CN109376995A - Financial data methods of marking, device, computer equipment and storage medium - Google Patents
Financial data methods of marking, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109376995A CN109376995A CN201811089672.9A CN201811089672A CN109376995A CN 109376995 A CN109376995 A CN 109376995A CN 201811089672 A CN201811089672 A CN 201811089672A CN 109376995 A CN109376995 A CN 109376995A
- Authority
- CN
- China
- Prior art keywords
- financial
- enterprise
- sample
- data
- financial data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000012549 training Methods 0.000 claims abstract description 21
- 238000004590 computer program Methods 0.000 claims description 27
- 238000011156 evaluation Methods 0.000 claims description 15
- 238000010276 construction Methods 0.000 claims description 5
- 238000013473 artificial intelligence Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 7
- 238000013480 data collection Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 241001269238 Data Species 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 238000005194 fractionation Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000003466 anti-cipated effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000007596 consolidation process Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 238000009827 uniform distribution Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/12—Accounting
- G06Q40/125—Finance or payroll
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- Data Mining & Analysis (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Technology Law (AREA)
- Operations Research (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
This application involves a kind of financial data methods of marking, device, computer equipment and storage mediums.The application realizes the artificial intelligence assessment to business finance state by building model of mind.Method includes: to receive scoring request, and enterprise's mark and financial data to be scored are carried in scoring request；It searches and identifies associated industry label with enterprise；It obtains and the associated financial Rating Model of industry label, wherein, financial Rating Model is obtained based on the history financial data training for dividing good positive and negative sample data set, and positive and negative sample data set is divided with the sample judge signal foundation for including at least one financial dimension；Financial data to be scored is input in financial Rating Model, financial appraisal result is obtained；Financial appraisal result is fed back into requesting terminal.Using this method can the financial situation to enterprise precisely assessed.
Description
Technical field
This application involves field of computer technology, set more particularly to a kind of financial data methods of marking, device, computer
Standby and storage medium.
Background technique
Classification problem has widely been studied in machine learning field, most of sorting algorithm, such as decision tree, mind
Through network, it is successfully applied to multiple fields.These criteria classification algorithms usually assume that the classification of training sample is uniform
Distribution.However, non-equilibrium data collection is the data set frequently encountered in practical application.For example, in the financial number based on enterprise
When according to carrying out the building of business finance Rating Model, often good business data collection quantity is far longer than bad business data collection.Based on not
Enterprise's Rating Model that balance sample trains, which is difficult to ensure, has very high classification performance.
It is traditional to solve the problems, such as that method has used by training dataset is unbalanced: lack sampling and over-sampling, although can
It solves the problems, such as that data set is unbalanced, but also brings inevitable defect simultaneously, such as data volume loss and model over-fitting
Problem.Business finance Rating Model is constructed using the method for traditional lack sampling and over-sampling, although being able to solve fine or not enterprise
The problem of sample size great disparity, but since the defects of loss of data makes the classification performance of the model constructed bad, to enterprise
The Evaluated effect of financial situation is bad.
Summary of the invention
Based on this, it is necessary in view of the above technical problems, provide it is a kind of can the financial situation to enterprise precisely commented
Financial data methods of marking, device, computer equipment and the storage medium estimated.
A kind of financial data methods of marking, which comprises
Scoring request is received, carries enterprise's mark and financial data to be scored in the scoring request；
It searches and identifies associated industry label with the enterprise；
It obtains and the associated financial Rating Model of the industry label, wherein the finance Rating Model is to be based on dividing
What the history financial data training of good positive and negative sample data set obtained, the positive and negative sample data set is to include at least one wealth
The sample of business dimension is judged signal foundation and is divided；
The financial data to be scored is input in the financial Rating Model, financial appraisal result is obtained；
The financial appraisal result is fed back into requesting terminal.
In one embodiment, the method also includes:
Disaggregated model building request is received, carries industry label in the request；
Enterprise's financial data is obtained, the corresponding enterprise's mark of the enterprise's financial data is associated with the industry label；
It obtains sample and judges signal, the sample judges the judge signal that signal includes at least a financial dimension；
The corresponding sample of the enterprise's financial data for calculating each enterprise judges signal value；
Signal and the judge signal value are judged according to the sample, and positive negative sample stroke is carried out to the enterprise's financial data
Point, obtain positive and negative sample data set；
Predtermined category device is trained according to the positive and negative sample data set to construct financial Rating Model；
By the financial Rating Model and the industry label associated storage.
In one embodiment, it pre-defines sample described in multiple groups and judges signal；The method also includes:
The sample according to multiple groups judges the multiple financial Rating Models of signal building, and multiple finance are scored
Model and the industry label associated storage；
The acquisition includes: that acquisition is associated with the industry label with the associated financial Rating Model of the industry label
Multiple financial Rating Models；
It is described that the financial data to be scored is input in the financial Rating Model, obtain financial appraisal result packet
It includes: the financial data to be scored being separately input in multiple financial Rating Models, multiple finance scoring knots are obtained
Fruit.
In one embodiment, described that signal and the judge signal value are judged to the business finance according to the sample
Data carry out positive negative sample division, obtain positive and negative sample data set, comprising:
Signal value is judged according to the sample to be ranked up enterprise mark；
The corresponding financial data of enterprise mark is subjected to positive negative sample division according to the sequence.
In one embodiment, after the acquisition enterprise's financial data, further includes:
Each enterprise is identified into corresponding enterprise's financial data according to the time cycle and is divided into multiple groups, every group of finance number
According to a corresponding time cycle unit；
The corresponding sample of financial data described in calculating every group judges signal value, judges signal value to every group according to the sample
The financial data carries out positive negative sample division, obtains positive and negative sample data set.
In one embodiment, the method also includes:
The financial data to be scored history financial data corresponding with enterprise's label is merged, obtain to
The financial data of qualitative evaluation；
It calculates the corresponding sample of the financial data to qualitative evaluation and judges signal value；
Signal value, which is judged, according to the sample determines the corresponding financial situation label of enterprise's label；
The financial data to be scored and the financial situation label are input in the financial Rating Model, obtained
Financial appraisal result.
A kind of financial data scoring apparatus, described device include:
Score request receiving module, for receiving scoring request, enterprise's mark is carried in the scoring request and wait score
Financial data；
Industry label lookup module identifies associated industry label with the enterprise for searching；
Rating Model obtains module, for obtaining and the associated financial Rating Model of the industry label, wherein the wealth
Business Rating Model is obtained based on the history financial data training for dividing good positive and negative sample data set, the positive and negative sample data
Collection is divided with the sample judge signal foundation for including at least one financial dimension；
Model score module is obtained for the financial data to be scored to be input in the financial Rating Model
Financial appraisal result；
As a result feedback module, for the financial appraisal result to be fed back to requesting terminal.
In one embodiment, described device further include:
Request module is modeled, for receiving disaggregated model building request, carries industry label in the request；
Financial data obtains module, for obtaining enterprise's financial data, the corresponding enterprise's mark of the enterprise's financial data
It is associated with the industry label；
Sample judges signal acquisition module, judges signal for obtaining sample, the sample judges signal and includes at least one
The judge signal of a financial dimension；
Signal value computing module is judged, the corresponding sample of the enterprise's financial data for calculating each enterprise judges letter
Number value；
Data set division module, for judging signal and the judge signal value to the business finance according to the sample
Data carry out positive negative sample division, obtain positive and negative sample data set；
Model construction module, for being trained predtermined category device to construct finance according to the positive and negative sample data set
Rating Model；
Model memory module is used for the financial Rating Model and the industry label associated storage.
A kind of computer equipment, including memory and processor, the memory are stored with computer program, the processing
The step of device realizes method described above when executing the computer program.
A kind of computer readable storage medium, is stored thereon with computer program, and the computer program is held by processor
The step of method described above is realized when row.
Above-mentioned financial data methods of marking, device, computer equipment and storage medium, financial data of the branch trade to enterprise
Carry out Rating Model building, and the Rating Model constructed trained financial data is carried out by the parameter of financial dimension it is positive and negative
Data set divides, so that the division of positive and negative sample set has controllability, can neatly adjust as needed, effectively prevents drawing
The positive and negative sample data branched away is obviously unbalance, and the accuracy for the model that the positive and negative sample training based on equal number goes out is more preferable,
The financial Rating Model obtained based on training makes the Evaluated effect to financial position of the enterprise more preferable, and assessment result can be than calibrated
Really reflect the true financial situation of enterprise.
Detailed description of the invention
Fig. 1 is the application scenario diagram of financial data methods of marking in one embodiment；
Fig. 2 is that flow diagram involved in financial Rating Model is constructed in one embodiment；
Fig. 3 is that flow diagram involved in financial Rating Model is constructed in another embodiment；
Fig. 4 is the flow diagram of financial data methods of marking in one embodiment；
Fig. 5 is the flow diagram of financial data methods of marking in another embodiment；
Fig. 6 is the structural block diagram of financial data scoring apparatus in one embodiment；
Fig. 7 is the internal structure chart of computer equipment in one embodiment.
Specific embodiment
It is with reference to the accompanying drawings and embodiments, right in order to which the objects, technical solutions and advantages of the application are more clearly understood
The application is further elaborated.It should be appreciated that specific embodiment described herein is only used to explain the application, not
For limiting the application.
Financial data methods of marking provided by the present application, can be applied in application environment as shown in Figure 1.Wherein, eventually
End 102 is communicated with server 104 by network by network.Terminal 102 is asked for sending building model request and scoring
It asks, user selects enterprise to be scored, industry etc. based on terminal.Server 104 is for constructing Rating Model and based on scoring mould
Type carries out business finance status assessment.Wherein, terminal 102 can be, but not limited to be various personal computers, laptop, intelligence
Energy mobile phone, tablet computer and portable wearable device, server 104 can use independent server either multiple servers
The server cluster of composition is realized.
In one embodiment, as shown in Fig. 2, providing a kind of financial Rating Model construction method, comprising the following steps:
Step 202, it receives the disaggregated model that terminal is sent and constructs request, carry industry label in the request.
Step 204: obtaining enterprise's financial data, the industry carried in the corresponding enterprise's mark of enterprise's financial data and request
Label is associated.
It is in advance each enterprise mark addition industry label of server storage.When carrying out disaggregated model building, first
Industry label corresponding to the disaggregated model of specified building.
Specifically, receiving the disaggregated model that terminal is sent constructs request, industry label is carried in the request.Server will be looked into
Look for the financial data of the corresponding all enterprises of industry specified in request.Belong to the finance of all enterprises of financial industry as used
Data construct financial industry business finance Rating Model.
Step 206, it obtains sample and judges signal, sample judges the judge signal that signal includes at least a financial dimension.
The sample of training pattern is the enterprise's financial data obtained in step 202.It is positive and negative for judging that sample judges signal
Sample attribute, in the present embodiment, it has been enterprise or bad enterprise that sample, which judges signal for judging enterprise,.Sample judges signal extremely
It less include the judge signal of a financial dimension.The judge signal of financial dimension can be debt ratio, net profit, stockholder's equity ratio
Rate takes in one or more being in debt in the financial parameters such as ratio, income from sales.
Sample judges the judge signal that signal can also include non-financial dimension, such as can also include the hard finger of business dimension
Whether whether mark, such as enterprise go bankrupt, break a contract.
Step 208, it calculates the corresponding sample of each enterprise's financial data and judges signal value.
According to the enterprise's financial data of all enterprises under setting industry label, sample is calculated according to enterprise's financial data and is judged
The corresponding sample of signal judges signal value.If it is multiple, each enterprise that the sample of financial dimension predetermined, which judges signal,
Financial data corresponds to multiple samples and judges signal value.Or directly acquired from enterprise's financial data sample judge signal it is corresponding
Sample judges signal value.
Judging signal such as first sample is " net profit ", then net profit data are obtained directly from the financial data of enterprise,
The net profit data are that the first sample of enterprise judges signal value.Judging signal such as the second sample is " income debt ratio ", enterprise
When there is no " income debt ratio " data in industry financial data, obtains the sample and judge the corresponding calculating function of signal, according to calculating
Function and enterprise's financial data calculate corresponding " the income debt ratio " numerical value of enterprise, and " income debt ratio " numerical value is the of enterprise
Two samples judge signal value.
Step 210, it judges signal according to sample and judges the enterprise's financial data that signal value will acquire and carry out positive negative sample
It divides, obtains positive and negative sample data set.
If only including financial dimension sample judge signal, according only to calculating judge signal value to each enterprise's wealth
Data of being engaged in carry out positive negative sample division.In one embodiment, judge signal threshold value is preset, by by enterprise's financial data
Corresponding judge signal value is compared with signal threshold value is judged, and determines that enterprise's financial data is positive sample data or negative sample
Data.
If the sample including multiple financial dimensions judges signal, each enterprise's financial data corresponds to multiple judge signals
Value.Signal setting is judged for each sample and judges signal threshold value, and signal value is judged by comparison and judges signal threshold value, according to
Default rule determines that enterprise's financial data is positive sample data or negative sample data.Default rule can be setting ratio
Judge signal value meet judge signal threshold value requirement.Such as include that two samples judge signals, two samples judge signal values with
Corresponding judge signal threshold value comparison is all satisfied positive sample requirement, and corresponding enterprise's financial data is divided into positive sample data.Arbitrarily
One sample judges signal value and is unsatisfactory for positive sample requirement with the comparison of corresponding judge signal threshold value, and corresponding enterprise's financial data is drawn
It is divided into negative sample data.
Step 212, predtermined category device is trained according to positive and negative sample data set to construct financial Rating Model.
Scheduled classifier can be two disaggregated models, be also possible to neural network model (such as convolutional neural networks mould
Type).
In one embodiment, the process of training classifier is that enterprise's financial data is input in classifier, and input is every
When a enterprise's financial data, expected output is inputted, then expected output is " good enterprise if it is positive sample financial data
Industry ", if it is negative sample enterprise's financial data, then expected finance output is " bad enterprise ".It is trained using training algorithm,
Training algorithm will adjust the value of a large amount of variable elements in classifier, this adjustment process makes classifier in given the case where inputting
Lower output gradually approaches anticipated output.When training set is as inputting, the output result and expected result of classifier are close enough
Or deconditioning when being unable to get closer result, classifier training is completed at this time.
Step 214, by the financial Rating Model of building and industry label associated storage.
Signal can be judged for certain some industry setting multiple groups sample.Every group of sample is judged including one or more in signal
The judge signal of a financial dimension.When financial Rating Model constructs, signal is judged according to multiple groups sample and determines the positive negative sample of multiple groups
Data set.Based on the multiple financial Rating Models of multiple groups positive and negative sample data set building, and by the multiple groups finance model of building with it is right
The industry label associated storage answered.
In the present embodiment, distinguishes industry and divide positive and negative sample data set using the judge signal of financial dimension, by fixed
The financial situation of the judge signal entry evaluation enterprise of the financial dimension of the setting industry of justice, using the result of entry evaluation as just
The standard that negative sample divides.Assessment rule (as signal threshold value is judged in adjustment) is adjusted flexibly due to that can pass through, effectively avoid
It is obviously unbalance to mark off the positive and negative sample size come.
The parameter that financial dimension is set in the present embodiment judges signal as model, even if in order to guarantee the accuracy divided
Signal threshold value is judged without judging the adjustment of signal threshold value or only finely tuning, due to the parameter value difference of the financial dimension between enterprise
It is different larger, the positive sample and negative sample quantity marked off can relative equilibrium, the positive and negative sample training based on equal number go out
The accuracy of model is more preferable.
In one embodiment, step 210, signal is judged according to sample and judges the business finance number that signal value will acquire
According to positive negative sample division is carried out, positive and negative sample data set is obtained, comprising: signal value is judged according to sample and is arranged enterprise's mark
Sequence；Enterprise is identified into corresponding financial data according to sequence and carries out positive negative sample division.
Specifically, arranging according to the ascending sequence of signal value or descending sequence is judged enterprise's mark
Sequence.Using accounting be the first setting ratio and the preceding enterprise of sequence identifies corresponding enterprise's financial data as positive sample/negative sample
Notebook data collection.Using accounting be the second setting ratio and the posterior enterprise of sequence identifies corresponding enterprise's financial data as negative sample
Sheet/positive sample data set.First setting ratio and the second setting ratio can be the same or different, the first setting ratio and
The sum of two setting ratios are less than 1.Sort preceding refer to since sequencing queue first place in preceding sequence, sequence refers to after from row
It sorts after starting forward calamity last position of sequence queue.Such as the corporate logo preferably company to sort preceding 30%, rear 10% mark
It is denoted as bad company.
Further, when it includes multiple that sample, which judges signal, the corresponding judge signal of signal is judged according to each sample
Value is ranked up enterprise's mark.Signal, which is judged, for each sample determines positive sample enterprise tag set and negative sample enterprise mark
Label set.The intersection of multiple positive sample enterprises tag set is final positive sample enterprise tag set, multiple negative sample enterprises
The intersection of tag set is final negative sample enterprise tag set.
In the present embodiment, by determining positive negative sample by setting accounting from sequence, it can guarantee positive and negative sample data quilt
Equilibrium is chosen.Removal sequence is used only sample and judges signal in the unconspicuous financial data of positive and negative sample characteristics in middle position
The different apparent financial data of value difference is trained, and the forecast result of model trained can be made more preferable.
In one embodiment, it as shown in figure 3, providing a kind of financial Rating Model construction method, specifically includes as follows
Step:
Step 302, disaggregated model building request is received, carries industry label in request.
Step 304, it searches and the associated enterprise's label of industry label, the corresponding enterprise's financial data of acquisition enterprise's label.
The enterprise's financial data of acquisition can be newest, the financial data of setting time length.Such as obtain enterprise 2018
The financial data of the year first quarter and the second quarter.Specific time span can go to set according to specific demand.
Step 306, each enterprise is identified into corresponding enterprise's financial data according to the time cycle and is divided into multiple groups, every group of wealth
The corresponding time cycle unit of data of being engaged in.
Time cycle can be a season, one month, a week.The specified time cycle is obtained, according to specified
Time cycle is divided into multiple groups to the enterprise's financial data of each enterprise of acquisition.If the specified time cycle is a season,
The enterprise's financial data of acquisition is the financial data in nearest two season, then can be divided into the enterprise's financial data of each enterprise
Two groups, every group of financial data corresponding season.It will may include the wealth in two seasons if the specified time cycle is one month
Business data are divided into 6 groups.
Step 308, it is retrieved as the sector label sample predetermined and judges signal, it is corresponding to calculate every group of financial data
Sample judges signal value, judges signal value according to sample and carries out positive negative sample division to every group of financial data, obtains positive negative sample
Data set.
It calculates the corresponding sample of every group of financial data and judges signal value, using every group of financial data as independent sample number
According to judging signal value according to its corresponding sample and determine sample data set belonging to it.
After determining positive and negative sample data set, negative sample in positive sample data and negative sample data set is counted in positive sample data set
This quantity can merge the sample in a fairly large number of data set if positive and negative sample data differs greatly.Implement at one
In example, two groups of corresponding enterprise mark or multiple groups sample data can be merged.In another embodiment, it can incite somebody to action
Two groups of corresponding enterprise mark and Time Continuous or multiple groups sample data merge.
Such as by two groups of the financial data fractionation group of enterprise A, after sample judges signal differentiation, two groups of wealth of enterprise A segmentation
Business data are divided into positive sample data set.Through quantitative comparison, sample size is more in positive sample data set, then by enterprise A
Corresponding two groups of financial datas are merged into one group of financial data.
Step 310, predtermined category device is trained according to positive and negative sample data set to construct financial Rating Model.
In the present embodiment, the financial data of an enterprise is split into multiple groups financial data, every group of wealth according to the time cycle
Business data are judged to correspond to after signal differentiates through sample generates a positive sample/negative sample.After fractionation, the financial number of a business
According to multiple samples are produced, the sample number of model training is increased.Positive sample is can be further assured that by sample consolidation strategy
With the relative equilibrium of negative sample quantity.
Further, it can be updated based on financial Rating Model of the newest enterprise's financial data to building.Such as prestore
Financial scoring model be to be constructed based on financial data in 2017.Finance can be rebuild according to financial data in 2018
Rating Model, when rebuilding financial Rating Model, can reset sample judge signal so that sample judge signal with most
New financial data is consistent, and is better able to embody the financial situation of newest financial data.Such as the financial number in some industry
A new financial index is increased in, then can judge signal for new financial index as sample.The enterprise rebuild
Financial Rating Model is more bonded the financial situation assessment of current generation, and financial situation prediction effect also can be more preferable.
Based on the finance model constructed in above-mentioned financial Rating Model building embodiment, a kind of business finance shape is provided
State methods of marking, as shown in figure 4, specifically comprising the following steps:
Step 402, scoring request is received, enterprise's mark and financial data to be scored are carried in scoring request.
Step 404, it searches enterprise and identifies associated industry label.
Step 406, the associated multiple financial Rating Models of industry label are obtained.
Multiple finance Rating Models are that the sample based on multiple groups financial dimension judges signal by training history financial data
It obtains, specifically, the sample based on every group of financial dimension, which judges signal, is marking off corresponding one group just for enterprise's financial data
Negative sample data set trains a financial Rating Model based on every group of division result.
Step 408, financial data to be scored is separately input in multiple financial Rating Models, obtains multiple finance and comments
Divide result.
Step 410, multiple financial appraisal results are sent to requesting terminal.
Further, it is also possible to which multiple financial appraisal results are carried out data processing, a final appraisal result is obtained, it is such as logical
It crosses weighting averaging and obtains final financial appraisal result.
In the present embodiment, branch trade scores to the financial data of enterprise, and is scored and believed based on different financial dimensions
Number multiple Rating Models of training fully consider that multiple groups difference financial dimension index carries out the synthesis of enterprise's financial data various dimensions and comments
Point, it can be than more actually reflecting the whole financial situation of the enterprise, the financial position of the enterprise effect that scores is more preferable.
In one embodiment, a kind of financial data methods of marking is provided, is specifically comprised the following steps:
Step 502, scoring request is received, enterprise's mark and financial data to be scored are carried in scoring request.
Step 504, it searches enterprise and identifies associated industry label, obtain the associated financial Rating Model of industry label.
Step 506, it obtains the corresponding sample of financial Rating Model and judges signal.
In constructing financial Rating Model, sample must be pre-defined and judge signal.Therefore, each of building finance scoring mould
Type corresponding one or one group of sample judge signal.
Step 508, financial data to be scored history financial data corresponding with enterprise's label is merged, obtain to
The financial data of qualitative evaluation.
Enterprise's history financial data in the setting time section continuous in time with financial data to be scored is obtained, is obtained
History financial data and financial data to be scored merge into the financial data to qualitative evaluation.
Financial data such as to be scored is the financial data of the third season in 2018 of enterprise A, the history finance number of acquisition
According to the financial data for the 2018 year second quarter continuous in time with the third season in 2018, the finance number of the second quarter in 2018
Data basis according to the financial data merged with third season financial data as qualitative evaluation.
Step 510, it calculates and judges signal value to the corresponding sample of financial data of qualitative evaluation, signal is judged according to sample
It is worth and determines the corresponding financial situation label of enterprise's label.
The step of sample judges signal value phase is calculated in this step and the step 308 for constructing finance model embodiment in Fig. 3
Together.The corresponding financial situation label of enterprise includes " good enterprise ", " bad enterprise " label.Determine the corresponding financial situation of financial data
Label is identical as the step of positive negative sample divides is carried out to financial data.Positive sample is identified as " good enterprise ", and negative sample is identified as
" bad enterprise ".
The corresponding financial position of the enterprise label of financial data after merging is to qualitatively judge the current affiliated finance of the enterprise
State.
Step 512, financial data to be scored and financial situation label are input in financial Rating Model, obtain finance
Appraisal result.
Qualitative financial situation label and financial data to be scored are input in financial Rating Model, commented by finance
Sub-model makes final quantitative assessment result.Financial Rating Model treats scoring by the financial situation label adjustment of input
There is excessively high or too low distortion situation in the scoring of financial data, the scoring effectively prevented.
In the present embodiment, introducing qualitative financial situation, i.e., financial situation, which scores, considers the financial situation of enterprise's history,
The scoring that may make more is bonded the true management position of enterprise, rather than because the exception of individual financial datas, causes wealth
Business condition grading is excessively high or too low.
It should be understood that although each step in the flow chart of Fig. 2-5 is successively shown according to the instruction of arrow,
These steps are not that the inevitable sequence according to arrow instruction successively executes.Unless expressly stating otherwise herein, these steps
Execution there is no stringent sequences to limit, these steps can execute in other order.Moreover, at least one in Fig. 2-5
Part steps may include that perhaps these sub-steps of multiple stages or stage are not necessarily in synchronization to multiple sub-steps
Completion is executed, but can be executed at different times, the execution sequence in these sub-steps or stage is also not necessarily successively
It carries out, but can be at least part of the sub-step or stage of other steps or other steps in turn or alternately
It executes.
In one embodiment, as shown in fig. 6, providing a kind of financial data scoring apparatus, device includes:
Score request receiving module 602, for receiving scoring request, enterprise's mark is carried in scoring request and wait score
Financial data.
Industry label lookup module 604 identifies associated industry label for searching enterprise.
Rating Model obtains module 606, for obtaining the associated financial Rating Model of industry label, wherein finance scoring
Model be by training history financial data obtain, using the financial parameter of setting as judge signal to history financial data into
The positive and negative sample data set of row divides.
Model score module 608 obtains finance and comments for financial data to be scored to be input in financial Rating Model
Divide result.
As a result feedback module 610, for financial appraisal result to be pushed to requesting terminal.
In one embodiment, device further include:
Request module is modeled, for receiving disaggregated model building request, carries industry label in request.
Financial data obtains module, for obtaining enterprise's financial data, the corresponding enterprise's mark of enterprise's financial data and row
Industry label is associated.
Sample judges signal acquisition module, judges signal for obtaining sample, sample judges signal and includes at least a wealth
The judge signal for dimension of being engaged in.
Signal value computing module is judged, the corresponding sample of the enterprise's financial data for calculating each enterprise judges signal
Value.
Data set division module, for signal is judged according to sample and judge the enterprise's financial data that will acquire of signal value into
The positive negative sample of row divides, and obtains positive and negative sample data set.
Model construction module, for being trained to predtermined category device according to positive and negative sample data set to construct finance scoring
Model.
Model memory module is used for financial Rating Model and industry label associated storage.
In one embodiment, it pre-defines multiple groups sample and judges signal, every group of sample, which is judged, includes at least one in signal
The judge signal of a financial dimension；Method further include:
Multi-model constructs module, for judging the multiple financial Rating Models of signal building according to multiple groups sample, and will be multiple
Financial Rating Model and industry label associated storage.
Rating Model obtains module, is also used to obtain the associated multiple financial Rating Models of industry label；
In one embodiment, model score module is also used to for financial data to be scored being input to finance scoring mould
In type, obtaining financial appraisal result includes: that financial data to be scored is separately input in multiple financial Rating Models, is obtained
Multiple finance appraisal results.
Data set division module is also used to judge signal value according to sample being ranked up to enterprise's mark；It will according to sequence
Enterprise identifies corresponding financial data and carries out positive negative sample division.
In one embodiment, device further include:
Data grouping module is more for being divided into the corresponding enterprise's financial data of each enterprise mark according to the time cycle
Group, the corresponding time cycle unit of every group of financial data；
Data set division module is also used to calculate the corresponding sample of every group of financial data and judges signal value, commented according to sample
Sentence signal value and positive negative sample division is carried out to every group of financial data, obtains positive and negative sample data set.
In one embodiment, device further include:
Data combiners block, for closing financial data to be scored history financial data corresponding with enterprise's label
And obtain the financial data to qualitative evaluation；
Merging data judges signal value computing module, for calculating the corresponding sample judge of financial data to qualitative evaluation
Signal value；
Financial situation label determining module determines the corresponding financial situation of enterprise's label for judging signal value according to sample
Label；
Model score module is also used to financial data to be scored and financial situation label being input to financial Rating Model
In, obtain financial appraisal result.
Specific about financial data scoring apparatus limits the limit that may refer to above for financial data methods of marking
Fixed, details are not described herein.Modules in above-mentioned financial data scoring apparatus can fully or partially through software, hardware and its
Combination is to realize.Above-mentioned each module can be embedded in the form of hardware or independently of in the processor in computer equipment, can also be with
It is stored in the memory in computer equipment in a software form, in order to which processor calls the above modules of execution corresponding
Operation.
In one embodiment, a kind of computer equipment is provided, which can be server, internal junction
Composition can be as shown in Figure 7.The computer equipment include by system bus connect processor, memory, network interface and
Database.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment
Include non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and data
Library.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating
The database of machine equipment is for storing financial Rating Model.The network interface of the computer equipment is used to pass through with external terminal
Network connection communication.To realize a kind of financial data methods of marking when the computer program is executed by processor.
It will be understood by those skilled in the art that structure shown in Fig. 7, only part relevant to application scheme is tied
The block diagram of structure does not constitute the restriction for the computer equipment being applied thereon to application scheme, specific computer equipment
It may include perhaps combining certain components or with different component layouts than more or fewer components as shown in the figure.
In one embodiment, a kind of computer equipment, including memory and processor are provided, which is stored with
Computer program, the processor perform the steps of reception scoring request, take in the scoring request when executing computer program
With enterprise's mark and financial data to be scored；It searches and identifies associated industry label with the enterprise；It obtains and the industry
The associated financial Rating Model of label, wherein the finance Rating Model is based on the history for dividing good positive and negative sample data set
Financial data training obtains, and the positive and negative sample data set is to be with the sample judge signal for including at least one financial dimension
According to what is divided；The financial data to be scored is input in the financial Rating Model, finance scoring knot is obtained
Fruit；The financial appraisal result is fed back into requesting terminal.
In one embodiment, it is also performed the steps of when processor executes computer program and receives disaggregated model building
It requests, industry label is carried in the request；Obtain enterprise's financial data, the corresponding enterprise's mark of the enterprise's financial data with
The industry label is associated；It obtains sample and judges signal, the sample judges signal and includes at least commenting for a financial dimension
Sentence signal；The corresponding sample of the enterprise's financial data for calculating each enterprise judges signal value；It is judged and is believed according to the sample
Number and the judges signal value to the positive negative sample division of enterprise's financial data progress, obtain positive and negative sample data set；According to
The positive and negative sample data set is trained predtermined category device to construct financial Rating Model；By the financial Rating Model with
The industry label associated storage.
In one embodiment, the sample according to multiple groups is also performed the steps of when processor executes computer program
The multiple financial Rating Models of signal building are judged, and multiple financial Rating Models are associated with the industry label and are deposited
Storage；The acquisition includes: that acquisition is associated multiple with the industry label with the associated financial Rating Model of the industry label
The finance Rating Model；It is described that the financial data to be scored is input in the financial Rating Model, obtain finance
Appraisal result includes: that the financial data to be scored is separately input in multiple financial Rating Models, is obtained multiple
Financial appraisal result.
In one embodiment, it also performs the steps of when processor executes computer program and is judged according to the sample
Signal value is ranked up enterprise mark；The corresponding financial data of enterprise mark is carried out according to the sequence
Positive negative sample divides.
In one embodiment, also performing the steps of when processor executes computer program will be every according to the time cycle
A enterprise identifies corresponding enterprise's financial data and is divided into multiple groups, the corresponding time cycle unit of every group of financial data；
The corresponding sample of financial data described in calculating every group judges signal value, judges signal value finance described in every group according to the sample
Data carry out positive negative sample division, obtain positive and negative sample data set.
In one embodiment, it also performs the steps of when processor executes computer program by the wealth to be scored
Business data history financial data corresponding with enterprise's label merges, and obtains the financial data to qualitative evaluation；It calculates
The corresponding sample of the financial data to qualitative evaluation judges signal value；Signal value, which is judged, according to the sample determines institute
State the corresponding financial situation label of enterprise's label；The financial data to be scored and the financial situation label are input to institute
It states in financial Rating Model, obtains financial appraisal result.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated
Machine program performs the steps of reception scoring request when being executed by processor, carried in the scoring request enterprise identify and to
The financial data of scoring；It searches and identifies associated industry label with the enterprise；It obtains and the associated finance of industry label
Rating Model, wherein the finance Rating Model is trained based on the history financial data for dividing good positive and negative sample data set
It arrives, the positive and negative sample data set is to be divided with the sample judge signal for including at least one financial dimension for foundation
's；The financial data to be scored is input in the financial Rating Model, financial appraisal result is obtained；By the finance
Appraisal result feeds back to requesting terminal.
In one embodiment, it is also performed the steps of when computer program is executed by processor and receives disaggregated model structure
Request is built, carries industry label in the request；Obtain enterprise's financial data, the corresponding enterprise's mark of the enterprise's financial data
It is associated with the industry label；It obtains sample and judges signal, the sample judges signal and includes at least a financial dimension
Judge signal；The corresponding sample of the enterprise's financial data for calculating each enterprise judges signal value；It is judged according to the sample
Signal and the judge signal value carry out positive negative sample division to the enterprise's financial data, obtain positive and negative sample data set；Root
Predtermined category device is trained according to the positive and negative sample data set to construct financial Rating Model；By the financial Rating Model
With the industry label associated storage.
In one embodiment, the sample according to multiple groups is also performed the steps of when computer program is executed by processor
The multiple financial Rating Models of this judge signal building, and multiple financial Rating Models are associated with the industry label
Storage；The acquisition includes: that acquisition and the industry label are associated more with the associated financial Rating Model of the industry label
A financial Rating Model；It is described that the financial data to be scored is input in the financial Rating Model, obtain wealth
Business appraisal result includes: that the financial data to be scored is separately input in multiple financial Rating Models, is obtained more
A finance appraisal result.
In one embodiment, it also performs the steps of when computer program is executed by processor and is commented according to the sample
Sentence signal value to be ranked up enterprise mark；According to the sequence by the enterprise identify the corresponding financial data into
The positive negative sample of row divides.
In one embodiment, also performing the steps of when computer program is executed by processor will according to the time cycle
Each enterprise identifies corresponding enterprise's financial data and is divided into multiple groups, the corresponding time cycle list of every group of financial data
Member；The corresponding sample of financial data described in calculating every group judges signal value, judges signal value described in every group according to the sample
Financial data carries out positive negative sample division, obtains positive and negative sample data set.
In one embodiment, also performing the steps of when computer program is executed by processor will be described to be scored
Financial data history financial data corresponding with enterprise's label merges, and obtains the financial data to qualitative evaluation；Meter
It calculates the corresponding sample of the financial data to qualitative evaluation and judges signal value；Signal value is judged according to the sample to determine
The corresponding financial situation label of enterprise's label；The financial data to be scored and the financial situation label are input to
In the finance Rating Model, financial appraisal result is obtained.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the computer program can be stored in a non-volatile computer
In read/write memory medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein,
To any reference of memory, storage, database or other media used in each embodiment provided herein,
Including non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
Each technical characteristic of above embodiments can be combined arbitrarily, for simplicity of description, not to above-described embodiment
In each technical characteristic it is all possible combination be all described, as long as however, the combination of these technical characteristics be not present lance
Shield all should be considered as described in this specification.
The several embodiments of the application above described embodiment only expresses, the description thereof is more specific and detailed, but simultaneously
It cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the art
It says, without departing from the concept of this application, various modifications and improvements can be made, these belong to the protection of the application
Range.Therefore, the scope of protection shall be subject to the appended claims for the application patent.
Claims (10)
1. a kind of financial data methods of marking, which comprises
Scoring request is received, carries enterprise's mark and financial data to be scored in the scoring request；
It searches and identifies associated industry label with the enterprise；
It obtains and the associated financial Rating Model of the industry label, wherein the finance Rating Model is to be based on dividing just
What the history financial data training of negative sample data set obtained, the positive and negative sample data set is to include at least one finance dimension
The sample of degree is judged signal foundation and is divided；
The financial data to be scored is input in the financial Rating Model, financial appraisal result is obtained；
The financial appraisal result is fed back into requesting terminal.
2. the method according to claim 1, wherein the method also includes:
Disaggregated model building request is received, carries industry label in the request；
Enterprise's financial data is obtained, the corresponding enterprise's mark of the enterprise's financial data is associated with the industry label；
It obtains sample and judges signal, the sample judges the judge signal that signal includes at least a financial dimension；
The corresponding sample of the enterprise's financial data for calculating each enterprise judges signal value；
Signal and the judge signal value are judged according to the sample, positive negative sample division is carried out to the enterprise's financial data, obtain
To positive and negative sample data set；
Predtermined category device is trained according to the positive and negative sample data set to construct financial Rating Model；
By the financial Rating Model and the industry label associated storage.
3. according to the method described in claim 2, it is characterized in that, sample described in pre-defined multiple groups judges signal；The side
Method further include:
The sample according to multiple groups judges the multiple financial Rating Models of signal building, and will multiple finance Rating Models
With the industry label associated storage；
The acquisition includes: that acquisition is associated multiple with the industry label with the associated financial Rating Model of the industry label
The finance Rating Model；
Described that the financial data to be scored is input in the financial Rating Model, obtaining financial appraisal result includes:
The financial data to be scored is separately input in multiple financial Rating Models, multiple financial appraisal results are obtained.
4. according to the method in claim 2 or 3, which is characterized in that described to judge signal and institute's commentary according to the sample
Sentence signal value and positive negative sample division carried out to the enterprise's financial data, obtains positive and negative sample data set, comprising:
Signal value is judged according to the sample to be ranked up enterprise mark；
The corresponding financial data of enterprise mark is subjected to positive negative sample division according to the sequence.
5. according to the method described in claim 2, it is characterized in that, after the acquisition enterprise's financial data, further includes:
Each enterprise is identified into corresponding enterprise's financial data according to the time cycle and is divided into multiple groups, every group of financial data pair
Answer a time cycle unit；
The corresponding sample of financial data described in calculating every group judges signal value, judges signal value described in every group according to the sample
Financial data carries out positive negative sample division, obtains positive and negative sample data set.
6. method according to claim 1-3, which is characterized in that the method also includes:
The financial data to be scored history financial data corresponding with enterprise's label is merged, is obtained to qualitative
The financial data of assessment；
It calculates the corresponding sample of the financial data to qualitative evaluation and judges signal value；
Signal value, which is judged, according to the sample determines the corresponding financial situation label of enterprise's label；
The financial data to be scored and the financial situation label are input in the financial Rating Model, finance are obtained
Appraisal result.
7. a kind of financial data scoring apparatus, which is characterized in that described device includes:
Score request receiving module, for receiving scoring request, carries enterprise's mark and wealth to be scored in the scoring request
Business data；
Industry label lookup module identifies associated industry label with the enterprise for searching；
Rating Model obtains module, for obtaining and the associated financial Rating Model of the industry label, wherein the finance are commented
Sub-model is obtained based on the history financial data training for dividing good positive and negative sample data set, and the positive and negative sample data set is
It is divided with the sample judge signal foundation for including at least one financial dimension；
Model score module obtains finance for the financial data to be scored to be input in the financial Rating Model
Appraisal result；
As a result feedback module, for the financial appraisal result to be fed back to requesting terminal.
8. device according to claim 7, which is characterized in that described device further include:
Request module is modeled, for receiving disaggregated model building request, carries industry label in the request；
Financial data obtains module, for obtaining enterprise's financial data, the corresponding enterprise's mark of the enterprise's financial data and institute
It is associated to state industry label；
Sample judges signal acquisition module, judges signal for obtaining sample, the sample judges signal and includes at least a wealth
The judge signal for dimension of being engaged in；
Signal value computing module is judged, the corresponding sample of the enterprise's financial data for calculating each enterprise judges signal
Value；
Data set division module, for judging signal and the judge signal value to the enterprise's financial data according to the sample
Positive negative sample division is carried out, positive and negative sample data set is obtained；
Model construction module, for being trained to predtermined category device according to the positive and negative sample data set to construct finance scoring
Model；
Model memory module is used for the financial Rating Model and the industry label associated storage.
9. a kind of computer equipment, including memory and processor, the memory are stored with computer program, feature exists
In the step of processor realizes any one of claims 1 to 6 the method when executing the computer program.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
The step of method described in any one of claims 1 to 6 is realized when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811089672.9A CN109376995A (en) | 2018-09-18 | 2018-09-18 | Financial data methods of marking, device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811089672.9A CN109376995A (en) | 2018-09-18 | 2018-09-18 | Financial data methods of marking, device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109376995A true CN109376995A (en) | 2019-02-22 |
Family
ID=65405237
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811089672.9A Pending CN109376995A (en) | 2018-09-18 | 2018-09-18 | Financial data methods of marking, device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109376995A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110321342A (en) * | 2019-05-27 | 2019-10-11 | 平安科技（深圳）有限公司 | Business valuation studies method, apparatus and storage medium based on intelligent characteristic selection |
| CN110414709A (en) * | 2019-06-18 | 2019-11-05 | 重庆金融资产交易所有限责任公司 | Debt risk intelligent Forecasting, device and computer readable storage medium |
| CN110472660A (en) * | 2019-07-09 | 2019-11-19 | 深圳壹账通智能科技有限公司 | Abnormal deviation data examination method, device, computer equipment and storage medium |
| CN111783829A (en) * | 2020-05-29 | 2020-10-16 | 广发证券股份有限公司 | Financial anomaly detection method and device based on multi-label learning |
| CN111882417A (en) * | 2020-07-30 | 2020-11-03 | 深圳市原点参数信息技术有限公司 | Financial analysis method and device |
| CN112668820A (en) * | 2019-10-16 | 2021-04-16 | 北京国双科技有限公司 | Data processing method, data processing device, storage medium and electronic equipment |
| CN112767967A (en) * | 2020-12-30 | 2021-05-07 | 深延科技(北京)有限公司 | Voice classification method and device and automatic voice classification method |
| CN113269626A (en) * | 2021-06-03 | 2021-08-17 | 北京航空航天大学 | Financial manipulation behavior identification method and device, electronic equipment and medium |
| CN113298630A (en) * | 2020-02-24 | 2021-08-24 | 上海方付通商务服务有限公司 | Method, device, equipment and medium for processing benchmark financial data of small and micro business industry |
| CN114219596A (en) * | 2021-12-14 | 2022-03-22 | 中国平安财产保险股份有限公司 | Data processing method based on decision tree model and related equipment |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101149833A (en) * | 2007-05-15 | 2008-03-26 | 佟辛 | Public information based dynamically interactive type enterprise finance model construction and operation method |
| CN105719073A (en) * | 2016-01-18 | 2016-06-29 | 苏州汇誉通数据科技有限公司 | Enterprise credit evaluation system and method |
-
2018
- 2018-09-18 CN CN201811089672.9A patent/CN109376995A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101149833A (en) * | 2007-05-15 | 2008-03-26 | 佟辛 | Public information based dynamically interactive type enterprise finance model construction and operation method |
| CN105719073A (en) * | 2016-01-18 | 2016-06-29 | 苏州汇誉通数据科技有限公司 | Enterprise credit evaluation system and method |
Non-Patent Citations (2)
| Title |
|---|
| 周梦洁: "基于同行业非平衡数据流的动态企业财务困境预警研究", 中国优秀硕士学位论文全文数据库 经济与管理科学辑, no. 07, 15 July 2018 (2018-07-15), pages 7 - 48 * |
| 孙琳: "基于Logit模型的中小企业信用风险评估", 中国优秀硕士学位论文全文数据库 经济与管理科学辑, no. 09, 15 September 2013 (2013-09-15), pages 31 - 45 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110321342A (en) * | 2019-05-27 | 2019-10-11 | 平安科技（深圳）有限公司 | Business valuation studies method, apparatus and storage medium based on intelligent characteristic selection |
| CN110414709A (en) * | 2019-06-18 | 2019-11-05 | 重庆金融资产交易所有限责任公司 | Debt risk intelligent Forecasting, device and computer readable storage medium |
| CN110472660A (en) * | 2019-07-09 | 2019-11-19 | 深圳壹账通智能科技有限公司 | Abnormal deviation data examination method, device, computer equipment and storage medium |
| WO2021004132A1 (en) * | 2019-07-09 | 2021-01-14 | 深圳壹账通智能科技有限公司 | Abnormal data detection method, apparatus, computer device, and storage medium |
| CN112668820A (en) * | 2019-10-16 | 2021-04-16 | 北京国双科技有限公司 | Data processing method, data processing device, storage medium and electronic equipment |
| CN113298630A (en) * | 2020-02-24 | 2021-08-24 | 上海方付通商务服务有限公司 | Method, device, equipment and medium for processing benchmark financial data of small and micro business industry |
| CN111783829A (en) * | 2020-05-29 | 2020-10-16 | 广发证券股份有限公司 | Financial anomaly detection method and device based on multi-label learning |
| CN111882417A (en) * | 2020-07-30 | 2020-11-03 | 深圳市原点参数信息技术有限公司 | Financial analysis method and device |
| CN112767967A (en) * | 2020-12-30 | 2021-05-07 | 深延科技(北京)有限公司 | Voice classification method and device and automatic voice classification method |
| CN113269626A (en) * | 2021-06-03 | 2021-08-17 | 北京航空航天大学 | Financial manipulation behavior identification method and device, electronic equipment and medium |
| CN114219596A (en) * | 2021-12-14 | 2022-03-22 | 中国平安财产保险股份有限公司 | Data processing method based on decision tree model and related equipment |
| CN114219596B (en) * | 2021-12-14 | 2024-04-12 | 中国平安财产保险股份有限公司 | Data processing method and related equipment based on decision tree model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109376995A (en) | Financial data methods of marking, device, computer equipment and storage medium | |
| US10943186B2 (en) | Machine learning model training method and device, and electronic device | |
| CN107766929B (en) | Model analysis method and device | |
| CN105718490A (en) | Method and device for updating classifying model | |
| CN109816221A (en) | Decision of Project Risk method, apparatus, computer equipment and storage medium | |
| WO2022042123A1 (en) | Image recognition model generation method and apparatus, computer device and storage medium | |
| CN109360105A (en) | Product risks method for early warning, device, computer equipment and storage medium | |
| CN106156809A (en) | For updating the method and device of disaggregated model | |
| CN112329843B (en) | Call data processing method, device, equipment and storage medium based on decision tree | |
| CN106228399A (en) | A kind of stock trader's customer risk preference categories method based on big data | |
| US20220027814A1 (en) | Environmental, social, and governance (esg) performance trends | |
| CN104834479A (en) | Method and system for automatically optimizing configuration of storage system facing cloud platform | |
| CN111931055B (en) | Object recommendation method, object recommendation device and electronic equipment | |
| CN109583682A (en) | Recognition methods, device and the computer equipment of business finance fraud risk | |
| CN108182633A (en) | Loan data processing method, device, computer equipment and storage medium | |
| CN109767071A (en) | User credit ranking method, device, computer equipment and storage medium | |
| CN112785005A (en) | Multi-target task assistant decision-making method and device, computer equipment and medium | |
| US20130046710A1 (en) | Methods and system for financial instrument classification | |
| CN110362798A (en) | Ruling information retrieval analysis method, device, computer equipment and storage medium | |
| CN114881343A (en) | Short-term load prediction method and device of power system based on feature selection | |
| CN109754135A (en) | Behavior of credit data processing method, device, storage medium and computer equipment | |
| CN109559232A (en) | Transaction data processing method, device, computer equipment and storage medium | |
| CN108986786B (en) | Voice interaction equipment rating method, system, computer equipment and storage medium | |
| CN110705889A (en) | Enterprise screening method, device, equipment and storage medium | |
| CN114820618B (en) | Defect detection model training method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |