CN109711696A - Enterprise's methods of marking, device, medium and electronic equipment - Google Patents
Enterprise's methods of marking, device, medium and electronic equipment Download PDFInfo
- Publication number
- CN109711696A CN109711696A CN201811554838.XA CN201811554838A CN109711696A CN 109711696 A CN109711696 A CN 109711696A CN 201811554838 A CN201811554838 A CN 201811554838A CN 109711696 A CN109711696 A CN 109711696A
- Authority
- CN
- China
- Prior art keywords
- enterprise
- characteristic
- data
- evaluated
- rating model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present embodiments relate to data analysis technique fields, provide a kind of enterprise's methods of marking, device, medium and electronic equipment, this method comprises: obtaining the characteristic to be evaluated about Target Enterprise, wherein, the characteristic to be evaluated be credit feature data, it is character of innovation data, operation characteristic, history debt characteristic, at least a kind of in qualification of paying taxes characteristic；The fitting coefficient of every category feature parameter is obtained, and enterprise's Rating Model is determined according to the fitting coefficient and corresponding characteristic parameter of the every category feature parameter got；The characteristic to be evaluated is inputted into enterprise's Rating Model, obtains scoring of the output of enterprise's Rating Model as the Target Enterprise.Technical solution provided in this embodiment evaluates the character pair of Target Enterprise by way of scoring, can provide for user and provide the evaluation data of different angle to Target Enterprise, while improving the confidence level of enterprise's scoring.
Description
Technical field
The present invention relates to data analysis technique fields, in particular to a kind of enterprise's methods of marking, enterprise's scoring dress
It sets, computer-readable medium and electronic equipment.
Background technique
With economic continuous development, contact between enterprise with exchange it is more and more.It will be apparent that mutual between enterprise
Understanding is very helpful to the development of each enterprise.But the evaluation in current valuation of enterprise method to a certain enterprise
Mode is relatively simple.For example, only evaluate from single specific concept enterprise, so that other enterprises can not have it
Comprehensive cognition can not comprehensively understand this enterprise.Meanwhile use that current valuation of enterprise method is generally only summarized " excellent,
It is good, in, it is poor " etc. recapitulative, degree word enterprise is evaluated, cause user only can be right from above-mentioned degree word
Target Enterprise has a rough evaluation.
It should be noted that information is only used for reinforcing the reason to background of the invention disclosed in above-mentioned background technology part
Solution, therefore may include the information not constituted to the prior art known to persons of ordinary skill in the art.
Summary of the invention
The embodiment of the present invention is designed to provide a kind of enterprise's methods of marking, enterprise's scoring apparatus, computer-readable Jie
Matter and electronic equipment, so overcome at least to a certain extent the evaluation method of the method for evaluation goal enterprise in the prior art compared with
Single problem.
Other characteristics and advantages of the invention will be apparent from by the following detailed description, or partially by the present invention
Practice and acquistion.
According to a first aspect of the embodiments of the present invention, a kind of enterprise's methods of marking is provided, the application program is arranged in
In goal systems, comprising:
Obtain the characteristic to be evaluated about Target Enterprise, wherein the characteristic to be evaluated is credit feature number
According at least a kind of in, character of innovation data, operation characteristic, history debt characteristic, qualification of paying taxes characteristic；
Obtain the fitting coefficient of every category feature parameter, and according to the fitting coefficient of the every category feature parameter got with
And corresponding characteristic parameter determines enterprise's Rating Model；And
The characteristic to be evaluated is inputted into enterprise's Rating Model, the output for obtaining enterprise's Rating Model is made
For the scoring of the Target Enterprise.
In some embodiments of the invention, the fitting coefficient of every category feature parameter is obtained, comprising:
Feature extraction is carried out to the historical data of many enterprises got and obtains the first training set, first training set
It include: at least a kind of and every enterprise score data in the multiclass feature data of every enterprise；
Based on first training set, the first Logic Regression Models are trained by iterative algorithm, and according to training
The first Logic Regression Models afterwards obtain the fitting coefficient of every category feature parameter.
In some embodiments of the invention, after the fitting coefficient for obtaining every category feature parameter, further includes:
Feature extraction is carried out to the historical data of many enterprises got and obtains the second training set, second training set
It include: that the fitting coefficient of at least a kind of, described every category feature parameter in the multiclass feature data of every enterprise and every look forward to
The score data of industry；
Based on second training set, the second Logic Regression Models are trained by iterative algorithm, to obtain enterprise
Rating Model.
In some embodiments of the invention, before obtaining enterprise's Rating Model, further includes:
The verifying collection to second Logic Regression Models is obtained, the verifying collection includes: the multiclass feature of every enterprise
The fitting coefficient of at least a kind of, every category feature parameter in data and the score data of every enterprise；
The second Logic Regression Models after training are verified by verifying collection, and obtain verification result；
If the verification result meets preset verification condition, using the second Logic Regression Models after training as described in
Enterprise's Rating Model.
In some embodiments of the invention, enterprise's methods of marking, further includes:
If the verification result is unsatisfactory for preset verification condition, continue to train described second to state Logic Regression Models,
Until the verification result obtained after verifying to the second Logic Regression Models after training meets preset verification condition.
In some embodiments of the invention, the preset verification condition includes: in accuracy rate, recall rate and AUC
It is at least one.
In some embodiments of the invention, the characteristic to be evaluated about Target Enterprise is obtained, comprising:
Feature is carried out by the empirical data of valuation of enterprise, enterprise's ranking and the accumulation of business event side of issuing official to mention
It takes, obtains the characteristic to be evaluated about Target Enterprise.
According to a second aspect of the embodiments of the present invention, a kind of enterprise's scoring apparatus is provided, comprising:
Characteristic to be evaluated obtains module, for obtaining the characteristic to be evaluated about Target Enterprise, wherein described
Characteristic to be evaluated is credit feature data, character of innovation data, operation characteristic, history debt characteristic, pays taxes
At least one of qualification characteristic；
Enterprise's Rating Model determining module, for obtaining the fitting coefficient of every category feature parameter, and according to the institute got
The fitting coefficient and corresponding characteristic parameter for stating every category feature parameter determine enterprise's Rating Model；
Grading module obtains enterprise's scoring mould for the characteristic to be evaluated to be inputted enterprise's Rating Model
Scoring of the output of type as the Target Enterprise.
According to a third aspect of the embodiments of the present invention, a kind of computer-readable medium is provided, computer is stored thereon with
Program realizes enterprise's methods of marking as described in first aspect in above-described embodiment when described program is executed by processor.
According to a fourth aspect of the embodiments of the present invention, a kind of electronic equipment is provided, comprising: one or more processors；
Storage device, for storing one or more programs, when one or more of programs are held by one or more of processors
When row, so that one or more of processors realize enterprise's methods of marking as described in first aspect in above-described embodiment.
Technical solution provided in an embodiment of the present invention can include the following benefits:
In the technical solution provided by some embodiments of the present invention, on the one hand, by obtaining about Target Enterprise
Credit feature data, character of innovation data, operation characteristic, history debt characteristic, in qualification of paying taxes characteristic
It is at least a kind of, and as characteristic to be evaluated, characteristic enterprise to be evaluated Rating Model is come with further true
Enterprise set the goal about the corresponding scoring of above-mentioned characteristic to be evaluated, and then to the correspondence of Target Enterprise by way of scoring
Characteristic evaluating can provide for user and provide the evaluation data of different angle to Target Enterprise, meanwhile, the mode of scoring can be with
A more specific valuation of enterprise structure is provided for user, is compared different enterprises convenient for user.On the other hand, with
All kinds of characteristics and fitting coefficient corresponding with all kinds of characteristics pass through as the training sample for determining enterprise's Rating Model
Valuation of enterprise model is determined based on big data training machine learning model, is conducive to improve the accuracy to Target Enterprise scoring,
Meanwhile different fitting coefficients characterizes weight of all kinds of characteristics during training enterprise's Rating Model, is conducive into one
Improve the confidence level of the enterprise's scoring obtained by valuation of enterprise model in step ground.
It should be understood that above general description and following detailed description be only it is exemplary and explanatory, not
It can the limitation present invention.
Detailed description of the invention
The drawings herein are incorporated into the specification and forms part of this specification, and shows and meets implementation of the invention
Example, and be used to explain the principle of the present invention together with specification.It should be evident that the accompanying drawings in the following description is only the present invention
Some embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.In the accompanying drawings:
Fig. 1 shows the flow diagram of enterprise's methods of marking of embodiment according to the present invention；
Fig. 2 shows the flow diagrams of the method for the determination fitting coefficient of embodiment according to the present invention；
Fig. 3 shows the flow diagram of the method for the determination enterprise Rating Model of embodiment according to the present invention；
Fig. 4 shows the structural schematic diagram of enterprise's scoring apparatus of embodiment according to the present invention；
Fig. 5 shows the structural schematic diagram for being suitable for the computer system for the electronic equipment for being used to realize the embodiment of the present invention.
Specific embodiment
Example embodiment is described more fully with reference to the drawings.However, example embodiment can be with a variety of shapes
Formula is implemented, and is not understood as limited to example set forth herein；On the contrary, thesing embodiments are provided so that the present invention will more
Fully and completely, and by the design of example embodiment comprehensively it is communicated to those skilled in the art.
In addition, described feature, structure or characteristic can be incorporated in one or more implementations in any suitable manner
In example.In the following description, many details are provided to provide and fully understand to the embodiment of the present invention.However,
It will be appreciated by persons skilled in the art that technical solution of the present invention can be practiced without one or more in specific detail,
Or it can be using other methods, constituent element, device, step etc..In other cases, it is not shown in detail or describes known side
Method, device, realization or operation are to avoid fuzzy each aspect of the present invention.
Block diagram shown in the drawings is only functional entity, not necessarily must be corresponding with physically separate entity.
I.e., it is possible to realize these functional entitys using software form, or realized in one or more hardware modules or integrated circuit
These functional entitys, or these functional entitys are realized in heterogeneous networks and/or processor device and/or microcontroller device.
Flow chart shown in the drawings is merely illustrative, it is not necessary to including all content and operation/step,
It is not required to execute by described sequence.For example, some operation/steps can also decompose, and some operation/steps can close
And or part merge, therefore the sequence actually executed is possible to change according to the actual situation.
For the evaluation method to Target Enterprise that the prior art provides, there are following defects.One is: to a certain enterprise
Evaluation method it is relatively simple.For example, only evaluating from single specific concept enterprise, so that other enterprises can not be right
It has a comprehensive cognition, can not comprehensively understand this enterprise.Another kind is: current valuation of enterprise method is generally only summarized
It uses degree words such as " excellent middle differences " to evaluate enterprise, causes user that can look forward to from above-mentioned degree word target to the greatest extent
Industry has a rough evaluation.As it can be seen that the prior art provide the evaluation method to Target Enterprise there are evaluation method it is single and
Evaluate exact degree problem to be improved.
A kind of enterprise's methods of marking, enterprise's scoring apparatus, computer-readable medium and electronics provided in an embodiment of the present invention
Equipment, and then the above-mentioned problems in the prior art is overcome at least to a certain extent.Fig. 1 shows reality according to the present invention
The flow diagram for applying enterprise's methods of marking of example, with reference to Fig. 1, enterprise's methods of marking, comprising:
Step S101 obtains the characteristic to be evaluated about Target Enterprise, wherein the characteristic to be evaluated is letter
With characteristic, character of innovation data, operation characteristic, history debt characteristic, in qualification of paying taxes characteristic extremely
Few one kind；
Step S102 obtains the fitting coefficient of every category feature parameter, and according to the every category feature parameter got
Fitting coefficient and corresponding characteristic parameter determine enterprise's Rating Model；And
The characteristic to be evaluated is inputted enterprise's Rating Model, obtains enterprise's Rating Model by step S103
Export the scoring as the Target Enterprise.
In the technical solution of embodiment illustrated in fig. 1, by obtaining about the credit feature data of Target Enterprise, character of innovation
It is data, operation characteristic, history debt characteristic, at least a kind of in qualification of paying taxes characteristic, and as to
Evaluating characteristic data, characteristic to be evaluated come determine Target Enterprise about the corresponding scoring of above-mentioned characteristic to be evaluated, into
And the character pair of Target Enterprise is evaluated by way of scoring, it can be provided for user and provide different angles to Target Enterprise
The evaluation data of degree, meanwhile, the mode of scoring can provide a more specific valuation of enterprise structure for user, be convenient for user
Different enterprises is compared.On the other hand, made with all kinds of characteristics and fitting coefficient corresponding with all kinds of characteristics
For determine enterprise's Rating Model training sample, by determining valuation of enterprise model based on big data training machine learning model,
Be conducive to improve the accuracy to Target Enterprise scoring, meanwhile, different fitting coefficients characterizes all kinds of characteristics and looks forward in training
Weight during industry Rating Model is conducive to further improve the credible of the enterprise's scoring obtained by valuation of enterprise model
Degree.
The specific implementation of each step shown in Fig. 1 is described in detail below:
In the exemplary embodiment, in step s101, the tool of the characteristic to be evaluated about Target Enterprise is obtained
Body implementation are as follows: carried out by the empirical data of valuation of enterprise, enterprise's ranking and the accumulation of business event side issued to official
Feature extraction obtains the characteristic to be evaluated about Target Enterprise.By the accuracy for guaranteeing characteristic to be evaluated
To further increase the accuracy of enterprise's scoring, to provide enterprise's scoring of closer to reality for user.
In the exemplary embodiment, in step s 102, multiple fitting coefficients, and a fitting coefficient and one kind are obtained
Characteristic parameter is corresponding, for the sample after being multiplied with character pair data as training enterprise's Rating Model.
In the exemplary embodiment, Fig. 2 shows the methods of the acquisition fitting coefficient of embodiment according to the present invention
Flow diagram.With reference to Fig. 2, the method provided in this embodiment for obtaining fitting parameter, comprising:
Step S201 carries out feature extraction to the historical data of many enterprises got and obtains the first training set, described
First training set includes: at least a kind of and every enterprise score data in the multiclass feature data of every enterprise；With
And
Step S202 is based on first training set, is trained by iterative algorithm to the first Logic Regression Models, and
The fitting coefficient of every category feature parameter is obtained according to the first Logic Regression Models after training.
In the exemplary embodiment, in step s 201, the specific acquisition modes of the historical data of many enterprises can also
With are as follows: feature extraction is carried out by the empirical data of valuation of enterprise, enterprise's ranking and the accumulation of business event side issued to official,
Obtain the characteristic to be evaluated about Target Enterprise.By guarantee many enterprises historical data accuracy with into one
Step improves the accuracy of fitting parameter, the accuracy for improving enterprise's scoring is ultimately facilitated, to provide closer to reality for user
Enterprise scoring.
In the exemplary embodiment, at least to the above-mentioned M category feature data of the scoring of every enterprise and this enterprise, family
It is a kind of related.Specifically, can be characterized between each feature and enterprise's scoring according to by the corresponding fitting coefficient of each feature
Relationship.Further, the M category feature data { X that feature extraction obtains is carried out with the historical data of enterprise, N family1,X2,…,Xm}
And the score data of enterprise, N family is sample, is trained to the first regression model, obtains the fitting parameter { A of output1,
A2,…,Am}。
In the exemplary embodiment, a kind of method of enterprise's Rating Model in determining step S103 is provided.Fig. 3 is shown
The flow diagram of the method for determining enterprise's Rating Model.With reference to Fig. 3, the method for determining enterprise's Rating Model includes step
S301- step S306.
In step S301, feature extraction is carried out to the historical data of many enterprises got and obtains the second training set,
Second training set includes: the fitting of at least a kind of, described every category feature parameter in the multiclass feature data of every enterprise
The score data of coefficient and every enterprise.
Illustratively, fitting coefficient therein is that the method provided by embodiment illustrated in fig. 2 obtains.
In step s 302, it is based on second training set, the second Logic Regression Models are instructed by iterative algorithm
Practice, to obtain enterprise's Rating Model.
Illustratively, the second Logic Regression Models are trained by iterative algorithm, to obtain enterprise's Rating Model
Specific implementation may is that
Firstly, to the historical data of the enterprise, N family got, and further progress feature extraction obtains the second training set.
Specifically, the second training set includes: the M of the 1st enterprise1Category feature data { X1,X2,…,XM1}+corresponding fitting coefficient { A1,
A2,…,AM1..., i-th enterprise MiCategory feature data { X1,X2,…,XMi}+corresponding fitting coefficient { A1,A2,…,
AMi..., enterprise, N family MNCategory feature data { X1,X2,…,XMN}+corresponding fitting coefficient { A1,A2,…,AMN}；And every family
Score data { the Y of enterprise1,Y2,…,Yi,…,YN, wherein " characteristic+fitting coefficient " of i-th enterprise with it is corresponding
Score data YiFoundation has corresponding relationship, and i value is more than or equal to 1 and is less than or equal to N.Further, according to above-mentioned second training
Collect one Logic Regression Models (being denoted as the second Logic Regression Models) of training；By iterative process at least once (i.e. training process)
Whether the second Logic Regression Models after verifying training meet verification condition afterwards.
Illustratively, for i-th enterprise, characteristic { X1,X2,…,XMi, fitting coefficient { A1,A2,…,AMiAnd
The score data Y of i-th enterpriseiRelationship can indicate are as follows: Yi=A1*X1+A2*X2+…+AMi*XMi。
It illustratively, can be with after step S302 is trained the second Logic Regression Models by the second training set
Further the second Logic Regression Models after training are verified.Specific verification process such as step S303- step S306
It is shown:
In step S303, the verifying collection to second Logic Regression Models is obtained, the verifying collection includes: every enterprise
The fitting coefficient of at least a kind of, every category feature parameter in the multiclass feature data of industry and the score data of every enterprise.
In step s 304, the second Logic Regression Models after training are verified by verifying collection, and obtained
Verification result.
In step S305, judge that the verification result meets preset verification condition.
If the verification result meets preset verification condition, S306 is thened follow the steps: the second logic after training is returned
Return model as enterprise's Rating Model.
If the verification result is unsatisfactory for preset verification condition, step S302 is continued to execute.Continue described in training
Second Logic Regression Models, until the verification result obtained after verifying to the second Logic Regression Models after training meets in advance
If verification condition.
Illustratively, above-mentioned verification condition can be at least one of accuracy rate, recall rate and AUC.It is tested by following
Card process can reduce the generation of poor fitting, over-fitting, to further increase the accuracy of enterprise's scoring.
In the exemplary embodiment, characteristic verifying concentrated substitutes into the second logic for completing at least an iteration
Regression model determines that the specific implementation of verification result includes:
Firstly, after substituting into completion at least the second Logic Regression Models of an iteration according to the characteristic that verifying is concentrated
Output data obtains following: true positives TP, true negative TN, under-referral FN and pseudo- positive FP.Wherein, TP is to utilize completion at least one
In the characteristic that second Logic Regression Models of secondary iteration concentrate verifying positive class belong to after judging be still positive class number
Mesh, TN judge negative class in the characteristic of verifying concentration using the second Logic Regression Models for completing at least an iteration
Belong to afterwards be still negative class number, FN using complete at least an iteration the second Logic Regression Models to verifying concentrate feature
In data negative class belong to after being judged be positive class number, FP utilizes the second Logic Regression Models for completing at least an iteration
In the characteristic concentrated to verifying positive class belong to after judging be negative class number.Positive class and negative class refer to manually to training
The two categories of sample characteristics data mark, i.e., manually mark some sample characteristics data and belong to specific class, then the sample is special
Sign data belong to positive class, and the sample characteristics data for being not belonging to the certain kinds then belong to negative class.
Secondly, according to true positives TP, true negative TN, under-referral FN and puppet positive FP calculating completion at least once iteration the
The verification result of two Logic Regression Models.
Verifying index is introduced by taking accuracy rate, recall rate as an example, specific:
Accuracy rate p and recall rate r are calculated separately according to formula one and formula two；
P=TP/ (TP+FP) formula one,
R=TP/ (TP+FN) formula two.
If the second Logic Regression Models for completing at least an iteration to x are verified result after verifying are as follows:
Accuracy rate verification result p1, p2 ..., px and recall rate verification result r1, r2 ..., rx.
Verify the corresponding setting condition of index are as follows: accuracy rate verification result is greater than p ' and then imposes a condition to meet accuracy rate,
Otherwise it is unsatisfactory for accuracy rate setting condition and recall rate verification result is greater than r ' then to meet recall rate setting condition, otherwise
It is unsatisfactory for recall rate setting condition.
In the exemplary embodiment, it in the case where verification result satisfaction verifying index corresponding setting condition, is tested
Second Logic Regression Models of card then can be used as enterprise's Rating Model；When verification result is unsatisfactory for imposing a condition, then to quilt
Second Logic Regression Models of verifying continue iteration until the verification result of second Logic Regression Models meets setting condition.
It in the exemplary embodiment, can be with when judging whether verification result meets the verifying corresponding setting condition of index
Using accuracy rate or recall rate as verifying index, i.e., accuracy/recall rate, which meets, imposes a condition；It can also be simultaneously with accurate
Simultaneously as verifying index, i.e., accuracy and recall rate, which meet, imposes a condition for rate and recall rate.
It should be noted that specific verification mode is formulated according to actual needs, it is not limited to the above accuracy rate and/or calls together
The rate of returning is verified as verifying index.
In the exemplary embodiment, verifying index can also be AUC, specific:
In the exemplary embodiment, puppet positive rate FPR and true positive rate TPR is determined using formula three and formula four,
FPR=FP/ (FP+TN) formula three,
TPR=TP/ (TP+FN) formula four.
Further, using FPR as abscissa, TPR is ordinate, draws Receiver operating curve (Receiver
Operating Characteristic curve, abbreviation ROC curve).Wherein, ROC curve is the feature of each index obtained
Curve for showing the relationship between each index, and further calculates area AUC under ROC curve.ROC curve is to obtain
The indicatrix of each index, for showing the relationship between each index, area under AUC, that is, ROC curve, AUC is bigger, then model
Predictive value is higher, and then can be verified by AUC to the second Logic Regression Models are completed.And by evaluation result be AUC value
Maximum model for receiving characteristic to be evaluated, and exports the scoring to Target Enterprise as enterprise's Rating Model.
It can be also used for appointing M category feature data by the Rating Model that the technical solution of embodiment illustrated in fig. 3 provides
A kind of or several predictions scored.
Illustratively, characteristic X is carried out to Target Enterprise2(character of innovation data) corresponding enterprise's scoring carries out pre-
It surveys, to obtain about " character of innovation data X2" enterprise score Y1.It then can be by the character of innovation data X of Target Enterprise2And this
The corresponding fitting parameter A of category feature data2It is input to above-mentioned enterprise's Rating Model, output is enterprise scoring Y1。YsFor concentrating
The innovation ability of enterprise is embodied, mainly with data dimensions such as trademark patent, copyright, software copyright, talent of high caliber's deposits
It is assessed.
Illustratively, characteristic X is carried out simultaneously to Target Enterprise1(credit feature data) and characteristic X3(operation is special
Sign data) scoring of corresponding enterprise predicted, to obtain about " credit feature data X1+ operation characteristic X3" enterprise
Score Y2.It then can be by the 1. credit feature data X of Target Enterprise1Fitting parameter A corresponding with such characteristic1And 2.
Run characteristic X3Fitting parameter A corresponding with such characteristic3It is input to above-mentioned enterprise's Rating Model, output is to look forward to
Industry scoring Y2。Y2Both it is used to reflect the reflection passing credit standing of enterprise, is mainly related to from the administration of justice and tell, administrative penalty, debt paying ability, go through
The dimensions assessments such as history debt rating, qualification of paying taxes grading, but reflection presents the day-to-day operations ability of enterprise, mainly uses the talent
The data dimensions such as recruitment, brand image, bidding, external cooperation are assessed.Example is comprehensive to one kind of Target Enterprise at this time
Close assessment.
As it can be seen that enterprise's Rating Model that the technical solution provided through this embodiment obtains, is suitable for any multiclass feature
The case where parameter, the feature classification acquired for a certain Target Enterprise is more, then the grade form that enterprise's Rating Model determines
Sign is more comprehensive, is predicted or is assessed suitable for integrally scoring Target Enterprise, provided the user with and comment the entirety of Target Enterprise
Value.Meanwhile when the characteristic parameter for obtaining a certain Target Enterprise is not easy or is inconvenient or user wishes to obtain a certain Target Enterprise
The corresponding scoring of certain a kind of/a few category feature parameter be that can be assessed the feature of its unitary class using this model, increase
Applicability.Meanwhile user can be made to be assessed from various dimensions Target Enterprise according to oneself demand.
The device of the invention embodiment introduced below can be used for executing the above-mentioned enterprise's methods of marking of the present invention.
Fig. 4 shows the structural schematic diagram of enterprise's scoring apparatus of embodiment according to the present invention.With reference to Fig. 4, enterprise is commented
Separating device 400, comprising: characteristic to be evaluated obtains module 401, enterprise's Rating Model determining module 402 and grading module
403。
Wherein, characteristic to be evaluated obtains module 401 and is used to obtain the characteristic to be evaluated about Target Enterprise,
In, the characteristic to be evaluated is credit feature data, character of innovation data, operation characteristic, history debt characteristic
According at least one of, qualification of paying taxes characteristic；
Enterprise's Rating Model determining module 402 is used to obtain the fitting coefficient of every category feature parameter, and according to getting
The fitting coefficient and corresponding characteristic parameter of every category feature parameter determine enterprise's Rating Model；
Grading module 403 is used to the characteristic to be evaluated inputting enterprise's Rating Model, obtains enterprise's scoring
Scoring of the output of model as the Target Enterprise.
In the exemplary embodiment, enterprise's Rating Model determining module 402 includes: first acquisition unit and first
Training unit.
Wherein, first acquisition unit is used to carry out feature extraction to the historical data of many enterprises got to obtain first
Training set, first training set include: commenting at least a kind of in the multiclass feature data of every enterprise and every enterprise
Divided data；And first training unit be used for be based on first training set, by iterative algorithm to the first Logic Regression Models
It is trained, and obtains the fitting coefficient of every category feature parameter according to the first Logic Regression Models after training.
In the exemplary embodiment, enterprise's Rating Model determining module 402 includes: second acquisition unit and second
Training unit.
After the fitting coefficient that the first training unit obtains every category feature parameter, the second acquisition unit is used for obtaining
The historical data of many enterprises got carries out feature extraction and obtains the second training set, and second training set includes: every enterprise
The fitting coefficient of at least a kind of, described every category feature parameter in the multiclass feature data of industry and the scoring number of every enterprise
According to；And second training unit is used to be based on second training set, by iterative algorithm to the second Logic Regression Models
It is trained, to obtain enterprise's Rating Model.
In the exemplary embodiment, enterprise's Rating Model determining module 402 includes: third acquiring unit and verifying
Unit.
Wherein, the third acquiring unit is used to obtain the verifying collection to second Logic Regression Models, the verifying
Collection includes: at least one kind, the fitting coefficient of every category feature parameter and every enterprise in the multiclass feature data of every enterprise
Score data；And the authentication unit is used to carry out the second Logic Regression Models after training by the verifying collection
Verifying, and obtain verification result.
In the exemplary embodiment, if the verification result meets preset verification condition, enterprise's scoring mould
Type determining module 402 is also used to the second Logic Regression Models after training as enterprise's Rating Model.
In the exemplary embodiment, if the verification result is unsatisfactory for preset verification condition, second training
Module continues to train described second to state Logic Regression Models, until after being verified to the second Logic Regression Models after training
To verification result meet preset verification condition.
In the exemplary embodiment, the preset verification condition may is that in accuracy rate, recall rate and AUC extremely
Few one kind.
In the exemplary embodiment, the characteristic to be evaluated obtains module 401 and is specifically used for: by sending out official
Valuation of enterprise, enterprise's ranking and the empirical data of business event side's accumulation of cloth carry out feature extraction, obtain described about target
The characteristic to be evaluated of enterprise.
Each functional module and above-mentioned enterprise's methods of marking due to enterprise's scoring apparatus of example embodiments of the present invention
Example embodiment the step of it is corresponding, therefore for undisclosed details in apparatus of the present invention embodiment, please refer in the present invention
The embodiment for the enterprise's methods of marking stated.
Below with reference to Fig. 5, it illustrates the computer systems 500 for the electronic equipment for being suitable for being used to realize the embodiment of the present invention
Structural schematic diagram.The computer system 500 of electronic equipment shown in Fig. 5 is only an example, should not be to the embodiment of the present invention
Function and use scope bring any restrictions.
As shown in figure 5, computer system 500 includes central processing unit (CPU) 501, it can be read-only according to being stored in
Program in memory (ROM) 502 or be loaded into the program in random access storage device (RAM) 503 from storage section 508 and
Execute various movements appropriate and processing.In RAM 503, it is also stored with various programs and data needed for system operatio.CPU
501, ROM 502 and RAM 503 is connected with each other by bus 504.Input/output (I/O) interface 505 is also connected to bus
504。
I/O interface 505 is connected to lower component: the importation 506 including keyboard, mouse etc.；It is penetrated including such as cathode
The output par, c 507 of spool (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.；Storage section 508 including hard disk etc.；
And the communications portion 509 of the network interface card including LAN card, modem etc..Communications portion 509 via such as because
The network of spy's net executes communication process.Driver 510 is also connected to I/O interface 505 as needed.Detachable media 511, such as
Disk, CD, magneto-optic disk, semiconductor memory etc. are mounted on as needed on driver 510, in order to read from thereon
Computer program be mounted into storage section 508 as needed.
Particularly, according to an embodiment of the invention, may be implemented as computer above with reference to the process of flow chart description
Software program.For example, the embodiment of the present invention includes a kind of computer program product comprising be carried on computer-readable medium
On computer program, which includes the program code for method shown in execution flow chart.In such reality
It applies in example, which can be downloaded and installed from network by communications portion 509, and/or from detachable media
511 are mounted.When the computer program is executed by central processing unit (CPU) 501, executes and limited in the system of the application
Above-mentioned function.
It should be noted that computer-readable medium shown in the present invention can be computer-readable signal media or meter
Calculation machine readable storage medium storing program for executing either the two any combination.Computer readable storage medium for example can be --- but not
Be limited to --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor system, device or device, or any above combination.Meter
The more specific example of calculation machine readable storage medium storing program for executing can include but is not limited to: have the electrical connection, just of one or more conducting wires
Taking formula computer disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable type may be programmed read-only storage
Device (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device,
Or above-mentioned any appropriate combination.In the present invention, computer readable storage medium can be it is any include or storage journey
The tangible medium of sequence, the program can be commanded execution system, device or device use or in connection.And at this
In invention, computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including but unlimited
In electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be that computer can
Any computer-readable medium other than storage medium is read, which can send, propagates or transmit and be used for
By the use of instruction execution system, device or device or program in connection.Include on computer-readable medium
Program code can transmit with any suitable medium, including but not limited to: wireless, electric wire, optical cable, RF etc. are above-mentioned
Any appropriate combination.
Flow chart and block diagram in attached drawing are illustrated according to the system of various embodiments of the invention, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, program segment or code of table, a part of above-mentioned module, program segment or code include one or more
Executable instruction for implementing the specified logical function.It should also be noted that in some implementations as replacements, institute in box
The function of mark can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are practical
On can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it wants
It is noted that the combination of each box in block diagram or flow chart and the box in block diagram or flow chart, can use and execute rule
The dedicated hardware based systems of fixed functions or operations is realized, or can use the group of specialized hardware and computer instruction
It closes to realize.
Being described in unit involved in the embodiment of the present invention can be realized by way of software, can also be by hard
The mode of part realizes that described unit also can be set in the processor.Wherein, the title of these units is in certain situation
Under do not constitute restriction to the unit itself.
As on the other hand, present invention also provides a kind of computer-readable medium, which be can be
Included in electronic equipment described in above-described embodiment；It is also possible to individualism, and without in the supplying electronic equipment.
Above-mentioned computer-readable medium carries one or more program, when the electronics is set by one for said one or multiple programs
When standby execution, so that the electronic equipment realizes such as above-mentioned enterprise's methods of marking as described in the examples.
For example, the electronic equipment may be implemented as shown in Figure 1: step S101 is obtained about Target Enterprise
Characteristic to be evaluated, wherein the characteristic to be evaluated is credit feature data, character of innovation data, operation characteristic
According to, it is at least a kind of in history debt characteristic, qualification of paying taxes characteristic；And step S102, by the spy to be evaluated
It levies data and inputs enterprise's Rating Model, obtain scoring of the output of enterprise's Rating Model as the Target Enterprise.
For another example, each step as shown in Figure 2 or Figure 3 may be implemented in the electronic equipment.
It should be noted that although being referred to several modules or list for acting the equipment executed in the above detailed description
Member, but this division is not enforceable.In fact, embodiment according to the present invention, it is above-described two or more
Module or the feature and function of unit can embody in a module or unit.Conversely, an above-described mould
The feature and function of block or unit can be to be embodied by multiple modules or unit with further division.
Through the above description of the embodiments, those skilled in the art is it can be readily appreciated that example described herein is implemented
Mode can also be realized by software realization in such a way that software is in conjunction with necessary hardware.Therefore, according to the present invention
The technical solution of embodiment can be embodied in the form of software products, which can store non-volatile at one
Property storage medium (can be CD-ROM, USB flash disk, mobile hard disk etc.) in or network on, including some instructions are so that a calculating
Equipment (can be personal computer, server, touch control terminal or network equipment etc.) executes embodiment according to the present invention
Method.
Those skilled in the art after considering the specification and implementing the invention disclosed here, will readily occur to of the invention its
Its embodiment.This application is intended to cover any variations, uses, or adaptations of the invention, these modifications, purposes or
Person's adaptive change follows general principle of the invention and including the undocumented common knowledge in the art of the present invention
Or conventional techniques.The description and examples are only to be considered as illustrative, and true scope and spirit of the invention are by following
Claim is pointed out.
It should be understood that the present invention is not limited to the precise structure already described above and shown in the accompanying drawings, and
And various modifications and changes may be made without departing from the scope thereof.The scope of the present invention is limited only by the attached claims.
Claims (10)
1. a kind of enterprise's methods of marking characterized by comprising
Obtain the characteristic to be evaluated about Target Enterprise, wherein the characteristic to be evaluated is credit feature data, wound
It is new feature data, operation characteristic, history debt characteristic, at least a kind of in qualification of paying taxes characteristic；
Obtain the fitting coefficient of every category feature parameter, and according to the fitting coefficient of the every category feature parameter got and right
The characteristic parameter answered determines enterprise's Rating Model；
The characteristic to be evaluated is inputted into enterprise's Rating Model, obtains the output of enterprise's Rating Model as institute
State the scoring of Target Enterprise.
2. enterprise's methods of marking according to claim 1, which is characterized in that the fitting coefficient of every category feature parameter is obtained,
Include:
Feature extraction is carried out to the historical data of many enterprises got and obtains the first training set, the first training set packet
It includes: at least a kind of and every enterprise score data in the multiclass feature data of every enterprise；
Based on first training set, the first Logic Regression Models are trained by iterative algorithm, and according to training after
First Logic Regression Models obtain the fitting coefficient of every category feature parameter.
3. enterprise's methods of marking according to claim 2, which is characterized in that in the fitting coefficient for obtaining every category feature parameter
Later, further includes:
Feature extraction is carried out to the historical data of many enterprises got and obtains the second training set, the second training set packet
It includes: the fitting coefficient of at least a kind of, described every category feature parameter in the multiclass feature data of every enterprise and every enterprise
Score data；
Based on second training set, the second Logic Regression Models are trained by iterative algorithm, to obtain enterprise's scoring
Model.
4. enterprise's methods of marking according to claim 3, which is characterized in that before obtaining enterprise's Rating Model, also wrap
It includes:
The verifying collection to second Logic Regression Models is obtained, the verifying collection includes: the multiclass feature data of every enterprise
In at least a kind of, every category feature parameter fitting coefficient and every enterprise score data；
The second Logic Regression Models after training are verified by verifying collection, and obtain verification result；
If the verification result meets preset verification condition, using the second Logic Regression Models after training as the enterprise
Rating Model.
5. enterprise's methods of marking according to claim 4, which is characterized in that further include:
If the verification result is unsatisfactory for preset verification condition, continue to train described second to state Logic Regression Models, until
The verification result obtained after verifying to the second Logic Regression Models after training meets preset verification condition.
6. enterprise's methods of marking according to claim 4 or 5, which is characterized in that the preset verification condition includes: standard
At least one of true rate, recall rate and AUC.
7. enterprise's methods of marking according to any one of claims 1 to 5, which is characterized in that obtain about Target Enterprise
Characteristic to be evaluated, comprising:
Feature extraction is carried out by the empirical data of valuation of enterprise, enterprise's ranking and the accumulation of business event side issued to official,
Obtain the characteristic to be evaluated about Target Enterprise.
8. a kind of enterprise's scoring apparatus characterized by comprising
Characteristic to be evaluated obtains module, for obtaining the characteristic to be evaluated about Target Enterprise, wherein described to be evaluated
Valence characteristic is credit feature data, character of innovation data, operation characteristic, history debt characteristic, qualification of paying taxes
At least one of characteristic；
Enterprise's Rating Model determining module, it is described every for obtaining the fitting coefficient of every category feature parameter, and according to what is got
The fitting coefficient of category feature parameter and corresponding characteristic parameter determine enterprise's Rating Model；
Grading module obtains enterprise's Rating Model for the characteristic to be evaluated to be inputted enterprise's Rating Model
Export the scoring as the Target Enterprise.
9. a kind of computer-readable medium, is stored thereon with computer program, which is characterized in that described program is executed by processor
Enterprise methods of marking of the Shi Shixian as described in any one of claims 1 to 7.
10. a kind of electronic equipment characterized by comprising
One or more processors；
Storage device, for storing one or more programs, when one or more of programs are by one or more of processing
When device executes, so that one or more of processors realize the scoring side, enterprise as described in any one of claims 1 to 7
Method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811554838.XA CN109711696A (en) | 2018-12-19 | 2018-12-19 | Enterprise's methods of marking, device, medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811554838.XA CN109711696A (en) | 2018-12-19 | 2018-12-19 | Enterprise's methods of marking, device, medium and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109711696A true CN109711696A (en) | 2019-05-03 |
Family
ID=66255858
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811554838.XA Pending CN109711696A (en) | 2018-12-19 | 2018-12-19 | Enterprise's methods of marking, device, medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109711696A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109636242A (en) * | 2019-01-03 | 2019-04-16 | 深圳壹账通智能科技有限公司 | Enterprise's methods of marking, device, medium and electronic equipment |
| CN110544050A (en) * | 2019-09-12 | 2019-12-06 | 南京岳智信息技术有限公司 | Scientific and technological enterprise innovation capability evaluation method based on machine learning |
| CN111967788A (en) * | 2020-08-27 | 2020-11-20 | 天津大学 | Target enterprise determination method and device, first electronic equipment and storage medium |
| JP2022035965A (en) * | 2020-08-20 | 2022-03-04 | 株式会社日立製作所 | Intelligent supplier managing system and intelligent supplier managing method |
| CN115204717A (en) * | 2022-07-27 | 2022-10-18 | 海南锦赟安全技术服务有限公司 | Security level classification method, device, equipment and readable storage medium |
-
2018
- 2018-12-19 CN CN201811554838.XA patent/CN109711696A/en active Pending
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109636242A (en) * | 2019-01-03 | 2019-04-16 | 深圳壹账通智能科技有限公司 | Enterprise's methods of marking, device, medium and electronic equipment |
| CN110544050A (en) * | 2019-09-12 | 2019-12-06 | 南京岳智信息技术有限公司 | Scientific and technological enterprise innovation capability evaluation method based on machine learning |
| JP2022035965A (en) * | 2020-08-20 | 2022-03-04 | 株式会社日立製作所 | Intelligent supplier managing system and intelligent supplier managing method |
| JP7181334B2 (en) | 2020-08-20 | 2022-11-30 | 株式会社日立製作所 | Intelligent supplier management system and intelligent supplier management method |
| CN111967788A (en) * | 2020-08-27 | 2020-11-20 | 天津大学 | Target enterprise determination method and device, first electronic equipment and storage medium |
| CN115204717A (en) * | 2022-07-27 | 2022-10-18 | 海南锦赟安全技术服务有限公司 | Security level classification method, device, equipment and readable storage medium |
| CN115204717B (en) * | 2022-07-27 | 2024-03-22 | 海南锦赟安全技术服务有限公司 | Security level classification method, device, equipment and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109711696A (en) | Enterprise's methods of marking, device, medium and electronic equipment | |
| CN109636242A (en) | Enterprise's methods of marking, device, medium and electronic equipment | |
| CN110147456B (en) | Image classification method and device, readable storage medium and terminal equipment | |
| CN109919684A (en) | For generating method, electronic equipment and the computer readable storage medium of information prediction model | |
| US20220413839A1 (en) | Software component defect prediction using classification models that generate hierarchical component classifications | |
| CN109242496A (en) | Prediction technique, device and the computer-readable medium of the means of payment | |
| US11080608B2 (en) | Agent aptitude prediction | |
| CN109726396A (en) | Semantic matching method, device, medium and the electronic equipment of question and answer text | |
| CN111127178A (en) | Data processing method and device, storage medium and electronic equipment | |
| CN110222220A (en) | Image processing method, device, computer-readable medium and electronic equipment | |
| CN109345368A (en) | Credit estimation method, device, electronic equipment and storage medium based on big data | |
| CN108846077A (en) | Semantic matching method, device, medium and the electronic equipment of question and answer text | |
| CN109118224A (en) | Proof of work method, apparatus, medium and the electronic equipment of block chain network | |
| CN109726655A (en) | Customer service evaluation method, device, medium and equipment based on Emotion identification | |
| CN109685645A (en) | User credit methods of risk assessment and device, storage medium | |
| CN110059112A (en) | Usage mining method and device based on machine learning, electronic equipment, medium | |
| CN109857431A (en) | Code revision method and device, computer-readable medium and electronic equipment | |
| CN109344154A (en) | Data processing method, device, electronic equipment and storage medium | |
| CN109344142A (en) | Data processing method, device, electronic equipment and storage medium | |
| CN109960650A (en) | Application assessment method, apparatus, medium and electronic equipment based on big data | |
| CN109784352A (en) | A kind of method and apparatus for assessing disaggregated model | |
| CN109409619A (en) | Prediction technique, device, medium and the electronic equipment of public sentiment trend | |
| CN110009229A (en) | Supply chain management method, device, storage medium and equipment based on block chain | |
| CN109636520A (en) | Processing method, device, medium and the electronic equipment of order | |
| CN108733712A (en) | A kind of question answering system evaluation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 201, room 518000, building A, No. 1, front Bay Road, Qianhai Shenzhen Guangdong Shenzhen Hong Kong cooperation zone (Qianhai business secretary) Applicant after: Pingan Pu Hui Enterprise Management Co., Ltd. Address before: 518000 Guangdong city of Shenzhen province Qianhai Shenzhen Hong Kong cooperation zone before Bay Road No. 1 building 201 room A Applicant before: Pingan Pu Hui Enterprise Management Co., Ltd. |
|
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |