CN105894372A - Method and device for predicting group credit - Google Patents
Method and device for predicting group credit Download PDFInfo
- Publication number
- CN105894372A CN105894372A CN201610414335.7A CN201610414335A CN105894372A CN 105894372 A CN105894372 A CN 105894372A CN 201610414335 A CN201610414335 A CN 201610414335A CN 105894372 A CN105894372 A CN 105894372A
- Authority
- CN
- China
- Prior art keywords
- colony
- population
- default probability
- data
- individual
- 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
Abstract
The invention relates to a method and device for predicting group credit. The method comprises the following steps: acquiring selected data, dividing the selected data into first class group data containing individuals of known default situations and second class group data containing individuals of unknown default situations; selecting a corresponding predefined group default probability; gaining the group default probability of the first class group data according to the selected predefined group default probability to calculate a first variable; obtaining a group characteristic variable of the first class group data; training according to the first variable and the group characteristic variable of the first class group data to obtain a group default prediction model; predicting a group default for the second class group data according to the group default prediction model to obtain the group default probability of the second class group data; transforming the group default probability of the second class group data into corresponding group credit scores. With the adoption of the method and the device for predicting the group credit, the group credit is evaluated.
Description
Technical field
The present invention relates to computer application field, particularly relate to a kind of method and apparatus predicting colony's credit.
Background technology
Along with developing rapidly of computer technology and network technology, the exchange activity between user is more and more frequent,
User creates the data of magnanimity in communication process, and the information represented by these data exists the true and false, when one
User transmits fake information to another user, may cause damage this another user.To this end, generally
The credit of individual subscriber can be estimated.And some service needed colony completes jointly, whole colony is believed
It is estimated with needs, to reduce the risk of colony's promise breaking.But, in tradition credit evaluation mode not
Colony's credit is estimated.
Summary of the invention
Based on this, it is necessary to problem colony's credit cannot being estimated for tradition credit evaluation mode,
There is provided a kind of method predicting colony's credit, it is achieved the assessment to colony's credit.
Additionally, there is a need to provide a kind of device predicting colony's credit, it is achieved the assessment to colony's credit.
A kind of method predicting colony's credit, including:
Obtain selected data, described selected data be divided into first kind population data and Equations of The Second Kind population data,
Described first kind population data comprises the individuality of known violation of agreement, described Equations of The Second Kind population data comprises
The individuality of unknown violation of agreement；
Business demand according to selected data selects corresponding predefined colony Default Probability；
The colony's promise breaking asking for described first kind population data according to the predefined colony Default Probability chosen is general
Rate, calculates the predefined colony promise breaking chosen general according to colony's Default Probability of described first kind population data
First variable of colony's promise breaking that rate is corresponding；
Obtain the population characteristic variable of described first kind population data；
Population characteristic variable according to described first variable and described first kind population data is trained obtaining group
Body violation correction model；
According to described colony violation correction model, described Equations of The Second Kind population data is carried out colony's violation correction to obtain
Colony's Default Probability of described Equations of The Second Kind colony；
Colony's Default Probability of described Equations of The Second Kind colony is converted into colony's credit score of correspondence.
A kind of device predicting colony's credit, including:
First acquisition module, is used for obtaining selected data, and described selected data is divided into first kind population data
With Equations of The Second Kind population data, described first kind population data comprises the individuality of known violation of agreement, described
Two types of populations data comprise the individuality of unknown violation of agreement；
Choose module, select corresponding predefined colony promise breaking general for the business demand according to selected data
Rate；
Ask for module, for asking for described first kind colony number according to the predefined colony Default Probability chosen
According to colony's Default Probability, calculate choose predetermined according to the colony Default Probability of described first kind population data
First variable of colony's promise breaking that adopted colony's Default Probability is corresponding；
Second acquisition module, for obtaining the population characteristic variable of described first kind population data；
Model building module, for according to described first variable and the population characteristic of described first kind population data
Variable is trained obtaining colony's violation correction model；
Prediction module, for carrying out group according to described colony violation correction model to described Equations of The Second Kind population data
Body violation correction obtains colony's Default Probability of described Equations of The Second Kind colony；
Conversion module, for being converted into colony's credit of correspondence by colony's Default Probability of described Equations of The Second Kind colony
Score value.
The method and apparatus of above-mentioned prediction colony credit, comprises known violation of agreement by being divided into by population data
The Equations of The Second Kind population data that individual first kind population data is individual with comprising unknown violation of agreement, according to first
Types of populations data calculate the first variable that predefined colony Default Probability is corresponding, obtain the group of first kind colony
Body characteristics variable, is trained obtaining group according to the population characteristic variable of the first variable and first kind population data
Body violation correction model, according to colony's Default Probability of colony's violation correction model prediction Equations of The Second Kind population data,
Colony's Default Probability of Equations of The Second Kind colony is converted into colony's credit score of correspondence, by colony's credit score
The credit situation of colony can be reflected, it is achieved that the assessment to colony's credit, and can be to comprising unknown promise breaking feelings
Individual colony's credit of condition is predicted.
Accompanying drawing explanation
Fig. 1 is the internal structure schematic diagram of electronic equipment in an embodiment；
Fig. 2 is the flow chart of the method predicting colony's credit in an embodiment；
Fig. 3 is the flow chart obtaining population characteristic variable in an embodiment；
Fig. 4 is the structured flowchart of the device predicting colony's credit in an embodiment；
Fig. 5 is the structured flowchart of the device predicting colony's credit in another embodiment；
Fig. 6 is the structured flowchart of the device predicting colony's credit in another embodiment；
Fig. 7 is the structured flowchart of the device predicting colony's credit in another embodiment.
Detailed description of the invention
In order to make the purpose of the present invention, technical scheme and advantage clearer, below in conjunction with accompanying drawing and reality
Execute example, the present invention is further elaborated.Only should be appreciated that specific embodiment described herein
Only in order to explain the present invention, it is not intended to limit the present invention.
Fig. 1 is the internal structure schematic diagram of electronic equipment in an embodiment.As it is shown in figure 1, this electronic equipment
Including the processor connected by system bus, non-volatile memory medium, built-in storage, network interface,
Display screen and input equipment.Wherein, the non-volatile memory medium storage of electronic equipment has operating system, also
Including a kind of device predicting colony's credit, the device of this prediction colony credit is used for realizing a kind of prediction colony
The method of credit.This processor is used for providing calculating and control ability, supports the operation of whole electronic equipment.
Built-in storage in electronic equipment is that the operation of the device of the prediction colony credit in non-volatile memory medium carries
For environment, can store computer-readable instruction in this built-in storage, this computer-readable instruction is by described place
When reason device performs, described processor can be made to perform a kind of method predicting colony's credit.Network interface is used for
Network service etc. is carried out with other equipment.The display screen of electronic equipment can be LCDs or electronic ink
Water display screens etc., input equipment can be the touch layer covered on display screen, it is also possible to is electronic equipment casing
The button of upper setting, trace ball or Trackpad, it is also possible to be external keyboard, Trackpad or mouse etc..Should
Electronic equipment can be mobile phone, panel computer or personal digital assistant or Wearable etc..Electronic equipment
Can be also server or high in the clouds etc..Server can form with independent server or multiple server
Server cluster realizes.It will be understood by those skilled in the art that the structure shown in Fig. 1, be only and this
The block diagram of the part-structure that application scheme is relevant, is not intended that the electronics being applied thereon the application scheme
The restriction of equipment, concrete electronic equipment can include than shown in figure more or less of parts, or group
Close some parts, or there is different parts layouts.
Fig. 2 is the flow chart of the method predicting colony's credit in an embodiment.As in figure 2 it is shown, it is a kind of pre-
The method surveying colony's credit, runs on the electronic equipment in Fig. 1, including:
Step 202, obtains selected data, this selected data is divided into first kind population data and Equations of The Second Kind colony
Data, comprise the individuality of known violation of agreement in this first kind population data, wrap in this Equations of The Second Kind population data
Individuality containing unknown violation of agreement.
In the present embodiment, selected data can be the data obtained from social networks, or from the network number of magnanimity
Data selected according to.The individuality of known violation of agreement refers to whether this individuality known breaks a contract, i.e. promise breaking or
Do not break a contract.The individuality of unknown violation of agreement refers to that individual violation of agreement is unclear or indefinite etc..
The individual violation of agreement comprised in first kind population data is the most known.Equations of The Second Kind population data comprises
Individual violation of agreement can be partly it is known that part be unknown or all unknown.
Before step 202, the method for above-mentioned prediction colony credit also includes: disobey according to predefined colony
The about corresponding threshold value of probabilistic settings, and build according to each predefined colony Default Probability and corresponding threshold value
First variable of corresponding colony's promise breaking, and the Default Probability of colony is more than or equal to this threshold value, then this colony
For promise breaking colony.
In the present embodiment, the Default Probability of this predefined colony include first predefine colony's Default Probability,
Second predefine colony's Default Probability, the 3rd predefine colony's Default Probability, the 4th to predefine colony's promise breaking general
Rate, the 5th predefine one or more in colony's Default Probability.
Predefined colony Default Probability includes (a1) to (a5):
(a1) the first ratio of the number of individuals that violations occur in colony and individual in population sum is obtained,
This first ratio is predefined colony's Default Probability as first.
In the present embodiment, with reference to personal credit promise breaking definition, it is assumed that the promise breaking of each individuality is independent event,
The then user of certain proportion P in a colony having individuality, had N*P individuality to go out in nearest D days
Existing violations, then this colony's Default Probability is P, is designated as first and predefines colony's Default Probability.Colony refers to
The corporations being made up of one or more individualities or group etc..Individuality refers to each independent person in colony.Individual
Single natural person or single legal person (as being enterprise, businessman etc.) etc. can be referred to.
(a2) obtaining the factor of influence of population size, factor of influence and first according to this population size make a reservation for
Colony's Default Probability of justice obtains second and predefines colony's Default Probability.
In the present embodiment, the venture influence after breaking a contract because of different scales colony is different, and larger colony disobeys
The actual influence caused the most afterwards is much larger than the impact of small population size.Because Default Probability is in [0,1] space, group
Body scale is as a promise breaking factor of influence, and colony's Default Probability of small scale suitably reduces.Advise according to colony
The factor of influence of mould and the first predefined colony Default Probability P obtain second and predefine colony's Default Probability P'.
P'=(1+ αN)-1P formula (1)
In formula (1), α is less than 1, and α is the factor of influence of population size, and N is individual in population sum.
(a3) obtain the first promise breaking function and the property value of forward attribute of individual in population, obtain each each and every one
The product of the property value of the forward attribute of body and the first promise breaking function and forward attribute with individual in population
Property value and seek ratio the second ratio being worth to, then the sum of this second ratio with this individual in population is asked
Ratio is worth to the 3rd ratio, and as the 3rd, the 3rd ratio is predefined colony's Default Probability, or, obtain
The property value of the negative sense attribute of the first promise breaking function and individual in population, obtains the negative sense attribute of each individuality
The product of the inverse of property value and the first promise breaking function and the property value of negative sense attribute with individual in population
Reciprocal and seek ratio the second ratio being worth to, then the sum of the second ratio Yu this individual in population is sought ratio
Obtain the 3rd ratio, the 3rd ratio is predefined colony's Default Probability as the 3rd, wherein, the first promise breaking
If individual promise breaking in function, then the value of the first promise breaking function is 1, if individuality is not broken a contract, then and the first promise breaking function
Value be 0.
In the present embodiment, individual forward attribute refers to become positively related attribute with risk of loss.In this colony
Individual forward attribute is power of influence, the accrediting amount, individual credit score or technorati authority.Individual negative sense belongs to
Property refers to become the attribute of negative correlation with risk of loss.The negative sense attribute of this individual in population is individual income data
Or stability data.The mark that individual credit score obtains after referring to be estimated individual credit.Technorati authority
Refer to individual prestige degree in colony.Stability data refers to the stability data of individual credit.
As a example by power of influence in individual forward attribute, individual power of influence I value in colony, individual shadow
The power of sound is the biggest, and I value is the biggest.The individuality that power of influence is big is in key position on Information Communication, right after its promise breaking
The impact of colony's promise breaking also can be big.
Obtaining I value can use PageRank to calculate the rank of each individuality according to the graph of a relation of individual in population
(sequence) value, or calculate this I value by other influences power algorithm.
Obtain I value to include (b1) to (b3):
(b1) data representing that groups of individuals belongs to relation are obtained, true according to the data that this expression groups of individuals belongs to relation
Determine the colony belonging to individuality.
Specifically, representing that groups of individuals belongs to the data of relation can be group in social networking application, interest corporations, corporations
The circle data etc. found.If can confirm that the colony belonging to individuality in expression groups of individuals belongs to the data of relation,
Then this individuality is directly belonged to affiliated colony.If individuality does not has an affiliated colony, then according to individual good
Friend's relational network carries out colony's division, makees according to the colony that good friend in individual friend relation network is at most affiliated
For the colony belonging to this individuality, or according to the colony belonging to individual hail fellow as belonging to this individuality
Colony etc..
(b2) according to individual interaction data, the cohesion of personal relationship is calculated.
Specifically, individual ownership good friend is sorted according to mutual frequency, after normalizing to [0,1] space, value
Size represents the cohesion between individuality and good friend.
(b3) according to the topological structure tectonic network graph of a relation between individual in population, by cohesion as net
The weight on limit in network graph of a relation, according to cyberrelationship figure calculate each individuality affect force value.
Specifically, the topological structure between individual in population refers to that the mutual relation between individual in population is formed
Topological structure.PageRank or LeaderRank, H index etc. is used to calculate node according to cyberrelationship figure
Importance and the algorithm of power of influence, calculate each individuality affects force value or ranking value.Page Rank algorithm and
LeaderRank algorithm is two kinds carries out the algorithm of importance ranking to Node Contraction in Complex Networks.
Power of influence I value being after max-min normalizes to [0,1] interval is I', calculates Default Probability P of colony "
For:
In formula (2), f (k) is the first promise breaking function, and k represents that kth is individual, and N represents individual in population
Sum.
If forward attribute is the accrediting amount, individual credit score or technorati authority, directly by formula (2)
Power of influence I value replaces with the value that the accrediting amount, individual credit score or technorati authority are corresponding.If negative sense attribute is individual
Power of influence I value in body income data or stability data, the most just formula (2) replaces with individual income number
According to or the inverse of value corresponding to stability data.
(a4) this second is predefined colony's Default Probability predefines colony's Default Probability with the 3rd and be multiplied and obtain
4th predefines colony's Default Probability.
Specifically, considering the Default Probability after the impact of individuality and corporations' scale is the 4th to predefine colony
Default Probability P " '.
(a5) first predefine colony's Default Probability to predefine in colony's Default Probability individual promise breaking to the 4th general
Rate all replaces with the behavior that individual promise breaking occurs and obtains four kind the 5th and predefine Default Probability.
