, consumer's risk grade really locking equipment determine identified whether user is abnormal user when,
It can be recorded based on the trading activity of user, the user property of the user is determined.Identify user due to being to record in advance
The user entered, therefore transactional operation can be initiated by the server of financial institution, transactional operation will produce a friendship each time
Easy behavior record determines that equipment can extract the All Activity behavior record of the user based on the user identifier for having identified user.
In S402, each trading activity record is directed respectively into preset regulation coefficient and determines model, is obtained
The regulation coefficient of each trading activity record.
In the present embodiment, a certain each trading activity record for having identified user is imported into preset adjustment respectively is
Cover half type really is counted, the corresponding regulation coefficient of each trading activity record is determined, since executing the relevant operation of S403.Tool
Body, determine that the mode of the corresponding regulation coefficient of each trading activity record can be:It extracts in trading activity record
Transaction feature value, the transaction feature value can be:The information such as trading frequency, transaction amount, transaction attribute, transaction address, need
What is illustrated is that transaction attribute is specifically used for providing that the transaction is positive transaction or reverse side transaction, and front transaction is to meet transaction row
For the transactional operation of specification, such as refund, the as scheduled transactional operations such as delivery interest as scheduled;And reverse side trading activity is then to violate
The transactional operation of trading activity rule, for example, it is overdue do not refund, user's lost contact situations such as.Each transaction feature value is being determined
Afterwards, which is imported into the transformation model of preset credit regulation coefficient, such as can be a preset Hash letter
Number determines the corresponding credit regulation coefficient of trading activity record.
Optionally, before the credit regulation coefficient of each trading activity record of determination, the row of the user can be determined first
To be accustomed to, therefore can be from the historical behavior characteristic value for recording the determining user according to each trading activity.Specifically, which hands over
Easy characteristic value can be the mean value based on the corresponding behavioural characteristic value of each trading activity record.Optionally, the historical trading
Characteristic value can also be the standard deviation based on the corresponding behavioural characteristic value of each trading activity record, for determining the user's
The floating situation of historical trading behavior, in this case, the process for calculating historical trading characteristic value are as follows:
Wherein, HistoryValue is the historical trading characteristic value, and HistoryNum is the number of trading activity record;
HistoryFigiFor the behavioural characteristic value of i-th of trading activity record;For the mean value of behavioural characteristic value.
Also, it can be calculated by following formula when calculating regulation coefficient;
Wherein, Adjust is credit regulation coefficient, and TradeValue is trading activity characteristic value, ApFor predetermined coefficient.
In S403, based on the regulation coefficient and each risk class for having identified user, adjust separately it is each
Identify the risk class of user.
In the present embodiment, equipment is determined after calculating the corresponding credit regulation coefficient of each trading activity record,
Understand the credit regulation coefficient based on the user and identified the original risk class corresponding to user, to having identified user's
Risk class is adjusted, wherein the process of calculating can be on the basis of the initial value of risk class is with each adjustment
Number carries out accumulation operations.It is preferably based on each trading activity record and the difference between current time, determines the transaction row
Initial value for the adjustment weight of record, adjustment weight and risk class based on each regulation coefficient is weighted, tool
The computation model of body is as follows:
Wherein, Cditt(usern) it is n-th of risk class for having identified user after adjustment;T0It is corresponding for current time
Time value;TtradeFor the transaction moment of trading activity record;Tradenum is the number of trading activity record, CditAtradeFor
Regulation coefficient;Cdit0(usern) be n-th of risk class for having identified user initial value.
In S404, if the risk class after adjustment is more than preset risk threshold value, use has been identified described in identification
Family is the abnormal user.
In the present embodiment, equipment is determined after to having identified that the risk class of user is adjusted, it can be by each adjustment
Risk class afterwards is compared with risk threshold value, determines whether the user is abnormal user, if the risk factor is less than or waits
In risk threshold value, then identify that the user is normal users, if conversely, the risk factor of the user is identified more than risk threshold value
The user is abnormal user.
In embodiments of the present invention, identified that the risk class of user is adjusted to each by trading activity record,
So as to adjust the user property of each user in time, to improve the accuracy rate of calculation risk grade.
Fig. 5 shows a kind of specific implementation stream of the determination method for consumer's risk grade that fifth embodiment of the invention provides
Cheng Tu.It is shown in Figure 5, relative to embodiment described in Fig. 1, a kind of determination method of consumer's risk grade provided in this embodiment
Further include:S501~S503, specific details are as follows:
Further, the customer data base includes local data base and cloud database;The local data base
Original state is service response state;The original state of the cloud database is synchrodata reception state;The determination side
Method further includes:
In S501, if receiving the transaction request for having identified user, one is created in the local data base
A trading activity about the transaction request records;The trading activity record is grasped for recording the response of the transaction request
Make.
In the present embodiment, locking equipment is associated with two databases to consumer's risk grade really, and one is stored in local
Local data and one be cloud database for backup information.Above-mentioned two database may be used network attached
Storage (Network Attached Storage, NAS) mode realizes the high availability scheme of Neo4j Community Editions, without purchase
Data backup can also be realized by buying Neo4j commercial versions, improve safety.In this case, local data base and high in the clouds data
Library is applied not only to isochronous transaction behavior record, additionally it is possible to and user synchronizes each customer relationship network for having identified user, due to
Neo4j databases are the databases for storing network topology mechanism, therefore when storing customer relationship network with very high
Storage efficiency.
In the present embodiment, local data base original state can be set as service response state, that is, respond Client-initiated
When transactional operation process, the data that can be recorded, and will be generated during transactional operation by local data base generate transaction
Behavior record.Cloud database can be set as synchrodata reception state, and local data base can at predetermined intervals will be new
The data increment of increasing backs up in the cloud server, to realize the purpose of data backup.
In the present embodiment, if determining that equipment receives has identified Client-initiated transaction request, due to local data base
To respond the higher database of priority, therefore a trading activity about the transaction request can be created in the local database
Record, and the related data responded in the transaction request is recorded in trading activity record.
In S502, if current time meets synchronous trigger condition, by the transaction of the local data library storage
Behavior record is synchronized to the cloud database.
In the present embodiment, synchronous trigger condition includes trigger conditions and time triggered condition.For example, determination is set
It is standby the trading activity record stored and/or customer relationship Network Synchronization to be uploaded into high in the clouds data at predetermined intervals
In library, above-mentioned upload operation can also be executed in preset each timing node;It is of course also possible to when newly-increased data volume reaches
Above-mentioned simultaneously operating is executed when data-quantity threshold.It is not defined one by one herein.
In S503, if detecting the local data base, there are data write-in exceptions, by the cloud server
Working condition is adjusted to the service response state, when receiving transaction request again, to be created by the cloud database
It builds and the trading activity of store transaction request records.
In the present embodiment, if determine equipment detect it is stupid it is doomed dead according to inventory when being written abnormal, then it represents that local data
Library can not continue to complete storage operation, need to carry out abnormality processing, in this case, in order to proceed to respond to Client-initiated
Transaction request determines that the working condition of cloud server can be adjusted to server responsive state by equipment, by local data base
Working condition is adjusted to data synchronous regime, in subsequently received transaction request, be created by cloud database
And the trading activity record of store transaction request, improve the robustness of whole system.
Optionally, after the abnormal conditions of local data base have been repaired, the data that can will be increased newly in cloud database
It is synchronized to local data base, and the working condition of local data base is adjusted to service response state, by the work of cloud server
It is adjusted to data synchronous regime as state.
In embodiments of the present invention, data backup, Neng Gouti are carried out by the way that local data base and cloud database is arranged
The robustness of the determination method of high consumer's risk grade.
It should be understood that the size of the serial number of each step is not meant that the order of the execution order in above-described embodiment, each process
Execution sequence should be determined by its function and internal logic, the implementation process without coping with the embodiment of the present invention constitutes any limit
It is fixed.
Fig. 6 shows a kind of structure diagram for consumer's risk grade locking equipment really that one embodiment of the invention provides, should
The consumer's risk grade each unit that locking equipment includes really is used to execute each step in the corresponding embodiments of Fig. 1.Please specifically it join
Read the associated description in the embodiment corresponding to Fig. 1 and Fig. 1.For convenience of description, portion related to the present embodiment is illustrated only
Point.
Referring to Fig. 6, locking equipment includes the consumer's risk grade really:
User information receiving unit 61, the user information for receiving user to be identified;
Keyword sequence determination unit 62 is used for each key in the user information and preset keyword dictionary
Word is matched, and determines the keyword sequence of the user to be identified;
Association user recognition unit 63, for having identified use according to each in the keyword sequence and customer data base
The keyword sequence at family determines the user to be identified and each matching degree identified between user, and chooses the matching
Degree is more than the identification user of preset matching degree threshold value, the association user as the user to be identified;
Customer relationship network creation unit 64, for being based on the association user, during establishment is with the user to be identified
The customer relationship network of the heart, and mark the abnormal user for including in the customer relationship network;The abnormal user is specially wind
Dangerous grade is more than the user of preset risk threshold value;
Risk class determination unit 65, described according to the number of abnormal user in the customer relationship network, determining
The risk class of user to be identified.
Optionally, customer relationship network creation unit 64 includes:
More degree association user recognition units, for obtaining each association user for having identified user, described in determination
The M degree association users of user to be identified;Wherein, in the associated path between the M degree association user and the user to be identified
There are M-1 user nodes;M is the positive integer more than or equal to 1;
More degree user network creating units, are used for according to the M degree association user, during establishment is with the user to be identified
The M degree customer relationship nets of the heart, and mark the abnormal user for including in the M degree customer relationship net.
Optionally, the user information includes multiple information projects, and the keyword sequence of the user includes multiple keys
Lexon sequence, each keyword subsequence correspond to a described information project;
Keyword sequence determination unit 62 includes:
Similarity calculated, for be based on the keyword subsequence, calculate separately the user to be identified with it is each
The similarity of identical information project between user is identified;
Matching degree computing unit is counted for the similarity of each described information project to be imported into matching degree transformation model
Calculate the matching degree;The matching degree transformation model is:
Wherein, Q is the matching degree;BkIt is similar between k-th of described information project of user described in any two
Degree;αkFor the matching weight of k-th of described information project;N is the number of information project.
Optionally, locking equipment further includes the consumer's risk grade really:
Trading activity records acquiring unit, for obtaining each trading activity record for having identified user;
Regulation coefficient determination unit, it is true for each trading activity record to be directed respectively into preset regulation coefficient
Cover half type obtains the regulation coefficient of each trading activity record;
Risk class adjustment unit, for based on the regulation coefficient and each risk class for having identified user, dividing
Each risk class for having identified user is not adjusted;
Abnormal user recognition unit identifies if being more than preset risk threshold value for the risk class after adjusting
It is described to have identified that user is the abnormal user.
Optionally, the customer data base includes local data base and cloud database;At the beginning of the local data base
Beginning state is service response state;The original state of the cloud database is synchrodata reception state;The consumer's risk
Really locking equipment further includes grade:
Transaction request receiving unit, if for receiving the transaction request for having identified user, in the local number
It is recorded according to a trading activity about the transaction request is created in library;The trading activity record is for recording the transaction
The response of request operates;
Data synchronisation unit, if meeting synchronous trigger condition for current time, by the local data library storage
The trading activity recording synchronism is to the cloud database;
Exception response unit, if there are data write-in exceptions for detecting the local data base, by the high in the clouds
The working condition of server is adjusted to the service response state, when receiving transaction request again, to pass through the high in the clouds
Database creates and the trading activity record of store transaction request.
Therefore, locking equipment can equally not depend on historical trading note to consumer's risk grade provided in an embodiment of the present invention really
It records to determine the risk class of user, and can be based on the number for the abnormal user for including in the association user of the user, sentence
Whether the disconnected user belongs to the member in abnormal user group, and the difference based on number obtains the risk class of the user, to
Risk class also can be determined quickly for newly added user, effective management and control is carried out to transaction request, reduces capital loss
Risk.
Fig. 7 is a kind of schematic diagram for consumer's risk grade locking equipment really that another embodiment of the present invention provides.Such as Fig. 7 institutes
Show, locking equipment 7 includes the consumer's risk grade of the embodiment really:Processor 70, memory 71 and it is stored in the storage
In device 71 and the computer program 72 that can be run on the processor 70, for example, consumer's risk grade determination program.It is described
Processor 70 realizes the step in the determination embodiment of the method for above-mentioned each consumer's risk grade when executing the computer program 72
Such as S101 shown in FIG. 1 to S105 suddenly,.Alternatively, the processor 70 realized when executing the computer program 72 it is above-mentioned each
The function of each unit in device embodiment, such as the function of module 61 to 65 shown in Fig. 6.
Illustratively, the computer program 72 can be divided into one or more units, one or more of
Unit is stored in the memory 71, and is executed by the processor 70, to complete the present invention.One or more of lists
Member can complete the series of computation machine program instruction section of specific function, and the instruction segment is for describing the computer journey
Implementation procedure of the sequence 72 in the consumer's risk grade really locking equipment 7.For example, the computer program 72 can be divided
At user information receiving unit, keyword sequence determination unit, association user recognition unit, customer relationship network creation unit with
And risk class determination unit, each unit concrete function are as described above.
Really locking equipment 7 can be desktop PC, notebook, palm PC and high in the clouds clothes to the consumer's risk grade
The computing devices such as business device.Really locking equipment may include the consumer's risk grade, but be not limited only to, processor 70, memory 71.
It will be understood by those skilled in the art that Fig. 7 is only the example of consumer's risk grade locking equipment 7 really, do not constitute to user
The restriction of risk class locking equipment 7 really, may include than illustrating more or fewer components, or the certain components of combination, or
The different component of person, such as locking equipment can also be set the consumer's risk grade including input-output equipment, network insertion really
Standby, bus etc..
Alleged processor 70 can be central processing unit (Central Processing Unit, CPU), can also be
Other general processors, digital signal processor (Digital Signal Processor, DSP), application-specific integrated circuit
(Application Specific Integrated Circuit, ASIC), ready-made programmable gate array (Field-
Programmable Gate Array, FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components etc..General processor can be microprocessor or the processor can also be any conventional processor
Deng.
The memory 71 can be the internal storage unit of consumer's risk grade locking equipment 7 really, such as user
The hard disk or memory of risk class locking equipment 7 really.The memory 71 can also be that the determination of the consumer's risk grade is set
Standby 7 External memory equipment, such as plug-in type hard disk that the consumer's risk grade is equipped on locking equipment 7 really, intelligent storage
Block (Smart Media Card, SMC), secure digital (Secure Digital, SD) card, flash card (Flash Card) etc..
Further, the memory 71 can also both include the consumer's risk grade really the internal storage unit of locking equipment 7 or
Including External memory equipment.The memory 71 is used to store the determination of the computer program and the consumer's risk grade
Other programs needed for equipment and data.The memory 71, which can be also used for temporarily storing, have been exported or will export
Data.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, it can also
It is that each unit physically exists alone, it can also be during two or more units be integrated in one unit.Above-mentioned integrated list
The form that hardware had both may be used in member is realized, can also be realized in the form of SFU software functional unit.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations;Although with reference to aforementioned reality
Applying example, invention is explained in detail, it will be understood by those of ordinary skill in the art that:It still can be to aforementioned each
Technical solution recorded in embodiment is modified or equivalent replacement of some of the technical features;And these are changed
Or replace, the spirit and scope for various embodiments of the present invention technical solution that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (10)
1. a kind of determination method of consumer's risk grade, which is characterized in that including:
Receive the user information of user to be identified;
The user information is matched with each keyword in preset keyword dictionary, determines the user to be identified
Keyword sequence;
According to each keyword sequence for having identified user in the keyword sequence and customer data base, wait knowing described in determination
Other user and each matching degree identified between user, and choose the identification that the matching degree is more than preset matching degree threshold value
User, the association user as the user to be identified;
Based on the association user, create with the customer relationship network of the user-center to be identified, and mark the user
The abnormal user for including in relational network;The abnormal user is specially the user that risk class is more than preset risk threshold value;
According to the number of abnormal user in the customer relationship network, the risk class of the user to be identified is determined.
2. determining method according to claim 1, which is characterized in that it is described to be based on the association user, it creates with described
The customer relationship network of user-center to be identified, and the abnormal user for including in the customer relationship network is marked, including:
Each association user for having identified user is obtained, with the M degree association users of the determination user to be identified;Wherein,
There are M-1 user nodes in associated path between the M degree association user and the user to be identified;M is to be more than or wait
In 1 positive integer;
According to the M degree association user, create with the M degree customer relationship nets of the user-center to be identified, and mark the M
The abnormal user for including in degree customer relationship net.
3. determining method according to claim 1, which is characterized in that the user information includes multiple information projects, institute
The keyword sequence for stating user includes multiple keyword subsequences, and each keyword subsequence corresponds to a described information item
Mesh;
It is described according to each keyword sequence for having identified user in the keyword sequence and customer data base, determine described in
User to be identified and each matching degree identified between user, including:
Based on the keyword subsequence, the user to be identified and each identical information between having identified user are calculated separately
The similarity of project;
The similarity of each described information project is imported into matching degree transformation model, calculates the matching degree;The matching degree
Transformation model is:
Wherein, Q is the matching degree;BkFor the similarity between k-th of described information project of user described in any two;αk
For the matching weight of k-th of described information project;N is the number of information project.
4. determining method according to claim 1-3 any one of them, which is characterized in that further include:
Obtain each trading activity record for having identified user;
Each trading activity record is directed respectively into preset regulation coefficient and determines model, obtains each transaction row
For the regulation coefficient of record;
Based on the regulation coefficient and each risk class for having identified user, each risk for having identified user is adjusted separately
Grade;
If the risk class after adjustment is more than preset risk threshold value, identify that user is that the exception is used described in identification
Family.
5. determining method according to claim 1, which is characterized in that the customer data base include local data base and
Cloud database;The original state of the local data base is service response state;The original state of the cloud database is
Synchrodata reception state;The determining method further includes:
If receiving the transaction request for having identified user, one is created in the local data base about the transaction
The trading activity of request records;The trading activity record is operated for recording the response of the transaction request;
If current time meets synchronous trigger condition, extremely by the trading activity recording synchronism of the local data library storage
The cloud database;
If detecting, there are data write-in exception, the working condition of the cloud server is adjusted to for the local data base
The service response state, when receiving transaction request again, to be created by the cloud database and store transaction is asked
The trading activity record asked.
6. a kind of consumer's risk grade locking equipment really, which is characterized in that locking equipment includes depositing to the consumer's risk grade really
Reservoir, processor and it is stored in the computer program that can be run in the memory and on the processor, the processing
Device realizes following steps when executing the computer program:
Receive the user information of user to be identified;
The user information is matched with each keyword in preset keyword dictionary, determines the user to be identified
Keyword sequence;
According to each keyword sequence for having identified user in the keyword sequence and customer data base, wait knowing described in determination
Other user and each matching degree identified between user, and choose the identification that the matching degree is more than preset matching degree threshold value
User, the association user as the user to be identified;
Based on the association user, create with the customer relationship network of the user-center to be identified, and mark the user
The abnormal user for including in relational network;The abnormal user is specially the user that risk class is more than preset risk threshold value;
According to the number of abnormal user in the customer relationship network, the risk class of the user to be identified is determined.
7. determining equipment according to claim 6, which is characterized in that it is described to be based on the association user, it creates with described
The customer relationship network of user-center to be identified, and the abnormal user for including in the customer relationship network is marked, including:
Each association user for having identified user is obtained, with the M degree association users of the determination user to be identified;Wherein,
There are M-1 user nodes in associated path between the M degree association user and the user to be identified;M is to be more than or wait
In 1 positive integer;
According to the M degree association user, create with the M degree customer relationship nets of the user-center to be identified, and mark the M
The abnormal user for including in degree customer relationship net.
8. determining equipment according to claim 6, which is characterized in that the user information includes multiple information projects, institute
The keyword sequence for stating user includes multiple keyword subsequences, and each keyword subsequence corresponds to a described information item
Mesh;
It is described according to each keyword sequence for having identified user in the keyword sequence and customer data base, determine described in
User to be identified and each matching degree identified between user, including:
Based on the keyword subsequence, the user to be identified and each identical information between having identified user are calculated separately
The similarity of project;
The similarity of each described information project is imported into matching degree transformation model, calculates the matching degree;The matching degree
Transformation model is:
Wherein, Q is the matching degree;BkFor the similarity between k-th of described information project of user described in any two;αk
For the matching weight of k-th of described information project;N is the number of information project.
It is arranged 9. being determined according to claim 6-8 any one of them, which is characterized in that the processor executes the computer
Following steps are also realized when program:
Obtain each trading activity record for having identified user;
Each trading activity record is directed respectively into preset regulation coefficient and determines model, obtains each transaction row
For the regulation coefficient of record;
Based on the regulation coefficient and each risk class for having identified user, each risk for having identified user is adjusted separately
Grade;
If the risk class after adjustment is more than preset risk threshold value, identify that user is that the exception is used described in identification
Family.
10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, feature to exist
In when the computer program is executed by processor the step of any one of such as claim 1 to 5 of realization the method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810436177.4A CN108564467A (en) | 2018-05-09 | 2018-05-09 | A kind of determination method and apparatus of consumer's risk grade |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810436177.4A CN108564467A (en) | 2018-05-09 | 2018-05-09 | A kind of determination method and apparatus of consumer's risk grade |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108564467A true CN108564467A (en) | 2018-09-21 |
Family
ID=63538095
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810436177.4A Pending CN108564467A (en) | 2018-05-09 | 2018-05-09 | A kind of determination method and apparatus of consumer's risk grade |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108564467A (en) |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109461070A (en) * | 2018-10-25 | 2019-03-12 | 深圳壹账通智能科技有限公司 | A kind of risk measures and procedures for the examination and approval, device, storage medium and server |
CN109471934A (en) * | 2018-10-09 | 2019-03-15 | 杭州安恒信息技术股份有限公司 | The financial risks clue method of excavation Internet-based |
CN109615256A (en) * | 2018-12-17 | 2019-04-12 | 武汉烽火信息集成技术有限公司 | Wisdom garden security protection risk control method, storage medium, electronic equipment and system |
CN109636575A (en) * | 2018-10-25 | 2019-04-16 | 平安科技(深圳)有限公司 | Terminal risk checking method, device, equipment and readable storage medium storing program for executing |
CN109636574A (en) * | 2018-10-25 | 2019-04-16 | 平安科技(深圳)有限公司 | Risk checking method, device, equipment and the storage medium of credit information |
CN109670934A (en) * | 2018-09-26 | 2019-04-23 | 深圳壹账通智能科技有限公司 | Personal identification method, equipment, storage medium and device based on user behavior |
CN109840778A (en) * | 2018-12-21 | 2019-06-04 | 上海拍拍贷金融信息服务有限公司 | The recognition methods of fraudulent user and device, readable storage medium storing program for executing |
CN110033170A (en) * | 2019-03-14 | 2019-07-19 | 阿里巴巴集团控股有限公司 | Identify the method and device of risk businessman |
CN110097258A (en) * | 2019-04-12 | 2019-08-06 | 平安普惠企业管理有限公司 | A kind of customer relationship network creating method, device and computer readable storage medium |
CN110197397A (en) * | 2019-05-15 | 2019-09-03 | 无线生活(北京)信息技术有限公司 | The partitioning method and device of node level |
CN111382284A (en) * | 2020-03-09 | 2020-07-07 | 中国建设银行股份有限公司 | Neo4j graphic database system, Neo4j graphic database system access method and device |
CN111861281A (en) * | 2020-08-05 | 2020-10-30 | 中国银行股份有限公司 | Knowledge graph-based risk employee discovery method and device |
CN113762684A (en) * | 2020-12-14 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | New user risk assessment method and device, electronic equipment and medium |
CN113807941A (en) * | 2020-12-29 | 2021-12-17 | 京东科技控股股份有限公司 | Risk detection method and device, computer equipment and storage medium |
CN117669594A (en) * | 2024-02-02 | 2024-03-08 | 智器云南京信息科技有限公司 | Big data relation network analysis method and system for abnormal information |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
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 |
CN106921504A (en) * | 2015-12-24 | 2017-07-04 | 阿里巴巴集团控股有限公司 | A kind of method and apparatus of the associated path for determining different user |
CN107194623A (en) * | 2017-07-20 | 2017-09-22 | 深圳市分期乐网络科技有限公司 | A kind of discovery method and device of clique's fraud |
CN107705036A (en) * | 2017-10-27 | 2018-02-16 | 杭州呯嘭智能技术有限公司 | Dynamic credit estimation method and system based on multi-dimensional data |
-
2018
- 2018-05-09 CN CN201810436177.4A patent/CN108564467A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
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 |
CN106921504A (en) * | 2015-12-24 | 2017-07-04 | 阿里巴巴集团控股有限公司 | A kind of method and apparatus of the associated path for determining different user |
CN107194623A (en) * | 2017-07-20 | 2017-09-22 | 深圳市分期乐网络科技有限公司 | A kind of discovery method and device of clique's fraud |
CN107705036A (en) * | 2017-10-27 | 2018-02-16 | 杭州呯嘭智能技术有限公司 | Dynamic credit estimation method and system based on multi-dimensional data |
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109670934A (en) * | 2018-09-26 | 2019-04-23 | 深圳壹账通智能科技有限公司 | Personal identification method, equipment, storage medium and device based on user behavior |
CN109471934A (en) * | 2018-10-09 | 2019-03-15 | 杭州安恒信息技术股份有限公司 | The financial risks clue method of excavation Internet-based |
CN109471934B (en) * | 2018-10-09 | 2021-08-17 | 杭州安恒信息技术股份有限公司 | Financial risk clue mining method based on Internet |
CN109636574A (en) * | 2018-10-25 | 2019-04-16 | 平安科技(深圳)有限公司 | Risk checking method, device, equipment and the storage medium of credit information |
CN109636575A (en) * | 2018-10-25 | 2019-04-16 | 平安科技(深圳)有限公司 | Terminal risk checking method, device, equipment and readable storage medium storing program for executing |
CN109461070A (en) * | 2018-10-25 | 2019-03-12 | 深圳壹账通智能科技有限公司 | A kind of risk measures and procedures for the examination and approval, device, storage medium and server |
CN109615256A (en) * | 2018-12-17 | 2019-04-12 | 武汉烽火信息集成技术有限公司 | Wisdom garden security protection risk control method, storage medium, electronic equipment and system |
CN109840778A (en) * | 2018-12-21 | 2019-06-04 | 上海拍拍贷金融信息服务有限公司 | The recognition methods of fraudulent user and device, readable storage medium storing program for executing |
CN110033170A (en) * | 2019-03-14 | 2019-07-19 | 阿里巴巴集团控股有限公司 | Identify the method and device of risk businessman |
US11379845B2 (en) | 2019-03-14 | 2022-07-05 | Advanced New Technologies Co., Ltd. | Method and device for identifying a risk merchant |
CN110097258A (en) * | 2019-04-12 | 2019-08-06 | 平安普惠企业管理有限公司 | A kind of customer relationship network creating method, device and computer readable storage medium |
CN110197397A (en) * | 2019-05-15 | 2019-09-03 | 无线生活(北京)信息技术有限公司 | The partitioning method and device of node level |
CN111382284A (en) * | 2020-03-09 | 2020-07-07 | 中国建设银行股份有限公司 | Neo4j graphic database system, Neo4j graphic database system access method and device |
CN111861281A (en) * | 2020-08-05 | 2020-10-30 | 中国银行股份有限公司 | Knowledge graph-based risk employee discovery method and device |
CN113762684A (en) * | 2020-12-14 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | New user risk assessment method and device, electronic equipment and medium |
CN113807941A (en) * | 2020-12-29 | 2021-12-17 | 京东科技控股股份有限公司 | Risk detection method and device, computer equipment and storage medium |
CN113807941B (en) * | 2020-12-29 | 2024-03-05 | 京东科技控股股份有限公司 | Risk detection method, risk detection device, computer equipment and storage medium |
CN117669594A (en) * | 2024-02-02 | 2024-03-08 | 智器云南京信息科技有限公司 | Big data relation network analysis method and system for abnormal information |
CN117669594B (en) * | 2024-02-02 | 2024-04-16 | 智器云南京信息科技有限公司 | Big data relation network analysis method and system for abnormal information |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108564467A (en) | A kind of determination method and apparatus of consumer's risk grade | |
CN108615119A (en) | A kind of recognition methods of abnormal user and equipment | |
Ludwig et al. | Ibm federated learning: an enterprise framework white paper v0. 1 | |
CN109146679B (en) | Intelligent contract calling method and device based on block chain and electronic equipment | |
CN109003078A (en) | Intelligent contract call method and device, electronic equipment based on block chain | |
CN108711085A (en) | A kind of response method and its equipment of transaction request | |
WO2018142764A1 (en) | Learned model generating method, learned model generating device, and learned model use device | |
CN109684330A (en) | User's portrait base construction method, device, computer equipment and storage medium | |
CN109493024A (en) | Digital asset trustship method, equipment and storage medium | |
CN110162695A (en) | A kind of method and apparatus of information push | |
CN109858007A (en) | Semantic analysis answering method and device, computer equipment and storage medium | |
CN109242612A (en) | A kind of method and apparatus of Products Show | |
CN107302573A (en) | A kind of information-pushing method, device, electronic equipment and storage medium | |
US20080087729A1 (en) | Federated Voting With Criteria-Based Discrimination | |
CN108229367A (en) | A kind of face identification method and device | |
CN101924786A (en) | Dynamic content preference and behavior sharing between computing devices | |
US9275125B1 (en) | System for organizing data from a plurality of users to create individual user profiles | |
CN110263311A (en) | A kind of generation method and equipment of Webpage | |
CN107729466A (en) | Construction method, device and the electronic equipment of relational network | |
CN108520471A (en) | It is overlapped community discovery method, device, equipment and storage medium | |
CN110266805A (en) | Information-pushing method, device, electronic equipment and readable medium | |
CN107845053A (en) | Personnel information management method, equipment, readable storage medium storing program for executing and electronic equipment | |
CN110428311A (en) | Bidding information recommendation method and Related product | |
EP4264472A1 (en) | Systems and methods for hard deletion of data across systems | |
CN107656959A (en) | A kind of message leaving method, device and message 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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180921 |