CN110135944A - Loan product recommended method, device, computer equipment and storage medium - Google Patents
Loan product recommended method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN110135944A CN110135944A CN201910298542.4A CN201910298542A CN110135944A CN 110135944 A CN110135944 A CN 110135944A CN 201910298542 A CN201910298542 A CN 201910298542A CN 110135944 A CN110135944 A CN 110135944A
- Authority
- CN
- China
- Prior art keywords
- loan
- creditor
- loan product
- parameter
- product
- 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 41
- 238000013461 design Methods 0.000 claims abstract description 68
- 238000011156 evaluation Methods 0.000 claims abstract description 36
- 238000012797 qualification Methods 0.000 claims abstract description 36
- 238000012937 correction Methods 0.000 claims description 21
- 239000000284 extract Substances 0.000 claims description 12
- 241000208340 Araliaceae Species 0.000 claims description 10
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims description 10
- 235000003140 Panax quinquefolius Nutrition 0.000 claims description 10
- 235000008434 ginseng Nutrition 0.000 claims description 10
- 238000005259 measurement Methods 0.000 claims description 10
- 238000013528 artificial neural network Methods 0.000 claims description 5
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 239000000047 product Substances 0.000 claims 45
- 239000006227 byproduct Substances 0.000 claims 1
- 238000007405 data analysis Methods 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 7
- 230000008901 benefit Effects 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 238000003062 neural network model Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008520 organization Effects 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
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Development Economics (AREA)
- Theoretical Computer Science (AREA)
- Technology Law (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
This application involves data analysis technique fields, more particularly to a kind of loan product recommended method, device, computer equipment and storage medium, include: the loan qualification evaluation information for obtaining creditor, the target loan product parameter of the creditor is obtained according to the loan qualification evaluation information；The loan product parameter for obtaining loan origination side obtains expected suggested design after being compared the target loan product parameter of the creditor with the loan product parameter of loan origination side, push the expected suggested design to the creditor；Creditor is obtained for the feedback information of the expected suggested design, forms final loan suggested design after optimizing according to the feedback information to the expected suggested design.The application effectively realizes the accurate matching of loan product, creates suitable loan product according to creditor's demand convenient for loan origination root.
Description
Technical field
This application involves data analysis technique fields more particularly to a kind of loan product recommended method, device, computer to set
Standby and storage medium.
Background technique
Loan is bank or other financial institutions by certain interest rate and the conditions such as must give back and carry out lending money-capital
A kind of credit activity form.The loan of broad sense refers to that loan, discount, overdraw etc. go out the general name of loan fund.The side that bank passes through loan
Formula launches away the currency concentrated and money-capital, can satisfy social enlarged reproduction to the needs of replenishment of funds, promotees
Into expanding economy, meanwhile, thus bank can also obtain income from the cost of capital, increase the accumulation of bank itself.
Currently, needing bank or financial institution to judge loan Man's Demands during loan, working as creditor
After the requirement for meeting corresponding loan product, corresponding loan product could be provided to the creditor.And if the creditor with
Loan product matching degree has differences, then cannot offer loans well to creditor.
Therefore, it is badly in need of setting up a perfect matching relationship between the bank and creditor to offer loans, to mention
Rise the purpose and specific aim of loan origination.
Summary of the invention
Based on this, it is necessary to aiming at the problem that lacking the timely information matches between loan origination side and demand for loan side,
A kind of loan product recommended method, device, computer equipment and storage medium are provided.
A kind of loan product recommended method, includes the following steps:
The loan qualification evaluation information for obtaining creditor, obtains the creditor's according to the loan qualification evaluation information
Target loan product parameter；
The loan product parameter for obtaining loan origination side, by the target loan product parameter of the creditor and loan origination
The loan product parameter of side obtains expected suggested design after being compared, push the expected suggested design to the creditor；
Creditor is obtained for the feedback information of the expected suggested design, the expection is pushed away according to the feedback information
The scheme of recommending forms final loan suggested design after optimizing.
In a wherein possible embodiment, the loan qualification evaluation information for obtaining creditor, according to the loan
Money qualification evaluation information obtains the target loan product parameter of the creditor, comprising:
It sends biological characteristic and extracts instruction to terminal where creditor, terminal is according to the life where receiving the creditor
The biological characteristic for the creditor that object feature extraction instruction acquisition is arrived；
Obtain creditor's identity information corresponding with the biological characteristic of the creditor；
Preset credit information registration form is traversed, the loan qualification evaluation information of the identity information of the creditor is obtained；
According to the corresponding relationship of preset loan the qualification evaluation information and loan product kind, the creditor is determined
Target loan product parameter.
In a wherein possible embodiment, the loan product parameter for obtaining loan origination side, by the loan
The target loan product parameter of people obtains expected suggested design, push after being compared with the loan product parameter of loan origination side
The expected suggested design is to the creditor, comprising:
The parameter for obtaining the loan product of each loan origination side, establishes loan origination data group；
The target loan product parameter for obtaining the creditor, by the target loan product parameter of the creditor according to pre-
If weight arranged after form a target loan product sequence；
The target loan product sequence is matched with the data in the loan origination data group, is obtained after matching
Expection loan product push to terminal where the creditor.
It is described to obtain creditor for the feedback letter of the expected suggested design in a wherein possible embodiment
Breath forms final loan suggested design after optimizing according to the feedback information to the expected suggested design, comprising:
Creditor is obtained for extracting the letter of the feature in the feedback information after the feedback information of the expected suggested design
Breath；
The characteristic information and the expected suggested design are entered into ginseng and carry out operation into fuzzy neural network algorithm, is joined out
After obtain recommended models of initially providing a loan；
The parameter of loan product in initial loan recommended models is entered ginseng to be modified into error correction model, after amendment
Obtain the final loan suggested design.
In a wherein possible embodiment, the parameter of the loan product for obtaining each loan origination side,
Establish loan origination data group, comprising:
It sends loan product information collection to instruct to loan origination side's terminal of each loan product to be uploaded, described in acquisition
The feedback information that loan origination side's terminal instructs the loan product information collection；
Extract the loan product parameter of the loan origination side included in the feedback information；
The loan is obtained after the loan product parameter of the loan origination side is arranged according to the time sequencing of upload
Money provides data group.
It is described by the target loan product sequence and the loan origination data in a wherein possible embodiment
Data in group are matched, and the expection loan product obtained after matching pushes to terminal where the creditor, comprising:
By the target loan product sequences segmentation at several data sub-blocks, institute in the target loan product sequence is calculated
The cryptographic Hash for each data sub-block for including；
N number of cryptographic Hash is extracted from calculating in resulting cryptographic Hash, wherein N is the natural number more than or equal to 2, described in generation
N product attribute information of target loan product sequence；
From the loan origination data group, the number of the numerical value of cryptographic Hash and the cryptographic Hash of any data sub-block is extracted
It is worth identical loan product information, pushes loan product information terminal where the creditor.
In a wherein possible embodiment, the parameter by loan product in initial loan recommended models enters ginseng and arrives
It is modified in error correction model, the final loan suggested design is obtained after amendment, comprising:
Theorem being stated by Grange, first step amendment being carried out to the parameter in the initial loan recommended models, amendment is public
Formula are as follows:
ΔYt=lag (Δ Y)-λ (μ t-1), in formula, μ t-1 is non-balancing error item, and λ is short-term correction parameter, Δ YtFor
Error difference, Δ Y are the parameter of measurement for measuring characteristic information and expected suggested design matching degree；
It will carry out assisting whole recurrence by the Grange statement corrected parameter of measurement of theorem, and obtain and assist whole vector；
The whole vector of association is input in the error correction model, obtains and assists whole regression parameter, it is whole according to the association
Regression parameter is finally borrowed after being modified the whole regression parameter of association to the initial loan recommended models as weight
Money suggested design.
A kind of loan product recommendation apparatus, including following module:
Object selection module is set as obtaining the loan qualification evaluation information of creditor, be evaluated according to the loan qualification
Information obtains the target loan product parameter of the creditor；
Expected scheme module is set as obtaining the loan product parameter of loan origination side, the target of the creditor is borrowed
Money product parameters obtain expected suggested design after being compared with the loan product parameter of loan origination side, push the expection and push away
Scheme is recommended to the creditor；
Final scheme module is set as obtaining creditor for the feedback information of the expected suggested design, according to described
Feedback information forms final loan suggested design after optimizing to the expected suggested design.
A kind of computer equipment, including memory and processor are stored with computer-readable instruction in the memory, institute
When stating computer-readable instruction and being executed by the processor, so that the processor executes the step of above-mentioned loan product recommended method
Suddenly.
A kind of storage medium being stored with computer-readable instruction, the computer-readable instruction are handled by one or more
When device executes, so that the step of one or more processors execute above-mentioned loan product recommended method.
Compared with current mechanism, in the application, pass through the loan for providing loan product desired by creditor and financial institution
Money product is accurately matched, to improve the efficiency that financial institution offers loans, while also creditor being enable to obtain in time
Take the loan product of suitable self-condition.By effectively being obtained to creditor's qualification information, thus preferably according to loan
The identity situation of people recommends suitable loan product to it.In addition, by by the target loan product of creditor and loan origination
The loan product of side carries out numerical value conversion, and the maximum loan product of matching degree is precisely obtained after being convenient for parameter relatively.
Detailed description of the invention
By reading the following detailed description of the preferred embodiment, various other advantages and benefits are common for this field
Technical staff will become clear.The drawings are only for the purpose of illustrating a preferred embodiment, and is not considered as to the application
Limitation.
Fig. 1 is a kind of overall flow figure of the loan product recommended method of the application in one embodiment；
Fig. 2 is the Object selection process signal in a kind of loan product recommended method of the application in one embodiment
Figure；
Fig. 3 is that the expected scheme generating process in a kind of loan product recommended method of the application in one embodiment is shown
It is intended to；
Fig. 4 is that the final scheme generating process in a kind of loan product recommended method of the application in one embodiment is shown
It is intended to；
Fig. 5 is a kind of structure chart of the loan product recommendation apparatus of the application 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, and
It is not used in restriction the application.
Those skilled in the art of the present technique are appreciated that unless expressly stated, singular " one " used herein, " one
It is a ", " described " and "the" may also comprise plural form.It is to be further understood that being arranged used in the description of the present application
Diction " comprising " refer to that there are the feature, integer, step, operation, element and/or component, but it is not excluded that in the presence of or addition
Other one or more features, integer, step, operation, element, component and/or their group.
Fig. 1 is a kind of overall flow figure of the loan product recommended method of the application in one embodiment, such as Fig. 1 institute
Show, a kind of loan product recommended method, comprising the following steps:
S1 obtains the loan qualification evaluation information of creditor, obtains the loan according to the loan qualification evaluation information
The target loan product parameter of people；
Specifically, all previous credit information of creditor is stored in credit information database, for example, A creditor is all previous
The amount of money of loan, if the information such as refund on schedule do not provide loan to it if the problem of not refunding repeatedly occurs in A creditor on schedule
Money.The essential informations such as age, the work income of A creditor are extracted if A loan is refunded per capita on time to determine that A creditor can
The loan product parameter of energy demand, these parameters are primarily referred to as the length of maturity, interest and air control model of loan product etc..
S2 obtains the loan product parameter of loan origination side, by the target loan product parameter of the creditor and loan
The loan product parameter of issuer obtains expected suggested design after being compared, push the expected suggested design to the loan
People；
Specifically, being carried out by the loan product parameter of the target loan product parameter of the creditor and loan origination side
When comparing, similarity comparison algorithm, such as cosine-algorithm, the calculating of Euclidean distance algorithm between the two similar can be used
Degree, if similarity is less than preset error threshold, two parameter matchings, general error threshold is 1% or less.By parameter ratio
After relatively, loan is pushed to using the loan product for the loan origination side that parameter coupling number ranking is preceding 3 as expected suggested design
People.
S3 obtains creditor for the feedback information of the expected suggested design, according to the feedback information to described pre-
Phase suggested design forms final loan suggested design after optimizing.
Specifically, creditor for the feedback information of expected loan scheme mainly include the case where 2 kinds it is possible, the first:
Select one or more kinds of loan products as loan product to be claimed from expected suggested design；Second: not from pre-
Any a loan product is selected in phase suggested design.For the first case, if the loan product of creditor's selection is more than one
Kind, then recommend a kind of loan product to be used as final loan suggested design according to optimization algorithm；And for second situation, then it needs
The loan product of reacquisition loan origination side is sent to creditor after re-starting matching again.
The present embodiment carries out accurate by the loan product for providing loan product desired by creditor and financial institution
Match, to improve the efficiency that financial institution offers loans, while creditor also being enable to obtain suitable self-condition in time
Loan product.
Fig. 2 is the Object selection process signal in a kind of loan product recommended method of the application in one embodiment
Figure, as shown, the S1, obtains the loan qualification evaluation information of creditor, obtains according to the loan qualification evaluation information
The target loan product parameter of the creditor, comprising:
S101, send biological characteristic and extract instruction to terminal where creditor, terminal where receiving the creditor according to
The biological characteristic extracts the biological characteristic for the creditor that instruction acquisition is arrived；
Specifically, terminal where the creditor mentions the biological characteristic after receiving biological characteristic and extracting instruction
Instruction fetch carries out Feature Words inquiry, and inquiring the biological characteristic and extracting instruction is that type of biological characteristic extracts, and is
Fingerprint extraction or iris texture extraction etc. are carried out to creditor, then again to the corresponding physical characteristics collecting equipment of starting to creditor
Biological characteristic extract.
S102, creditor's identity information corresponding with the biological characteristic of the creditor is obtained；
Specifically, the characteristic point in the biological characteristic of the creditor and the name of creditor input are obtained, by institute
It states the point of inherent feature corresponding to the name that characteristic point is inputted with the creditor to be compared, determines the loan if consistent
The artificially corresponding creditor of the name sends where re-entering the instruction to the creditor of name eventually if inconsistent
End.
S103, the preset credit information registration form of traversal obtain the loan qualification evaluation of the identity information of the creditor
Information；
Specifically, obtaining creditor's identity information, the credit information is retrieved according to creditor's identity information and is registered
The index list of table extracts all directory entries comprising creditor's identity information from the index list；Wherein,
Each directory entry corresponds to a qualification evaluation index.
In this step, can first it be inquired from the level-one master catalogue of index list during being indexed directory
Some keyword in creditor's identity information, for example be name, then again in the next stage of index list from catalogue
In inquire other keywords, for example be the age 35 years old, and so on obtain the directory entry of creditor's identity information.
S104, according to it is preset it is described loan qualification evaluation information and loan product kind corresponding relationship, determination described in
The target loan product parameter of creditor.
Specifically, loan qualification evaluation information progress binaryzation is obtained into the loan qualification evaluation information of binaryzation,
Table location information is registered in the credit information according to each evaluation information in creditor's qualification evaluation information, establishes two-value
Change loan qualification evaluation information matrix, the element in the binaryzation loan qualification evaluation information matrix is that the loan of binaryzation provides
Matter evaluation information；For example, location information of some evaluation information in the credit information registration form is the 3rd row, the 4th is arranged,
Then its position in binaryzation loan qualification evaluation information matrix is the 3rd row, the 4th column.
The Cultivar parameter for obtaining loan product traverses binaryzation loan qualification evaluation information matrix, obtain it is all with
The consistent element information of Cultivar parameter, using a most classification of consistent element information as the target of the creditor
The parameter of loan product.
The present embodiment, by effectively being obtained to creditor's qualification information, thus preferably according to the identity of creditor
Situation recommends suitable loan product to it.
Fig. 3 is that the expected scheme generating process in a kind of loan product recommended method of the application in one embodiment is shown
It is intended to, as shown, the S2, obtains the loan product parameter of loan origination side, by the target loan product of the creditor
Parameter obtains expected suggested design after being compared with the loan product parameter of loan origination side, push the expected suggested design
To the creditor, comprising:
S201, obtain each loan origination side loan product parameter, establish loan origination data group；
Specifically, obtaining the IP address of terminal where each loan origination side, loan hair is determined according to the IP address
The identity information for the side of putting is marked according to loan product of the identity information to the loan origination side, in order to carry out
Data search.The loan product that each loan origination side provides is gathered into a loan product set, summarizes these loans and produces
The loan origination data group is formed after product set.
S202, the target loan product parameter for obtaining the creditor, by the target loan product parameter of the creditor
A target loan product sequence is formed after being arranged according to preset weight；
Specifically, preset weight is obtained according to historical data, it can be according in all previous loan profile of creditor
Preference obtain, for example some creditor likes that the loan repayment period is long, then by the weight in loan repayment period.
S203, the target loan product sequence is matched with the data in the loan origination data group, is matched
The expection loan product obtained afterwards pushes to terminal where the creditor.
Wherein, the matched mode of parameter of measurement can be used when being matched, the parameter of measurement refers to for measuring
The parameter of data in target loan product sequence and the Data Matching degree in the loan origination data group.
It is trained to obtain revised measurement ginseng specifically, the parameter of measurement is input in BP neural network model
It counts, the formula in training process are as follows:
dtk=(ytk- ct)ct(1-ct)ejk, d in formulatkIndicate the difference of training front and back, ytkIndicate the parameter of input, ct
Indicate the reality output of t-th of neuron of output layer, ejkIt indicates input layer connection member, works as dtkWhen=0, then training terminates, at this time
Obtain the maximum loan product of matching degree.
The present embodiment is turned by the way that the loan product of the target loan product of creditor and loan origination side is carried out numerical value
It changes, the maximum loan product of matching degree is precisely obtained after being convenient for parameter relatively.
Fig. 4 is that the final scheme generating process in a kind of loan product recommended method of the application in one embodiment is shown
It is intended to, as shown, the S3, obtains creditor for the feedback information of the expected suggested design, according to the feedback letter
Breath forms final loan suggested design after optimizing to the expected suggested design, comprising:
S301, creditor is obtained for extracting the spy in the feedback information after the feedback information of the expected suggested design
Reference breath；
Specifically, being sent when sending expected loan product to the creditor, while to terminal where the creditor
Creditor in need is arranged to the tune of loan product satisfaction on the loan feedback form in the loan feedback form of one structuring
Information is looked into, what acquisition creditor filled in from the survey information does not meet the information of its target of providing a loan as characteristic information.
S302, the characteristic information and the expected suggested design are entered to join into fuzzy neural network algorithm and transported
It calculates, obtains recommended models of initially providing a loan after joining out；
Wherein, fuzzy neural network is similar with BP neural network, mainly by input layer, three layers of hidden layer and counts a layer structure
At except that blurring device is equipped in the second layer of three layers of hidden layer, to each node progress semantic ambiguity of this layer
Change reasoning.Therefore, fuzzy neural network model can preferably pure semantic text be analyzed.
S303, it the parameter of loan product in initial loan recommended models is entered into ginseng is modified into error correction model,
The final loan suggested design is obtained after amendment.
Specifically, can be repaired using first-order error when the error correction model used in this step carries out error correction
Just, it can also be corrected using second order error.
The present embodiment is modified expected suggested design by neural network model and error correction model, thus
To the best loan product of suitable creditor.
In one embodiment, the S201, obtain each loan origination side loan product parameter, establish and borrow
Money provides data group, comprising:
It sends loan product information collection to instruct to loan origination side's terminal of each loan product to be uploaded, described in acquisition
The feedback information that loan origination side's terminal instructs the loan product information collection；
Specifically, when sending loan product information collection instruction, according to the organization names of different loan origination mechanisms
It is sent, for example A bank receives loan product acquisition instructions prior to Z bank.
Extract the loan product parameter of the loan origination side included in the feedback information；
Wherein, it when carrying out feedback information extraction, if can also include in feedback information includes key, needs to feedback
Information be decrypted after obtain feedback information content.The key can be the existing institute such as Hash key, symmetric key
There is key form.If decryption is unsuccessful, loan product parameter is not obtained from the loan product issuer.
The loan is obtained after the loan product parameter of the loan origination side is arranged according to the time sequencing of upload
Money provides data group.
The present embodiment effectively analyzes feedback information, to get creditor for the evaluation feelings of loan product
Condition.
In one embodiment, the S203, will be in the target loan product sequence and the loan origination data group
Data matched, the expection loan product obtained after matching pushes to terminal where the creditor, comprising:
By the target loan product sequences segmentation at several data sub-blocks, institute in the target loan product sequence is calculated
The cryptographic Hash for each data sub-block for including；
Specifically, when loan product sequence is split can using the total amount of data of sequence as parameter enter ginseng to
It is calculated in machine function, the length of segmentation sub-block is obtained according to calculated result.
N number of cryptographic Hash is extracted from calculating in resulting cryptographic Hash, wherein N is the natural number more than or equal to 2, described in generation
N product attribute information of target loan product sequence；
Specifically, each cryptographic Hash corresponds to a product attribute value, identical cryptographic Hash corresponds to identical product
Attribute value, and each product attribute value corresponds to a product attribute information in product attribute list.It therefore, can basis
The product attribute table stored in cryptographic Hash ergodic data library, extracts the attribute information of loan product from product attribute table.
From the loan origination data group, the number of the numerical value of cryptographic Hash and the cryptographic Hash of any data sub-block is extracted
It is worth identical loan product information, pushes loan product information terminal where the creditor.
The present embodiment obtains the preferred plan of loan product push by cryptographic Hash, to make the recommendation side of loan product
Formula weight, to keep loan product matching more accurate.
In one embodiment, the S303, the parameter of loan product in initial loan recommended models entered into ginseng arrive error
It is modified in correction model, the final loan suggested design is obtained after amendment, comprising:
Theorem being stated by Grange, first step amendment being carried out to the parameter in the initial loan recommended models, amendment is public
Formula are as follows:
ΔYt=lag (Δ Y)-λ (μ t-1), in formula, μ t-1 is non-balancing error item, and λ is short-term correction parameter, Δ YtFor
Error difference, Δ Y are the parameter of measurement for measuring characteristic information and expected suggested design matching degree；
It will carry out assisting whole recurrence by the Grange statement corrected parameter of measurement of theorem, and obtain and assist whole vector；
The whole vector of association is input in the error correction model, obtains and assists whole regression parameter, it is whole according to the association
Regression parameter is finally borrowed after being modified the whole regression parameter of association to the initial loan recommended models as weight
Money suggested design.
In the present embodiment, there are many clear advantages for error correction model: a) use of first-order difference item eliminates variable
Trend factor that may be present, so as to avoid False value problem；B) use of first-order difference item is also eliminated model and may be deposited
Problems of Multiple Synteny；C) introducing of error correction item ensure that the information of variable level value is not ignored；D) due to
The stationarity of error correction item itself estimates the homing method of model classics, especially poor in model
Subitem can be used common t inspection and be examined with F to be chosen.Therefore, whether an important problem is exactly: between variable
Relationship can be stated by error correction model, on this question, Engle and propose within Granger 1987 it is famous
Grange states theorem.
The present embodiment is reduced by carrying out error correction because calculating loan product matching error caused by error.
In one embodiment it is proposed that a kind of loan product recommendation apparatus, as shown in figure 5, including following module:
Object selection module 51 is set as obtaining the loan qualification evaluation information of creditor, be commented according to the loan qualification
Valence information obtains the target loan product parameter of the creditor；
Expected scheme module 52 is set as obtaining the loan product parameter of loan origination side, by the target of the creditor
Loan product parameter obtains expected suggested design after being compared with the loan product parameter of loan origination side, push the expection
Suggested design is to the creditor；
Final scheme module 53 is set as obtaining creditor for the feedback information of the expected suggested design, according to institute
It states and forms final loan suggested design after feedback information optimizes the expected suggested design.
In one embodiment it is proposed that a kind of computer equipment, the computer equipment includes memory and processor,
Computer-readable instruction is stored in memory, when computer-readable instruction is executed by processor, so that processor execution is above-mentioned
The step of loan product recommended method in each embodiment.
In one embodiment it is proposed that a kind of storage medium for being stored with computer-readable instruction, this is computer-readable
When instruction is executed by one or more processors, so that one or more processors execute the loan in the various embodiments described above
The step of Products Show method.Wherein, the storage medium can be non-volatile memory medium.
Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of above-described embodiment is can
It is completed with instructing relevant hardware by program, which can be stored in a computer readable storage medium, storage
Medium may include: read-only memory (ROM, Read Only Memory), random access memory (RAM, Random
Access Memory), disk or CD etc..
Each technical characteristic of embodiment described above can be combined arbitrarily, for simplicity of description, not to above-mentioned reality
It applies all possible combination of the technical characteristic in example to be all described, as long as however, lance is not present in the combination of these technical characteristics
Shield all should be considered as described in this specification.
The some exemplary embodiments of the application above described embodiment only expresses, wherein describe it is more specific and detailed,
But it cannot be understood as the limitations to the application the scope of the patents.It should be pointed out that for the ordinary skill of this field
For personnel, without departing from the concept of this application, various modifications and improvements can be made, these belong to the application
Protection scope.Therefore, the scope of protection shall be subject to the appended claims for the application patent.
Claims (10)
1. a kind of loan product recommended method characterized by comprising
The loan qualification evaluation information for obtaining creditor, obtains the target of the creditor according to the loan qualification evaluation information
Loan product parameter；
The loan product parameter for obtaining loan origination side, by the target loan product parameter of the creditor and loan origination side
Loan product parameter obtains expected suggested design after being compared, push the expected suggested design to the creditor；
Creditor is obtained for the feedback information of the expected suggested design, according to the feedback information to the expected recommendation side
Case forms final loan suggested design after optimizing.
2. loan product recommended method according to claim 1, which is characterized in that the loan qualification for obtaining creditor
Evaluation information obtains the target loan product parameter of the creditor according to the loan qualification evaluation information, comprising:
It sends biological characteristic and extracts instruction to terminal where creditor, terminal where receiving the creditor is special according to the biology
Sign extracts the biological characteristic for the creditor that instruction acquisition is arrived；
Obtain creditor's identity information corresponding with the biological characteristic of the creditor；
Preset credit information registration form is traversed, the loan qualification evaluation information of the identity information of the creditor is obtained；
According to the corresponding relationship of preset loan the qualification evaluation information and loan product kind, the mesh of the creditor is determined
Mark loan product parameter.
3. loan product recommended method according to claim 1, which is characterized in that the loan for obtaining loan origination side
Product parameters, after the target loan product parameter of the creditor is compared with the loan product parameter of loan origination side
To expected suggested design, the expected suggested design is pushed to the creditor, comprising:
The parameter for obtaining the loan product of each loan origination side, establishes loan origination data group；
The target loan product parameter for obtaining the creditor, by the target loan product parameter of the creditor according to preset
Weight forms a target loan product sequence after being arranged；
The target loan product sequence is matched with the data in the loan origination data group, what is obtained after matching is pre-
Phase loan product pushes to terminal where the creditor.
4. loan product recommended method according to claim 1, which is characterized in that the acquisition creditor is for described pre-
The feedback information of phase suggested design forms final loan after optimizing according to the feedback information to the expected suggested design
Suggested design, comprising:
Creditor is obtained for extracting the characteristic information in the feedback information after the feedback information of the expected suggested design；
The characteristic information and the expected suggested design are entered into ginseng and carry out operation into fuzzy neural network algorithm, after joining out
To initial loan recommended models；
The parameter of loan product in initial loan recommended models is entered ginseng to be modified into error correction model, is obtained after amendment
The final loan suggested design.
5. loan product recommended method according to claim 3, which is characterized in that described to obtain each loan origination
The parameter of the loan product of side, establishes loan origination data group, comprising:
It sends loan product information collection to instruct to loan origination side's terminal of each loan product to be uploaded, obtains the loan
The feedback information that issuer terminal instructs the loan product information collection；
Extract the loan product parameter of the loan origination side included in the feedback information；
The loan hair is obtained after the loan product parameter of the loan origination side is arranged according to the time sequencing of upload
Put data group.
6. loan product recommended method according to claim 3, which is characterized in that described by the target loan product sequence
Column are matched with the data in the loan origination data group, and the expection loan product obtained after matching pushes to the loan
Terminal where people, comprising:
By the target loan product sequences segmentation at several data sub-blocks, calculate included in the target loan product sequence
Each data sub-block cryptographic Hash；
N number of cryptographic Hash is extracted from calculating in resulting cryptographic Hash, wherein N is the natural number more than or equal to 2, generates the target
N product attribute information of loan product sequence；
From the loan origination data group, the numerical value and the numerical value phase of the cryptographic Hash of any data sub-block of cryptographic Hash are extracted
Same loan product information pushes loan product information terminal where the creditor.
7. loan product recommended method according to claim 4, which is characterized in that in the recommended models that will initially provide a loan
The parameter of loan product enters ginseng and is modified into error correction model, and the final loan suggested design is obtained after amendment, wraps
It includes:
Theorem is stated by Grange, and first step amendment, correction formula are carried out to the parameter in the initial loan recommended models
Are as follows:
ΔYt=lag (Δ Y)-λ (μ t-1), in formula, μ t-1 is non-balancing error item, and λ is short-term correction parameter, Δ YtFor error
Difference, Δ Y are the parameter of measurement for measuring characteristic information and expected suggested design matching degree；
It will carry out assisting whole recurrence by the Grange statement corrected parameter of measurement of theorem, and obtain and assist whole vector；
The whole vector of association is input in the error correction model, obtains and assists whole regression parameter, according to the whole recurrence of association
Parameter, obtaining finally providing a loan after the whole regression parameter of association is modified the initial loan recommended models as weight pushes away
Recommend scheme.
8. a kind of loan product recommendation apparatus characterized by comprising
Object selection module is set as obtaining the loan qualification evaluation information of creditor, according to the loan qualification evaluation information
Obtain the target loan product parameter of the creditor；
Expected scheme module is set as obtaining the loan product parameter of loan origination side, the target of the creditor is provided a loan and is produced
Product parameter obtains expected suggested design after being compared with the loan product parameter of loan origination side, push the expected recommendation side
Case is to the creditor；
Final scheme module is set as obtaining creditor for the feedback information of the expected suggested design, according to the feedback
Information forms final loan suggested design after optimizing to the expected suggested design.
9. a kind of computer equipment, which is characterized in that including memory and processor, being stored with computer in the memory can
Reading instruction, when the computer-readable instruction is executed by the processor, so that the processor executes such as claim 1 to 7
Any one of loan product recommended method described in claim the step of.
10. a kind of storage medium, which is characterized in that the storage medium is stored with computer-readable instruction, the storage medium
It can be read and write with device processed, when the computer-readable instruction is executed by one or more processors, so that at one or more
Device is managed to execute as described in any one of claims 1 to 7 claim the step of loan product recommended method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910298542.4A CN110135944A (en) | 2019-04-15 | 2019-04-15 | Loan product recommended method, device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910298542.4A CN110135944A (en) | 2019-04-15 | 2019-04-15 | Loan product recommended method, device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110135944A true CN110135944A (en) | 2019-08-16 |
Family
ID=67569877
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910298542.4A Pending CN110135944A (en) | 2019-04-15 | 2019-04-15 | Loan product recommended method, device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110135944A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210097606A1 (en) * | 2019-09-30 | 2021-04-01 | Volvo Car Corporation | Online vehicle subscription service including an automated credit check function |
| CN116630017A (en) * | 2023-05-06 | 2023-08-22 | 广州市良策网络技术有限公司 | Loan product automatic matching method and system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107729443A (en) * | 2017-09-29 | 2018-02-23 | 平安科技（深圳）有限公司 | Loan product promotion method, device and computer-readable recording medium |
| CN108876601A (en) * | 2018-09-18 | 2018-11-23 | 中国银行股份有限公司 | A kind of loan requests processing method, device, electronic equipment and storage medium |
| CN109389491A (en) * | 2018-09-27 | 2019-02-26 | 深圳壹账通智能科技有限公司 | Loan product screening technique, device, equipment and storage medium based on big data |
| CN109389490A (en) * | 2018-09-26 | 2019-02-26 | 深圳壹账通智能科技有限公司 | Loan product matching process, device, computer equipment and storage medium |
| CN109409780A (en) * | 2018-11-21 | 2019-03-01 | 平安科技（深圳）有限公司 | Changing process method, device, computer equipment and storage medium |
| CN109410032A (en) * | 2018-09-26 | 2019-03-01 | 深圳壹账通智能科技有限公司 | A kind of information processing method, server and computer storage medium |
-
2019
- 2019-04-15 CN CN201910298542.4A patent/CN110135944A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107729443A (en) * | 2017-09-29 | 2018-02-23 | 平安科技（深圳）有限公司 | Loan product promotion method, device and computer-readable recording medium |
| CN108876601A (en) * | 2018-09-18 | 2018-11-23 | 中国银行股份有限公司 | A kind of loan requests processing method, device, electronic equipment and storage medium |
| CN109389490A (en) * | 2018-09-26 | 2019-02-26 | 深圳壹账通智能科技有限公司 | Loan product matching process, device, computer equipment and storage medium |
| CN109410032A (en) * | 2018-09-26 | 2019-03-01 | 深圳壹账通智能科技有限公司 | A kind of information processing method, server and computer storage medium |
| CN109389491A (en) * | 2018-09-27 | 2019-02-26 | 深圳壹账通智能科技有限公司 | Loan product screening technique, device, equipment and storage medium based on big data |
| CN109409780A (en) * | 2018-11-21 | 2019-03-01 | 平安科技（深圳）有限公司 | Changing process method, device, computer equipment and storage medium |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210097606A1 (en) * | 2019-09-30 | 2021-04-01 | Volvo Car Corporation | Online vehicle subscription service including an automated credit check function |
| CN116630017A (en) * | 2023-05-06 | 2023-08-22 | 广州市良策网络技术有限公司 | Loan product automatic matching method and system |
| CN116630017B (en) * | 2023-05-06 | 2023-11-21 | 广州市良策网络技术有限公司 | Loan product automatic matching method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108475393A (en) | The system and method that decision tree is predicted are promoted by composite character and gradient | |
| CN108776922A (en) | Finance product based on big data recommends method and device | |
| Roßbach | Neural networks vs. random forests–does it always have to be deep learning | |
| CN111080338B (en) | User data processing method and device, electronic equipment and storage medium | |
| CN109063931A (en) | A kind of model method for predicting freight logistics driver Default Probability | |
| Ferdousy et al. | Combination of naive bayes classifier and K-Nearest Neighbor (cNK) in the classification based predictive models | |
| US8984022B1 (en) | Automating growth and evaluation of segmentation trees | |
| Hanslo et al. | Machine learning models to predict agile methodology adoption | |
| CN106780018A (en) | A kind of stock certificate data mining model reappeared based on history and system | |
| CN110135944A (en) | Loan product recommended method, device, computer equipment and storage medium | |
| Saheed et al. | Effectiveness of deep learning long short-term memory network for stock price prediction on graphics processing unit | |
| Sharma et al. | Stock market prediction using historical stock prices and dependence on other companies in automotive sector | |
| Prasanth et al. | Intelligent loan eligibility and approval system based on random forest algorithm using machine learning | |
| Aliaj et al. | Nowcasting inflation with Lasso‐regularized vector autoregressions and mixed frequency data | |
| CN111340102B (en) | Method and apparatus for evaluating model interpretation tools | |
| CN116167371A (en) | Product recommendation method and device, processor and electronic equipment | |
| CN109711848A (en) | A kind of matching system and its construction method, matching process of financial transaction | |
| Wu et al. | Customer churn prediction for commercial banks using customer-value-weighted machine learning models | |
| Sriman et al. | Price prediction of bitcoin and ethereum-A machine learning approach | |
| Yeh et al. | Multitask Learning for Time Series Data with 2D Convolution | |
| Wang | A Study on Early Warning of Financial Indicators of Listed Companies Based on Random Forest | |
| Gupta et al. | Classification-Based Credit Risk Analysis: The Case of Lending Club | |
| Oberoi et al. | Bankruptcy Prediction of Indian Banks Using Advanced Analytics | |
| Lee et al. | Evaluating missing data handling methods for developing building energy benchmarking models | |
| KR101725015B1 (en) | Appartus for home sales index prediction using artificial neural network and method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190816 |
|
| WD01 | Invention patent application deemed withdrawn after publication |