The behavior that individual promise breaking occurs can be promise breaking numerical value or promise breaking issue etc..5th predefines Default Probability is
A series of definition, i.e. first predefine colony's Default Probability to the 4th predefine in colony's Default Probability
Body Default Probability replaces with the behavior that individual promise breaking occurs, such as promise breaking amount, promise breaking issue etc., computational methods
Predefine colony's Default Probability with first and predefine colony's Default Probability to the 4th.
The first promise breaking letter that colony's Default Probability or the 4th predefines in colony's Default Probability is predefined for the 3rd
Number replaces with the second promise breaking function, and this second promise breaking function is the function relevant to promise breaking numerical value or promise breaking issue.
In the present embodiment, the second promise breaking function is the function relevant to promise breaking numerical value or promise breaking issue, can be according to promise breaking
The value of different range configuration the second promise breaking function of numerical value.Such as shown in formula (4).
In formula (4), f'(k) it is the second promise breaking function, k represents that kth is individual, and N represents individual in colony
Body sum.
The first predefined colony Default Probability can be selected pre-to the 5th according to the application demand to group risk
Colony's Default Probability of definition, according to threshold value T that the Default Probability configuration of each predefined colony is corresponding, when in advance
Colony's promise breaking it is when colony's Default Probability of definition is more than or equal to T.
The first variable of corresponding colony's promise breaking is built according to predefined colony Default Probability and corresponding threshold value
Y。
In formula (5), the first variable Y is designated 1 expression promise breaking, is designated 0 mark and does not breaks a contract.P0Table
Show colony's Default Probability.
Step 204, selects corresponding predefined colony Default Probability according to the business demand of selected data.
Specifically, according to business need select different predefined colony Default Probability.If business is every
It is the same that approximation is regarded in position individual promise breaking impact as, and the impact of colony's promise breaking is also nondistinctive, uses the
One predefines colony's Default Probability；If business only considers population size size, and does not consider individual impact
Diversity, then use second to predefine colony's Default Probability；If business only considers the diversity of individual impact,
And do not consider that the promise breaking that population size size is brought affects difference, use the 3rd to predefine colony's Default Probability；
If business had both considered individual power of influence difference, it is also desirable to the group influence that consideration population size difference causes is not
With, use the 4th to predefine colony's Default Probability.Four kind the 5th predefines colony's Default Probability also according to business
Demand can be chosen.
Step 206, asks for the colony of this first kind population data according to the predefined colony Default Probability chosen
Default Probability, calculates this predefined colony chosen according to colony's Default Probability of this first kind population data
First variable of colony's promise breaking that Default Probability is corresponding.
In the present embodiment, the violation of agreement of each individuality in first kind population data is it is known that pre-according to each
Colony's Default Probability of definition can ask for obtaining colony's Default Probability of this first kind population data.
Such as, predefine colony's Default Probability according to first and ask for colony's Default Probability of first kind population data,
If individual sum is 100 in first kind population data, individual promise breaking number is 20, then this first kind colony number
According to colony's Default Probability be 20/100=0.2.
It is corresponding that colony's Default Probability according to first kind population data calculates each predefined colony Default Probability
Colony promise breaking the first variable.
Step 208, obtains the population characteristic variable of this first kind population data.
In one embodiment, as it is shown on figure 3, the population characteristic variable of this acquisition first kind population data
Step includes:
Step 302, obtains the characteristic variable of individual credit score in first kind population data.
In the present embodiment, the characteristic variable of individual credit score includes in average, standard deviation, kurtosis, the degree of bias
One or more.
Individual credit score refers to the credit score that individual history credit data is estimated obtain.
Average is to represent that in one group of data, all data sums are again divided by the number of these group data.
Standard deviation is the quadratic sum that in one group of data, all data deduct its meansigma methods, and acquired results is divided by this group
The number of data, then income value is opened radical sign, the number of gained is the standard deviation of these group data.Standard deviation is anti-
Reflect one group of the most frequently used a kind of quantized versions of data discrete degree, be the important indicator representing degree of accuracy.
Kurtosis, also known as coefficient of kurtosis, characterizes probability density distribution curve characteristic number of peak value height at meansigma methods.
Assumed group comprises individuality, xiRepresent the credit score that i-th is individual, then kurtosis can be:
In formula (6), L is kurtosis, xiRepresenting the credit score that i-th is individual, x represents individual letter
By the meansigma methods of mark.
The degree of bias is statistical data distribution skew direction and the tolerance of degree, statistical data distribution degree of asymmetry
Numerical characteristic, characterizes the probability distribution density curve characteristic number relative to the asymmetric degree of meansigma methods.
In formula (7), C is the degree of bias, xiRepresenting the credit score that i-th is individual, x represents individual letter
By the meansigma methods of mark, N represents the sum of individual in population.
Step 304, constructs colony's attribute character variable according to the characteristic variable of this individuality credit score.
In the present embodiment, this group property characteristic variable includes population base property distribution characteristic variable, colony
Social behaviors distribution characteristics variable, group interest distribution characteristics variable, the on-line off-line characteristic variable of colony, group
In body geographic location feature variable and colony's topological characteristic variable one or more.
Structure colony primary attribute distribution characteristics variable.Population base property distribution characteristic variable can include colony
Age, sex, occupation, the characteristic variable such as average, standard deviation, kurtosis and the degree of bias of educational background etc..Colony
The average at age, standard deviation, kurtosis and the degree of bias be respectively adopted the average at age of individual in population, standard
Difference, kurtosis and the degree of bias represent.Same, the sex of colony, occupation, the average of educational background, standard deviation, peak
Degree and the degree of bias use too the sex of individual in population, occupation, educational background average, standard deviation, kurtosis and
The degree of bias represents.
Structure colony Social behaviors distribution variable.Individual Social behaviors data can include that messaging, comment are sent out
Table, forward frequency, game login to enliven frequency, all kinds social networking application logs in and enlivens the statistics spy such as frequency
Levy.
Structure group interest distribution characteristics variable.By individual shopping, reading, news, video, music,
The interest diggings such as search, structure represents the interest statistical nature of colony.
Structure colony o2o (on-line off-line) characteristic variable.Service under the line used under individual line, payment etc.,
According to different COS classification；Construct on-line off-line service or the frequency of COS, amount of money statistical nature
Deng.As commonly used made a reservation, the service average of frequency, the standard deviation such as special train, use make a reservation, special train etc.
The average of the amount of money of service, standard deviation etc..
Structure colony geographic location feature variable.By the position of mobile social activity location individuality, or can pass through
LBS service, as registered, navigating, the service such as neighbouring people, or the individual feelings allowed such as Wi-Fi, base station
Obtain individual geographic positional information, the position occurred by these Information Statistics individualities under condition, and add up appearance
The frequency of position and duration etc., obtain work, residence information, and structure colony location distribution feature becomes
Amount.
This step constructing colony's attribute character variable according to the characteristic variable of this individuality credit score includes: will
The characteristic variable of this individuality credit score is as this group property characteristic variable.
Build colony's topological characteristic variable.The relation between individual in population that builds forms colony's topological characteristic and becomes
Amount.Calculate the indexs such as the k-shell of colony's network, cluster coefficients, centrad.K-shell, cluster coefficients, in
Heart degree can react the compactness of group relation.
K-shell is used for calculating nodes power of influence, first finds out the node that all degree are 1, is deleted,
Then in remaining node, degree of continuing to search for is the node of 1, and deletes, until degree of not having is 1 in network
Node, the node k-shell value that all degree are 1 deleted before is entered as 1.By same procedure then
Degree of finding is the node of 2, and is entered as 2, the like, until all nodes have k-shell in network
Value, k-shell value is the highest, represents that power of influence is the biggest.
Cluster coefficients is to represent the coefficient of a nodes aggregation extent.In particular network, due to phase
Relation to high density junction point, node always trends towards setting up one group of tight membership credentials.
Centrad is the important node being capable of identify that in network, the significance level of node by the topological attribute of network,
Construction features and node particular location in a network determines.
Step 306, obtains population characteristic variable according to this group property characteristic variable.
Using group property characteristic variable as population characteristic variable.
Additionally, the data such as the title of colony, an attribute classified variable as training also can be obtained.
Degree of deep learning algorithm etc. can also be used to excavate hidden variable, as the input of colony's violation correction model
Variable.
Step 210, obtains colony according to the population characteristic variable of this first variable He this first kind population data and disobeys
About forecast model.
Population characteristic variable according to this first variable and this first kind population data uses regression algorithm to instruct
Practice or use degree of deep learning algorithm to be trained study and obtain colony's violation correction model.
Specifically, regression algorithm can be the tradition machine in normal service learning algorithm such as linear regression, logistic regression.With
Logistic regression illustrates.
Assuming there be n population characteristic variable, colony's Default Probability is:
In formula (8), P0For colony's Default Probability, βiAnd β0The parameter obtained for training, diSpecial for colony
Levy variable.
Or use the Structure learning methods such as degree of deep learning algorithm, learn colony's promise breaking and step 306 feature
The non-linear relation of variable, the final Default Probability predicting colony.
In addition, it is possible to by the first corresponding for each predefined colony Default Probability variable and the first population data
Population characteristic variable be trained, obtain predicting the outcome of colony's Default Probability, can filter out and first group
What the population characteristic variable match of volume data obtained predict the outcome the first optimal variable.
Step 212, carries out colony's violation correction according to this colony's violation correction model to this Equations of The Second Kind population data
Obtain colony's Default Probability of this Equations of The Second Kind population data.
Because of the promise breaking of first kind population data and Equations of The Second Kind population data, whether problem has mathematics dependency, and
And easily observation or calculating, can be to the population characteristic of Equations of The Second Kind population data structure with first kind population data
The population characteristic variable that variable is identical, then uses first kind population data to train the colony's violation correction obtained
Model, is predicted obtaining the colony of Equations of The Second Kind population data to colony's violation of agreement of Equations of The Second Kind population data
Default Probability.
Step 214, is converted into colony's credit score of correspondence by colony's Default Probability of this Equations of The Second Kind population data
Value.
In the present embodiment, colony's Default Probability [0,1] is divided into integer space according to distribution.Can use equifrequency,
Equidistant discretization method such as grade, is converted into colony's Default Probability integer, or colony's Default Probability is converted
For credit score.
Colony's Default Probability is converted into shown in the computing formula such as formula (9) of credit score.
In formula (9), S is colony's credit score, divides on the basis of base, and step is mark step-length, and p is pre-
The colony's Default Probability surveyed.
Additionally, may be used without formula (9) colony's Default Probability of first kind population data is converted into correspondence
Colony's credit score.
The method of above-mentioned prediction colony credit, is divided into population data and comprises the first of known violation of agreement individuality
The Equations of The Second Kind population data that types of populations data are individual with comprising unknown violation of agreement, according to first kind population data
Calculate the first variable that predefined colony Default Probability is corresponding, obtain the population characteristic variable of first kind colony,
Population characteristic variable according to the first variable and first kind population data is trained obtaining colony's violation correction mould
Type, according to colony's Default Probability of colony's violation correction model prediction Equations of The Second Kind population data, by the second monoid
Colony's Default Probability of body is converted into colony's credit score of correspondence, can reflect group by colony's credit score
The credit situation of body, it is achieved that the assessment to colony's credit, and can be to comprising the individual of unknown violation of agreement
Colony's credit is predicted.
In one embodiment, after the step of described acquisition selected data, above-mentioned prediction colony credit
Method also includes: judge in this selected data that the most individual violations of agreement are whether it is known that the most then basis
The business demand of selected data selects corresponding predefined colony Default Probability, predefined according to choose
Colony's Default Probability calculates colony's Default Probability of this selected data, and this colony's Default Probability is converted into right
The colony's credit score answered, if it is not, be then divided into first kind population data and Equations of The Second Kind colony by this selected data
Data, comprise the individuality of known violation of agreement in this first kind population data, wrap in this Equations of The Second Kind population data
Individuality containing unknown violation of agreement.
If the individual violation of agreement in selected data all it is known that, can be directly separated according to predefined colony
About colony's Default Probability of this selected data of probability calculation.
In one embodiment, if the number of individuals of selected data is less than or equal to predetermined number threshold value, then can root
Colony's Default Probability of this selected data is calculated according to predefined colony Default Probability.
If the number of individuals of selected data is more than predetermined number threshold value, then judge whole individualities in this selected data
Whether violation of agreement is it is known that the most then select corresponding predefined group according to the business demand of selected data
Body Default Probability, the colony's promise breaking calculating this selected data according to the predefined colony Default Probability chosen is general
Rate, and this colony's Default Probability is converted into colony's credit score of correspondence, if it is not, then by this selected data
It is divided into first kind population data and Equations of The Second Kind population data, this first kind population data comprises known promise breaking feelings
The individuality of condition, comprises the individuality of unknown violation of agreement in described Equations of The Second Kind population data.
In one embodiment, after obtaining the step of population characteristic variable, the side of this prediction colony credit
Method also includes: ask for the dependency of each population characteristic variable and this first variable；According to this first variable
The dependency population characteristic variable sequentially screening out predetermined quantity from high to low.
In the present embodiment, each population characteristic variable d can be calculatediChi-square value or Pearson came with the first variable Y
Correlation coefficienies etc. obtain each population characteristic variable diDependency with the first variable Y.
By each population characteristic variable diWith the dependency of the first variable Y according to entering from high to low or from low to high
Row sequence.
From ranking results according to this first correlation of variables from high to low sequentially screen out predetermined quantity
Population characteristic variable.
Predetermined quantity can set as required, such as 1,2,3 etc..
The population characteristic variable high by filtering out dependency, can more Accurate Prediction colony Default Probability, and
Reduce amount of calculation.
Further, this is instructed according to the population characteristic variable of this first variable He this first kind population data
The step getting colony's violation correction model includes: according to this first variable and the predetermined quantity filtered out
The population characteristic variable of first kind population data is trained obtaining colony's violation correction model.
It should be noted that the colony's credit score obtained is a colony to be treated as an entity.No
But compensate for the problem such as individual behavior, sparse, the individual swindle of mark, additionally by the social attribute of colony,
Preferably show the groupment correlated characteristic of individuality.Directly use colony's credit score, both may be used for assessment
The financial status of some colony, awards the group risk early warning in colony's credit foundation, debt-credit and colony
Letter amount is adjusted, and it will be seen that again the financial status of colony, makees colony's portrait and more portray and dig
Pick.Assessment and calculating to single colony credit risk, can help to observe overall situation larger-scale colony promise breaking wind
The change of danger and migration trend, thus early warning occurrence of large-area systematic risk.
Furthermore, colony's credit predicts the outcome the estimation that can apply to colony's financial status, as a company,
The prediction of the Default Probability of enterprise and judgement, be converted into the reference frame of business standing prediction, be applied to enterprise
Reference scene.Equally instructing the credit of individual, each user has personal credit score value and place
Multiple credit score attributes such as colony's credit score, more fully understand user credit state.
Fig. 4 is the structured flowchart of the device predicting colony's credit in an embodiment.As shown in Figure 4, a kind of
The device of prediction colony credit, including the first acquisition module 402, choose module 404, ask for module 406,
Second acquisition module 408, model building module 410, prediction module 412 and conversion module 414.Wherein:
First acquisition module 402 is used for obtaining selected data, and this selected data is divided into first kind population data
With Equations of The Second Kind population data, this first kind population data comprises the individuality of known violation of agreement, this Equations of The Second Kind
Population data comprises the individuality of unknown violation of agreement.
Choose module 404 and select corresponding predefined colony promise breaking for the business demand according to selected data
Probability.
Ask for module 406 for asking for this first kind colony number according to the predefined colony Default Probability chosen
According to colony's Default Probability, calculate choose predefined according to the colony Default Probability of this first kind population data
Colony's Default Probability corresponding colony promise breaking the first variable.
Second acquisition module 408 is for obtaining the population characteristic variable of this first kind population data.
Model building module 410 becomes for the population characteristic according to this first variable and this first kind population data
Amount is trained obtaining colony's violation correction model.
Prediction module 412 is for carrying out colony according to this colony's violation correction model to this Equations of The Second Kind population data
Violation correction obtains colony's Default Probability of this Equations of The Second Kind colony.
Conversion module 414 for being converted into colony's credit of correspondence by colony's Default Probability of this Equations of The Second Kind colony
Score value.
The device of above-mentioned prediction colony credit, comprises known violation of agreement individuality by being divided into by population data
The Equations of The Second Kind population data that first kind population data is individual with comprising unknown violation of agreement, according to first kind colony
Data calculate the first variable that predefined colony Default Probability is corresponding, obtain the population characteristic of first kind colony
Variable, is trained obtaining colony's promise breaking according to the population characteristic variable of the first variable and first kind population data
Forecast model, according to colony's Default Probability of colony's violation correction model prediction Equations of The Second Kind population data, by
Colony's Default Probability of two types of populations is converted into colony's credit score of correspondence, can be anti-by colony's credit score
Mirror the credit situation of colony, it is achieved that the assessment to colony's credit, and can be to comprising unknown violation of agreement
Individual colony's credit is predicted.
Fig. 5 is the structured flowchart of the device predicting colony's credit in another embodiment.As it is shown in figure 5, one
The device of kind of prediction colony credit, including the first acquisition module 402, choose module 404, ask for module 406,
Second acquisition module 408, model building module 410, prediction module 412 and conversion module 414, also include
Build module 401.Wherein:
Build module 401 to be used for configuring corresponding threshold value according to predefined colony Default Probability, and according to respectively
Individual predefined colony Default Probability and corresponding threshold value build the first variable of corresponding colony's promise breaking, and group
Body Default Probability is more than or equal to described threshold value, then this colony is promise breaking colony.
The Default Probability of this predefined colony includes first predefining colony's Default Probability, second predefining group
Body Default Probability, the 3rd predefine colony's Default Probability, the 4th predefine colony's Default Probability, the 5th make a reservation for
In justice colony Default Probability one or more；
Obtain in colony the first ratio of the number of individuals that violations occur and individual in population sum, by this
One ratio predefines colony's Default Probability as first；
Obtain the factor of influence of population size, according to factor of influence and the first predefined group of this population size
Body Default Probability obtains second and predefines colony's Default Probability；
Obtain the property value of the forward attribute of the first promise breaking function and individual in population, just obtain each individuality
To the product of the property value of attribute and the first promise breaking function and the property value of forward attribute with individual in population
And seek ratio the second ratio being worth to, then the sum of this second ratio with this individual in population is asked ratio be worth
To the 3rd ratio, the 3rd ratio is predefined colony's Default Probability as the 3rd, or, obtain first and disobey
The about property value of the negative sense attribute of function and individual in population, obtains the property value of the negative sense attribute of each individuality
Inverse and the first promise breaking function product and property value reciprocal of negative sense attribute with individual in population
With seek ratio the second ratio being worth to, then the sum of the second ratio with this individual in population is asked than being worth to
Three ratios, predefine colony's Default Probability using the 3rd ratio as the 3rd, wherein, in the first promise breaking function
If individual promise breaking, then the value of the first promise breaking function is 1, if individuality is not broken a contract, then the value of the first promise breaking function is
0；
The forward attribute of this individual in population is power of influence, the accrediting amount, individual credit score or technorati authority,
The negative sense attribute of this individual in population is individual income data or stability data.
This second is predefined colony's Default Probability predefines colony's Default Probability with the 3rd and be multiplied that to obtain the 4th pre-
Definition colony Default Probability；
Predefine colony's Default Probability to predefine in colony's Default Probability individual Default Probability to the 4th equal by first
Replace with the behavior that individual promise breaking occurs to obtain four kind the 5th and predefine Default Probability.In one embodiment,
This second acquisition module 408 is additionally operable to obtain the characteristic variable of individual credit score in first kind population data；
Characteristic variable according to this individuality credit score constructs colony's attribute character variable；And according to this group property
Characteristic variable obtains population characteristic variable.
In one embodiment, the characteristic variable of this individuality credit score include average, standard deviation, kurtosis,
In the degree of bias one or more；This group property characteristic variable includes population base property distribution characteristic variable, group
Body Social behaviors distribution characteristics variable, group interest distribution characteristics variable, the on-line off-line characteristic variable of colony and
In colony's geographic location feature variable one or more；
This second acquisition module 408 is additionally operable to the characteristic variable of this individuality credit score as this group property
Characteristic variable.
In one embodiment, model building module 410 is additionally operable to according to this first variable and this first monoid
The population characteristic variable of volume data uses regression algorithm to be trained or uses degree of deep learning algorithm to be trained learning
Acquistion is to colony's violation correction model.
Fig. 6 is the structured flowchart of the device predicting colony's credit in another embodiment.As shown in Figure 6, one
The device of kind of prediction colony credit, including the first acquisition module 402, choose module 404, ask for module 406,
Second acquisition module 408, model building module 410, prediction module 412 and conversion module 414, also include
Judge module 416.Wherein:
Judge module 416 is for after obtaining selected data, it is judged that the most individual disobeying in this selected data
About whether situation is it is known that the most then choose module 404 to be additionally operable to the business demand selection according to selected data
Corresponding predefined colony Default Probability, asks for module 406 and is additionally operable to according to the promise breaking of predefined colony general
Rate calculates colony's Default Probability of described selected data, and conversion module 414 is additionally operable to general for the promise breaking of described colony
Rate is converted into colony's credit score of correspondence, if it is not, then described selected data is divided by the first acquisition module 402
For first kind population data and Equations of The Second Kind population data, this first kind population data comprises known violation of agreement
Individuality, this Equations of The Second Kind population data comprises the individuality of unknown violation of agreement.
Fig. 7 is the structured flowchart of the device predicting colony's credit in another embodiment.As it is shown in fig. 7, one
The device of kind of prediction colony credit, except including the first acquisition module 402, choosing module 404, ask for module
406, the second acquisition module 408, model building module 410, prediction module 412 and conversion module 414, also
Including computing module 418 and screening module 420.Wherein:
Computing module 418 is for, after obtaining the population characteristic variable of first kind population data, asking for each
Population characteristic variable and the dependency of this first variable.
Screening module 420 for according to this first correlation of variables from high to low sequentially screen out predetermined number
The population characteristic variable of amount.
This model building module 410 is additionally operable to the first kind according to this first variable with the predetermined quantity filtered out
The population characteristic variable of population data is trained obtaining colony's violation correction model.
In other embodiments, a kind of device predicting colony's credit can include that building module 401, first obtains
Module 402, choose module 404, ask for module the 406, second acquisition module 408, model building module 410,
In prediction module 412, conversion module 414, judge module 416, computing module 418 and screening module 420
The most possible combination.
One of ordinary skill in the art will appreciate that all or part of flow process realizing in above-described embodiment method,
Can be by computer program and complete to instruct relevant hardware, described program can be stored in one non-easily
In the property lost computer read/write memory medium, this program is upon execution, it may include such as the enforcement of above-mentioned each method
The flow process of example.Wherein, described storage medium can be magnetic disc, CD, read-only store-memory body (Read-Only
Memory, ROM) etc..
Embodiment described above only have expressed the several embodiments of the present invention, and it describes more concrete and detailed,
But therefore can not be interpreted as the restriction to the scope of the claims of the present invention.It should be pointed out that, for this area
Those of ordinary skill for, without departing from the inventive concept of the premise, it is also possible to make some deformation and
Improving, these broadly fall into protection scope of the present invention.Therefore, the protection domain of patent of the present invention should be with appended
Claim is as the criterion.
Claims (14)
1. the method predicting colony's credit, including:
Obtain selected data, described selected data be divided into first kind population data and Equations of The Second Kind population data,
Described first kind population data comprises the individuality of known violation of agreement, described Equations of The Second Kind population data comprises
The individuality of unknown violation of agreement；
Business demand according to selected data selects corresponding predefined colony Default Probability；
The colony's promise breaking asking for described first kind population data according to the predefined colony Default Probability selected is general
Rate, the predefined colony chosen according to colony's Default Probability calculating of described first kind population data disobeys
First variable of colony's promise breaking that about probability is corresponding；
Obtain the population characteristic variable of described first kind population data；
Population characteristic variable according to described first variable and described first kind population data is trained obtaining group
Body violation correction model；
According to described colony violation correction model, described Equations of The Second Kind population data is carried out colony's violation correction to obtain
Colony's Default Probability of described Equations of The Second Kind population data；
Colony's Default Probability of described Equations of The Second Kind population data is converted into colony's credit score of correspondence.
Method the most according to claim 1, it is characterised in that described method also includes:
Configure corresponding threshold value according to predefined colony Default Probability, and disobey according to each predefined colony
About probability and corresponding threshold value builds the first variable of corresponding colony's promise breaking, and colony's Default Probability more than or
Equal to described threshold value, the most described colony is promise breaking colony；
Predefined colony Default Probability include first predefine colony's Default Probability, second predefine colony disobey
About probability, the 3rd predefine colony's Default Probability, the 4th predefine colony's Default Probability, the 5th predefine group
In body Default Probability one or more；
Obtain the first ratio of the number of individuals that violations occur in colony and individual in population sum, by described
First ratio predefines colony's Default Probability as first；
Obtaining the factor of influence of population size, factor of influence and first according to described population size are predefined
Colony's Default Probability obtains second and predefines colony's Default Probability；
Obtain the property value of the forward attribute of the first promise breaking function and individual in population, just obtain each individuality
To the product of the property value of attribute and the first promise breaking function and the property value of forward attribute with individual in population
And seek ratio the second ratio being worth to, then the sum of described second ratio with described individual in population is sought ratio
It is worth to the 3rd ratio, described 3rd ratio is predefined colony's Default Probability as the 3rd, or, obtain
The property value of the negative sense attribute of the first promise breaking function and individual in population, obtains the negative sense attribute of each individuality
The product of the inverse of property value and the first promise breaking function and the property value of negative sense attribute with individual in population
Reciprocal and seek ratio the second ratio being worth to, then the sum of the second ratio with described individual in population is sought ratio
It is worth to the 3rd ratio, described 3rd ratio is predefined colony's Default Probability as the 3rd, wherein, first
If individual promise breaking in promise breaking function, then the value of the first promise breaking function is 1, if individuality is not broken a contract, then and the first promise breaking
The value of function is 0；
Predefine colony's Default Probability to predefine colony's Default Probability with the 3rd and be multiplied by described second and obtain the 4th
Predefined colony Default Probability；
Predefine colony's Default Probability to predefine in colony's Default Probability individual Default Probability to the 4th equal by first
Replace with the behavior that individual promise breaking occurs to obtain four kind the 5th and predefine Default Probability.
Method the most according to claim 1, it is characterised in that described acquisition described first kind colony number
According to the step of population characteristic variable include:
Obtain the characteristic variable of described first kind individual in population credit score；
Characteristic variable according to described individual credit score constructs colony's attribute character variable；
The population characteristic variable of described first kind population data is obtained according to described group property characteristic variable；
The characteristic variable of described individual credit score includes in average, standard deviation, kurtosis, the degree of bias a kind of or many
Kind；Described group property characteristic variable includes that population base property distribution characteristic variable, colony's Social behaviors divide
Cloth characteristic variable, group interest distribution characteristics variable, the on-line off-line characteristic variable of colony, colony geographical position
In characteristic variable and colony's topological characteristic variable one or more.
Method the most according to claim 1, it is characterised in that in the step of described acquisition selected data
Afterwards, described method also includes:
Judge in described selected data that the most individual violations of agreement are whether it is known that the most then according to selected number
According to business demand select corresponding predefined colony Default Probability, disobey according to the predefined colony chosen
About colony's Default Probability of selected data described in probability calculation, and described colony Default Probability is converted into correspondence
Colony's credit score, if it is not, then described selected data to be divided into first kind population data and Equations of The Second Kind colony
Data, comprise the individuality of known violation of agreement, described Equations of The Second Kind population data in described first kind population data
In comprise the individuality of unknown violation of agreement.
Method the most according to claim 1, it is characterised in that according to described first variable and described
The population characteristic variable of one types of populations data is trained obtaining the step of colony's violation correction model and includes:
Population characteristic variable according to described first variable and described first kind population data uses regression algorithm to enter
Row training or use degree of deep learning algorithm are trained study and obtain colony's violation correction model.
Method the most according to claim 1, it is characterised in that obtaining described first kind population data
Population characteristic variable step after, described method also includes:
Ask for the dependency of each population characteristic variable and described first variable；
Become according to the described first correlation of variables population characteristic sequentially screening out predetermined quantity from high to low
Amount；
The described population characteristic variable according to described first variable and described first kind population data is trained
Step to colony's violation correction model includes:
Population characteristic according to described first variable with the described first kind population data of the predetermined quantity filtered out
Variable is trained obtaining colony's violation correction model.
Method the most according to claim 2, it is characterised in that the forward attribute of described individual in population
For power of influence, the accrediting amount, individual credit score or technorati authority, the negative sense attribute of described individual in population is
Individual income data or stability data.
8. the device predicting colony's credit, it is characterised in that including:
First acquisition module, is used for obtaining selected data, and described selected data is divided into first kind population data
With Equations of The Second Kind population data, described first kind population data comprises the individuality of known violation of agreement, described
Two types of populations data comprise the individuality of unknown violation of agreement；
Choose module, select corresponding predefined colony promise breaking general for the business demand according to selected data
Rate；
Ask for module, for asking for described first kind colony number according to the predefined colony Default Probability chosen
According to colony's Default Probability, according to the colony Default Probability of described first kind population data calculate described in choose
First variable of colony's promise breaking that predefined colony Default Probability is corresponding；
Second acquisition module, for obtaining the population characteristic variable of described first kind population data；
Model building module, for according to described first variable and the population characteristic of described first kind population data
Variable is trained obtaining colony's violation correction model；
Prediction module, for carrying out group according to described colony violation correction model to described Equations of The Second Kind population data
Body violation correction obtains colony's Default Probability of described Equations of The Second Kind colony；
Conversion module, for being converted into colony's credit of correspondence by colony's Default Probability of described Equations of The Second Kind colony
Score value.
Device the most according to claim 8, it is characterised in that described device also includes:
Build module, for configuring corresponding threshold value according to predefined colony Default Probability, and according to each
Predefined colony Default Probability and corresponding threshold value build the first variable of corresponding colony's promise breaking, and colony
Default Probability is more than or equal to described threshold value, and the most described colony is promise breaking colony；
Described predefined colony Default Probability includes first predefining colony's Default Probability, second predefining group
Body Default Probability, the 3rd predefine colony's Default Probability, the 4th predefine colony's Default Probability, the 5th make a reservation for
In justice colony Default Probability one or more；
Obtain the first ratio of the number of individuals that violations occur in colony and individual in population sum, by described
First ratio predefines colony's Default Probability as first；
Obtaining the factor of influence of population size, factor of influence and first according to described population size are predefined
Colony's Default Probability obtains second and predefines colony's Default Probability；
Obtain the property value of the forward attribute of the first promise breaking function and individual in population, just obtain each individuality
To the product of the property value of attribute and the first promise breaking function and the property value of forward attribute with individual in population
And seek ratio the second ratio being worth to, then the sum of described second ratio with described individual in population is sought ratio
It is worth to the 3rd ratio, described 3rd ratio is predefined colony's Default Probability as the 3rd, or, obtain
The property value of the negative sense attribute of the first promise breaking function and individual in population, obtains the negative sense attribute of each individuality
The product of the inverse of property value and the first promise breaking function and the property value of negative sense attribute with individual in population
Reciprocal and seek ratio the second ratio being worth to, then the sum of the second ratio with described individual in population is sought ratio
It is worth to the 3rd ratio, described 3rd ratio is predefined colony's Default Probability as the 3rd, wherein, first
If individual promise breaking in promise breaking function, then the value of the first promise breaking function is 1, if individuality is not broken a contract, then and the first promise breaking
The value of function is 0；
Predefine colony's Default Probability to predefine colony's Default Probability with the 3rd and be multiplied by described second and obtain the 4th
Predefined colony Default Probability；
Predefine colony's Default Probability to predefine in colony's Default Probability individual Default Probability to the 4th equal by first
Replace with the behavior that individual promise breaking occurs to obtain four kind the 5th and predefine Default Probability.
Device the most according to claim 8, it is characterised in that described second acquisition module is additionally operable to
Obtain the characteristic variable of individual credit score in described first kind population data；According to described individual credit score
Characteristic variable structure colony attribute character variable；And obtain colony according to described group property characteristic variable
Characteristic variable；
The characteristic variable of described individual credit score includes in average, standard deviation, kurtosis, the degree of bias a kind of or many
Kind；Described group property characteristic variable includes that population base property distribution characteristic variable, colony's Social behaviors divide
Cloth characteristic variable, group interest distribution characteristics variable, the on-line off-line characteristic variable of colony and colony geographical position
In characteristic variable one or more.
11. devices according to claim 8, it is characterised in that described device also includes:
Judge module, for after described acquisition selected data, it is judged that the most individual in described selected data
Violation of agreement whether it is known that choose module described in the most then and be additionally operable to choose correspondence according to selected data
Predefined colony Default Probability, described in ask for module and be additionally operable to calculate according to predefined colony Default Probability
Colony's Default Probability of described selected data, described conversion module is additionally operable to convert described colony Default Probability
For corresponding colony's credit score, if it is not, described selected data is divided into first by the most described first acquisition module
Types of populations data and Equations of The Second Kind population data, comprise the individual of known violation of agreement in described first kind population data
Body, comprises the individuality of unknown violation of agreement in described Equations of The Second Kind population data.
12. devices according to claim 8, it is characterised in that described model building module is additionally operable to
Population characteristic variable according to described first variable and described first kind population data uses regression algorithm to instruct
Practice or use degree of deep learning algorithm to be trained study and obtain colony's violation correction model.
13. devices according to claim 8, it is characterised in that described device also includes:
Computing module, for after obtaining the population characteristic variable of described first kind population data, asks for each
Individual population characteristic variable and the dependency of described first variable；
Screening module, for according to described first correlation of variables from high to low sequentially screen out predetermined number
The population characteristic variable of amount；
Described model building module is additionally operable to described the according to described first variable and the predetermined quantity that filters out
The population characteristic variable of one types of populations data is trained obtaining colony's violation correction model.
14. devices according to claim 9, it is characterised in that the forward of described individual in population belongs to
Property be power of influence, the accrediting amount, individual credit score or technorati authority, the negative sense attribute of described individual in population
For individual income data or stability data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610414335.7A CN105894372B (en) | 2016-06-13 | 2016-06-13 | The method and apparatus for predicting colony's credit |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610414335.7A CN105894372B (en) | 2016-06-13 | 2016-06-13 | The method and apparatus for predicting colony's credit |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105894372A true CN105894372A (en) | 2016-08-24 |
| CN105894372B CN105894372B (en) | 2018-03-16 |
Family
ID=56729464
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610414335.7A Active CN105894372B (en) | 2016-06-13 | 2016-06-13 | The method and apparatus for predicting colony's credit |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105894372B (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107104978A (en) * | 2017-05-24 | 2017-08-29 | 赖洪昌 | A kind of network risks method for early warning based on deep learning |
| 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 |
| CN107292424A (en) * | 2017-06-01 | 2017-10-24 | 四川新网银行股份有限公司 | A kind of anti-fraud and credit risk forecast method based on complicated social networks |
| CN107798597A (en) * | 2017-10-09 | 2018-03-13 | 上海二三四五金融科技有限公司 | A kind of dynamic excessive risk visitor group detection method and system |
| CN107871277A (en) * | 2017-07-25 | 2018-04-03 | 平安普惠企业管理有限公司 | The method and computer-readable recording medium that server, customer relationship are excavated |
| CN108399564A (en) * | 2017-02-08 | 2018-08-14 | 腾讯科技（深圳）有限公司 | Credit-graded approach and device |
| CN108416662A (en) * | 2017-02-10 | 2018-08-17 | 腾讯科技（深圳）有限公司 | A kind of data verification method and device |
| CN108564460A (en) * | 2018-01-12 | 2018-09-21 | 阳光财产保险股份有限公司 | Real-time fraud detection method under internet credit scene and device |
| CN109670961A (en) * | 2018-09-26 | 2019-04-23 | 深圳壹账通智能科技有限公司 | Management method, device, terminal device and computer readable storage medium after throwing |
| WO2019114481A1 (en) * | 2017-12-13 | 2019-06-20 | 腾讯科技（深圳）有限公司 | Cluster type recognition method, apparatus, electronic apparatus, and storage medium |
| CN110263802A (en) * | 2019-04-04 | 2019-09-20 | 平安科技（深圳）有限公司 | Credit data analysing method and relevant device based on Density Clustering |
| CN111164633A (en) * | 2018-05-31 | 2020-05-15 | 重庆小雨点小额贷款有限公司 | Method and device for adjusting grading card model, server and storage medium |
| CN112561635A (en) * | 2020-12-11 | 2021-03-26 | 永安行科技股份有限公司 | Shared article renting and returning method and system based on credit system |
| CN112884571A (en) * | 2021-02-26 | 2021-06-01 | 中库(北京)数据系统有限公司 | Method and device for determining credit risk level |
| CN113362120A (en) * | 2021-08-11 | 2021-09-07 | 北京达佳互联信息技术有限公司 | Group determination method and device, electronic equipment and computer readable storage medium |
| CN113643125A (en) * | 2021-08-30 | 2021-11-12 | 天元大数据信用管理有限公司 | Credit line measuring and calculating method, equipment and medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040044505A1 (en) * | 2002-09-04 | 2004-03-04 | Richard Horwitz | Method and system for identifying risk factors |
| CN101996381A (en) * | 2009-08-14 | 2011-03-30 | 中国工商银行股份有限公司 | Method and system for calculating retail asset risk |
| CN102163205A (en) * | 2010-02-21 | 2011-08-24 | 施章祖 | Automatic excavation system for analogous customer groups |
-
2016
- 2016-06-13 CN CN201610414335.7A patent/CN105894372B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040044505A1 (en) * | 2002-09-04 | 2004-03-04 | Richard Horwitz | Method and system for identifying risk factors |
| CN101996381A (en) * | 2009-08-14 | 2011-03-30 | 中国工商银行股份有限公司 | Method and system for calculating retail asset risk |
| CN102163205A (en) * | 2010-02-21 | 2011-08-24 | 施章祖 | Automatic excavation system for analogous customer groups |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11816727B2 (en) | 2017-02-08 | 2023-11-14 | Tencent Technology (Shenzhen) Company Limited | Credit scoring method and server |
| US11170436B2 (en) | 2017-02-08 | 2021-11-09 | Tencent Technology (Shenzhen) Company Limited | Credit scoring method and server |
| CN108399564B (en) * | 2017-02-08 | 2021-03-19 | 腾讯科技（深圳）有限公司 | Credit scoring method and device |
| CN108399564A (en) * | 2017-02-08 | 2018-08-14 | 腾讯科技（深圳）有限公司 | Credit-graded approach and device |
| WO2018145586A1 (en) * | 2017-02-08 | 2018-08-16 | 腾讯科技（深圳）有限公司 | Credit scoring method and server |
| CN108416662A (en) * | 2017-02-10 | 2018-08-17 | 腾讯科技（深圳）有限公司 | A kind of data verification method and device |
| CN108416662B (en) * | 2017-02-10 | 2021-09-21 | 腾讯科技（深圳）有限公司 | Data verification method and device |
| CN107104978B (en) * | 2017-05-24 | 2019-12-24 | 赖洪昌 | Network risk early warning method based on deep learning |
| CN107104978A (en) * | 2017-05-24 | 2017-08-29 | 赖洪昌 | A kind of network risks method for early warning based on deep learning |
| CN107292424A (en) * | 2017-06-01 | 2017-10-24 | 四川新网银行股份有限公司 | A kind of anti-fraud and credit risk forecast method based on complicated social networks |
| CN107292424B (en) * | 2017-06-01 | 2020-01-21 | 四川新网银行股份有限公司 | Anti-fraud and credit risk prediction method based on complex social network |
| CN107194623B (en) * | 2017-07-20 | 2021-01-05 | 深圳市分期乐网络科技有限公司 | Group partner fraud discovery method and device |
| 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 |
| CN107871277A (en) * | 2017-07-25 | 2018-04-03 | 平安普惠企业管理有限公司 | The method and computer-readable recording medium that server, customer relationship are excavated |
| CN107798597A (en) * | 2017-10-09 | 2018-03-13 | 上海二三四五金融科技有限公司 | A kind of dynamic excessive risk visitor group detection method and system |
| WO2019114481A1 (en) * | 2017-12-13 | 2019-06-20 | 腾讯科技（深圳）有限公司 | Cluster type recognition method, apparatus, electronic apparatus, and storage medium |
| CN108564460B (en) * | 2018-01-12 | 2020-10-30 | 阳光财产保险股份有限公司 | Real-time fraud detection method and device in internet credit scene |
| CN108564460A (en) * | 2018-01-12 | 2018-09-21 | 阳光财产保险股份有限公司 | Real-time fraud detection method under internet credit scene and device |
| CN111164633A (en) * | 2018-05-31 | 2020-05-15 | 重庆小雨点小额贷款有限公司 | Method and device for adjusting grading card model, server and storage medium |
| CN111164633B (en) * | 2018-05-31 | 2024-01-05 | 重庆小雨点小额贷款有限公司 | Method and device for adjusting scoring card model, server and storage medium |
| CN109670961A (en) * | 2018-09-26 | 2019-04-23 | 深圳壹账通智能科技有限公司 | Management method, device, terminal device and computer readable storage medium after throwing |
| CN110263802A (en) * | 2019-04-04 | 2019-09-20 | 平安科技（深圳）有限公司 | Credit data analysing method and relevant device based on Density Clustering |
| CN110263802B (en) * | 2019-04-04 | 2023-07-25 | 平安科技（深圳）有限公司 | Credit data analysis method based on density clustering and related equipment |
| CN112561635A (en) * | 2020-12-11 | 2021-03-26 | 永安行科技股份有限公司 | Shared article renting and returning method and system based on credit system |
| CN112884571A (en) * | 2021-02-26 | 2021-06-01 | 中库(北京)数据系统有限公司 | Method and device for determining credit risk level |
| CN113362120A (en) * | 2021-08-11 | 2021-09-07 | 北京达佳互联信息技术有限公司 | Group determination method and device, electronic equipment and computer readable storage medium |
| CN113643125A (en) * | 2021-08-30 | 2021-11-12 | 天元大数据信用管理有限公司 | Credit line measuring and calculating method, equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105894372B (en) | 2018-03-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105894372A (en) | Method and device for predicting group credit | |
| Mustafa et al. | Modelling built-up expansion and densification with multinomial logistic regression, cellular automata and genetic algorithm | |
| CN110992169B (en) | Risk assessment method, risk assessment device, server and storage medium | |
| Sadjadi et al. | A bootstrapped robust data envelopment analysis model for efficiency estimating of telecommunication companies in Iran | |
| Cigizoglu | Application of generalized regression neural networks to intermittent flow forecasting and estimation | |
| CN109189867A (en) | Relationship discovery method, apparatus and storage medium based on Corporate Intellectual map | |
| Xia et al. | Modeling urban growth in a metropolitan area based on bidirectional flows, an improved gravitational field model, and partitioned cellular automata | |
| CN112508118B (en) | Target object behavior prediction method aiming at data offset and related equipment thereof | |
| CN106126741B (en) | A kind of electric network information secure and trusted work system based on big data | |
| CN108734327A (en) | A kind of data processing method, device and server | |
| CN108198040A (en) | Educational resource distribution method, device, computer equipment and storage medium | |
| Osman et al. | An integrated land use change model to simulate and predict the future of greater Cairo metropolitan region | |
| CN109063759A (en) | A kind of neural network structure searching method applied to the more attribute forecasts of picture | |
| CN108038734B (en) | Urban commercial facility spatial distribution detection method and system based on comment data | |
| Sarica et al. | Spatio-temporal dynamics of flood exposure in Shenzhen from present to future | |
| CN109840676A (en) | Air control method, apparatus, computer equipment and storage medium based on big data | |
| US10444062B2 (en) | Measuring and diagnosing noise in an urban environment | |
| Cai et al. | Discovering regions of anomalous spatial co-locations | |
| CN101295388A (en) | Credit estimation method and system | |
| CN110276387A (en) | A kind of generation method and device of model | |
| Lagarias | Exploring land use policy scenarios with the use of a cellular automata-based model: urban sprawl containment and sustainable development in Thessaloniki | |
| CN115033456A (en) | Method and device for monitoring performance of front end of intranet, computer equipment and storage medium | |
| CN107093149A (en) | Online friend relation strength assessment method and system | |
| CN113850669A (en) | User grouping method and device, computer equipment and computer readable storage medium | |
| CN110633890A (en) | Land utilization efficiency judgment method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |