CN108182633A - Loan data processing method, device, computer equipment and storage medium - Google Patents
Loan data processing method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN108182633A CN108182633A CN201810091577.6A CN201810091577A CN108182633A CN 108182633 A CN108182633 A CN 108182633A CN 201810091577 A CN201810091577 A CN 201810091577A CN 108182633 A CN108182633 A CN 108182633A
- Authority
- CN
- China
- Prior art keywords
- loan
- client
- continuous
- borrowing
- prediction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
This application involves a kind of loan data processing method, device, computer equipment and storage medium, methods to include：It obtains each first personally identifiable information for borrowing client and head in currently first loan client set and borrows application information, borrow application information according to personally identifiable information and head obtains the corresponding first continuous loan prediction probability of each first loan client using preset first prediction model；Obtain it is current first borrow in client set each first monthly mortgage of the client in preset time period of borrowing than data, obtained than data each first borrowing the corresponding second continuous loan prediction probability of client using preset second prediction model according to monthly mortgage；Each first corresponding synthesis of client of borrowing is obtained using preset third prediction model according to the first continuous loan prediction probability and the second continuous loan prediction probability and continues loan prediction probability；Continue loan prediction probability according to the corresponding synthesis of each head loans client each first loan client is ranked up to obtain ranking results, according to ranking results push loan product information.The present invention can save computer resource.
Description
Technical field
This application involves technical field of data processing, more particularly to a kind of loan data processing method, device, computer
Equipment and storage medium.
Background technology
With the rapid development of Internet technology, such as finance product, loan product of the financial product based on internet is more next
It is more.Wherein, loan product mainly includes small amount debt-credit and wholesale debt-credit, and small amount borrows or lends money the threshold due to its application than relatively low,
Favored by more and more people, therefore, for borrow or lend money the internet financing corporation for main business, how to perform small amount loan
The marketing of money is into the task of top priority.
In traditional technology, internet financing corporation carries out mainly randomly choosing visitor by background server during petty load marketing
Family carries out the push of loan product, and to cause success rate height of marketing, background server is generally required to carry out the push of magnanimity, be caused
The waste of computer resource.
Invention content
Based on this, it is necessary to which for above-mentioned technical problem, providing a kind of can save at the loan data of computer resource
Manage method, apparatus, computer equipment and storage medium.
A kind of loan data processing method, the method includes：
It obtains current head and borrows each first personally identifiable information for borrowing client and first loan application information in client set, according to institute
It states personally identifiable information and first application information of borrowing obtains each first loan client corresponding first using preset first prediction model
It is continuous to borrow prediction probability；
It obtains each head in current first loan client set and borrows monthly mortgage of the client in preset time period than data, according to described
Monthly mortgage obtains each first borrowing that client is corresponding by second continuous to borrow prediction probability using preset second prediction model than data；
According to the described first continuous prediction probability and second continuous prediction probability of borrowing borrowed using preset third prediction model
It obtains each first corresponding synthesis of client of borrowing and continues loan prediction probability；
According to it is each it is first borrow the corresponding synthesis of client continue borrow prediction probability to it is each it is first borrow client and be ranked up sorted
As a result, push loan product information according to the ranking results.
The generation step of first prediction model includes in one of the embodiments,：
It is general to obtain the personally identifiable information of each loan customer in history loan customer set, first loan application information and continuous loan
Rate；
First training sample is obtained according to the personally identifiable information, first loan application information and continuous probability of borrowing；
Model training is carried out according to first training sample and obtains the first prediction model.
The generation step of second prediction model includes in one of the embodiments,：
It obtains monthly mortgage of each loan customer in preset time period in history loan customer set and compares data；
Second training sample is obtained than data and continuous probability of borrowing according to the monthly mortgage；
Model training is carried out according to second training sample and obtains the second prediction model.
The generation step of the third prediction model includes in one of the embodiments,：
It is adopted according to the personally identifiable information of each loan customer in the history loan customer set and first application information of borrowing
The corresponding first continuous loan prediction probability of each loan customer is obtained with preset first prediction model；
It is used according to monthly mortgage of each loan customer in preset time period in the history loan customer set than data
Preset second prediction model obtains the corresponding second continuous loan prediction probability of each loan customer；
It is predicted according to the corresponding first continuous loan prediction probability of loan customer each in history loan customer set, the second continuous borrow
Probability and continuous probability of borrowing obtain third training sample, and carrying out model training according to the third training sample obtains third prediction mould
Type.
It is described in one of the embodiments, that loan prediction probability is continued to each head according to each first corresponding synthesis of client of borrowing
The step of client is ranked up to obtain ranking results, loan product information is pushed according to the ranking results is borrowed to include：
Continuous loan client set is obtained according to the history loan customer set；
Head is chosen in current first loan client set borrow client as current goal client according to the ranking results；
Each continuous loan client mappings in current goal client and continuous loan client set are borrowed into information and identity information to head
In the multi-C vector space of coordinate, to respectively obtain the vectorial coordinate of current goal client and the corresponding vector of each continuous loan client
Coordinate；
Calculate current goal client and the continuous vector distance borrowed in client set between each continuous loan client；
Corresponding continuous loan client is joined as the references object of current goal client when vector distance is less than preset value
Examine object set；
Loan product information is pushed according to the references object set.
Described the step of pushing loan product information according to the references object set, wraps in one of the embodiments,
It includes：
The continuous loan product of each references object in references object set is inquired, obtains continuous loan product set；
It obtains and continuous borrows the corresponding references object quantity of each continuous loan product in product set；
The corresponding target push product of current goal client is obtained according to each continuous corresponding references object quantity of product of borrowing.
The method further includes in one of the embodiments,：
Target push product is pushed into the corresponding terminal of current goal client.
A kind of loan data processing unit, which is characterized in that described device includes：
First continuous loan prediction probability acquisition module, for obtaining the current first individual for borrowing each first loan client in client set
Identity information and first loan application information, according to the personally identifiable information and first application information of borrowing using the preset first prediction mould
Type obtains the corresponding first continuous loan prediction probability of each first loan client；
Second continuous loan prediction probability acquisition module is being preset for obtaining each first loan client in currently first loan client set
Monthly mortgage in period obtains each first loan client than data according to the monthly mortgage than data using preset second prediction model
Corresponding second continuous loan prediction probability；
It is comprehensive to continue loan prediction probability module, it is general for being predicted according to the described first continuous loan prediction probability and the second continuous loan
Rate obtains each first corresponding synthesis of client of borrowing using preset third prediction model and continues loan prediction probability；
Sorting module continues loan prediction probability to each first loan client progress for borrowing the corresponding synthesis of client according to each head
Sequence obtains ranking results, and loan product information is pushed according to the ranking results.
A kind of computer equipment can be run on a memory and on a processor including memory, processor and storage
Computer program, the processor realize following steps when performing the computer program：
It obtains current head and borrows each first personally identifiable information for borrowing client and first loan application information in client set, according to institute
It states personally identifiable information and first application information of borrowing obtains each first loan client corresponding first using preset first prediction model
It is continuous to borrow prediction probability；
It obtains each head in current first loan client set and borrows monthly mortgage of the client in preset time period than data, according to described
Monthly mortgage obtains each first borrowing that client is corresponding by second continuous to borrow prediction probability using preset second prediction model than data；
According to the described first continuous prediction probability and second continuous prediction probability of borrowing borrowed using preset third prediction model
It obtains each first corresponding synthesis of client of borrowing and continues loan prediction probability；
According to it is each it is first borrow the corresponding synthesis of client continue borrow prediction probability to it is each it is first borrow client and be ranked up sorted
As a result, push loan product information according to the ranking results.
A kind of computer readable storage medium, is stored thereon with computer program, and the computer program is held by processor
Following steps are realized during row：
It obtains current head and borrows each first personally identifiable information for borrowing client and first loan application information in client set, according to institute
It states personally identifiable information and first application information of borrowing obtains each first loan client corresponding first using preset first prediction model
It is continuous to borrow prediction probability；
It obtains each head in current first loan client set and borrows monthly mortgage of the client in preset time period than data, according to described
Monthly mortgage obtains each first borrowing that client is corresponding by second continuous to borrow prediction probability using preset second prediction model than data；
According to the described first continuous prediction probability and second continuous prediction probability of borrowing borrowed using preset third prediction model
It obtains each first corresponding synthesis of client of borrowing and continues loan prediction probability；
According to it is each it is first borrow the corresponding synthesis of client continue borrow prediction probability to it is each it is first borrow client and be ranked up sorted
As a result, push loan product information according to the ranking results.
Above-mentioned loan data processing method, device, computer equipment and storage medium are getting current first loan client collection
When each first personally identifiable information for borrowing client and head borrow application information in conjunction, can letter of application be borrowed according to personally identifiable information and head
It ceases and the corresponding first continuous loan prediction probability of each first loan client is obtained using preset first prediction model, borrowed obtaining current head
When each first monthly mortgage for borrowing client in preset time period is than data in client set, can according to monthly mortgage than data using preset the
Two prediction models obtain it is each it is first borrow client it is corresponding second it is continuous borrow prediction probability, then continuous borrow prediction probability and the according to first
Two continuous loan prediction probabilities obtain each first corresponding synthesis of client of borrowing using preset third prediction model and continue loan prediction probability, most
Continue loan prediction probability according to the corresponding synthesis of each head loans client afterwards each first loan client is ranked up to obtain ranking results, root
Loan product information is pushed according to ranking results, due to being predicted twice from different dimensions, and predicted twice according to this
As a result integrated forecasting probability is obtained, which can react the possibility that client continues loan, and according to the size of the possibility
The push for having target is carried out, a large amount of computer resource can be saved.
Description of the drawings
Fig. 1 is the application scenario diagram of loan data processing method in one embodiment；
Fig. 2 is the flow diagram of loan data processing method in one embodiment；
Fig. 3 is the step flow chart that the first prediction model generates in step S202 in one embodiment；
Fig. 4 is the step flow chart that the second prediction model generates in step S206 in one embodiment；
Fig. 5 is the flow diagram of loan data processing method in another embodiment；
Fig. 6 is the structure diagram of loan data processing unit in one embodiment；
Fig. 7 is the structure diagram of loan data processing unit in another embodiment；
Fig. 8 is the internal structure chart of one embodiment Computer equipment.
Specific embodiment
It is with reference to the accompanying drawings and embodiments, right in order to which the object, technical solution and advantage for making 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.
The loan data processing method that the application provides, can be applied in application environment as shown in Figure 1.In the application
In environment, including at least server 110, marketing personnel's terminal 120, loan customer terminal 130, server 110 is to loan customer
Continuous loan probability predicted, according to prediction result push loan product sent out to loan customer terminal 130, while by prediction result
It send to marketing personnel's terminal 120.Wherein, marketing personnel's terminal 120 is communicated by network with server 110, loan customer
Terminal 130 is communicated by network with server 110.Wherein, marketing personnel's terminal 120, loan customer terminal 130 can with but
It is not limited to various personal computers, laptop, smart mobile phone, tablet computer and portable wearable device, server
110 can be realized with the server cluster that the either multiple servers of independent server form.
In one embodiment, as shown in Fig. 2, a kind of loan data processing method is provided, in this way applied to Fig. 1
In server for illustrate, include the following steps：
Step S202 obtains current head and borrows each first personally identifiable information for borrowing client and first loan letter of application in client set
Breath according to personally identifiable information and first borrow application information using preset first prediction model to obtain each first loan client corresponding
First continuous loan prediction probability.
Specifically, current first client set of borrowing refers to that current time is interior in loan platform toward one preset time is pushed forward
By the first set borrowed client and formed of behavior of lending, if current time is January 5, then current first client set of borrowing can be
There is the first of behavior of lending to borrow client in loan platform in December 25 to the 10 day time on January 5.Wherein, it is first to borrow what client referred to
It is the client only once to be provided a loan excessively in loan platform, personally identifiable information includes age, gender, occupation, monthly income etc., first
Loan application information refers to the application information when platform of providing a loan provide a loan for the first time, including the amount of the loan, the length of maturity.
In one embodiment, server can periodically collect loan data, and current first loan client is obtained according to loan data
Set, and obtain it is current first borrow in client set each first personally identifiable information for borrowing client and it is first borrow application information, such as every
It collects a loan data within 10 days, obtains current first loan client set；In another embodiment, server is receiving marketing
After the request of staff terminal, current slot loan data is collected, such as after the request for receiving marketing personnel's terminal, collects and works as
The loan data of 10 days of preceding time forward obtains current first loan client set, and obtain current first according to current loan data
Borrow each first personally identifiable information for borrowing client and first loan application information in client set.
Further, server by each first personally identifiable information for borrowing client got and receives application information input in advance
If the first prediction model, obtain each first borrowing client corresponding first and continuous borrowing prediction probability.Wherein, the first prediction model can be with
Machine learning is carried out by the loan data of the client for having continuous loan behavior to storing in server to train to obtain, wherein, continue and borrow
It refers to providing a loan in identical platform again.When carrying out model training, the model training mode of supervision may be used,
Such as regression tree model, Bayesian model, SVM (Support Vector Machine, support vector machines) etc..
In one embodiment, server can be according to the first continuous prediction probability of borrowing to each head in current first loan client set
It borrows client to be ranked up, obtains preliminary marketing priority orders, and ranking results are sent to marketing personnel's terminal, market people
Member's terminal can tentatively be marketed according to the ranking results.
Step S204 obtains each first monthly mortgage for borrowing client in preset time period in current first loan client set and compares number
According to obtaining each first borrowing client corresponding by second continuous to borrow prediction general using preset second prediction model than data according to monthly mortgage
Rate.
Specifically, monthly mortgage is than the ratio for month repayment amount and loan total amount, such as the Zhang San head total amounts borrowed
10000, the repayment amount of first month is 1000, and the repayment amount of second month is 2000, then the monthly mortgage of Zhang San's first month
Than being 10%, the repayment amount of second month is 20%.Preset time period is referred to can be according to specific marketing by marketing personnel
A period of time since the first month of refund that situation is previously set generally includes the time of at least two months, such as can
To be three months, five months etc..Server obtains each head in current first loan client set and borrows client in preset time period
Monthly mortgage refers to that server obtains each first borrow client in preset time period monthly mortgage ratio of each month than data, such as obtains each
Head borrows the monthly mortgage ratio of each month in continuous three months of client.
Further, in the present embodiment, server gets each first monthly mortgage for borrowing client in preset time period and compares number
According to rear, monthly mortgage is inputted into preset second prediction model than data and obtains each first borrowing client corresponding by second continuous to borrow prediction general
Rate.Wherein, the second prediction model can by the loan data to the client for having continuous loan behavior stored in server into
Row machine learning is trained to obtain.When carrying out model training, the model training mode of supervision, such as regression tree mould may be used
Type, Bayesian model, SVM (Support Vector Machine, support vector machines) etc..
In one embodiment, server can be according to the second continuous prediction probability of borrowing to working as after obtaining the second continuous loan prediction probability
Each first loan client is ranked up in preceding first loan client set, obtains preliminary marketing priority orders, and sequence is received and is sent out
It send to marketing personnel's terminal, marketing personnel's terminal can tentatively be marketed according to the ranking results.
Step S206, according to the first continuous prediction probability and the second continuous prediction probability of borrowing borrowed using preset third prediction model
It obtains each first corresponding synthesis of client of borrowing and continues loan prediction probability.
Specifically, it is comprehensive continuous to borrow prediction probability and each in client set first borrow client for characterizing current first borrow and carry out continuous loan
The possibility of behavior, the continuous loan prediction probability of synthesis is bigger, and it is bigger to represent that head loan clients carry out the continuous possibility borrowed.Third is predicted
Model carries out model training according to training data and obtains, for predict it is current first borrow in client set it is each it is first borrow client into
The continuous probability borrowed of row.In the present embodiment, server, will after the first continuous loan prediction probability, the second continuous loan prediction probability is obtained
First it is continuous borrow prediction probability, the second continuous prediction probability input third prediction model of borrowing obtains comprehensive continuous borrowing prediction probability.
Step S208 continues loan prediction probability according to the corresponding synthesis of each head loans client and each first loan client is ranked up
Ranking results are obtained, loan product information is pushed according to ranking results.
Specifically, server can continue according to the corresponding synthesis of each head loans client borrows probability to each first loan client progress
Descending or ascending order arrangement, obtain ranking results.Further, server pushes loan production according to ranking results to loan customer terminal
Product information, when ranking results are arranged for descending, selection comes the pushes customer loan production such as preceding 100 in the preset range of front
Product；When ranking results are arranged for ascending order, selection comes for example last 100 pushes customers loan of client in the preset range of front
Product.Wherein, loan product can be provide a loan the either random push loan product of the loan product currently promoted mainly of platform or
It is that the head borrows once browsed loan product of client etc..In one embodiment, it is continuous that a synthesis can also be preset
The threshold value of prediction probability is borrowed, only just carries out continuous loan product when the synthesis that head borrows client continues and borrows probability more than the predetermined threshold value
Push, so as to further save computer resource.
In one embodiment, since some loan products may have marketing limit, when being issued such as some loan products
Marketing limit is 100w, therefore can also be classified according to sequence to the first user for borrowing client set, if any 300 clients'
When, when carrying out descending arrangement, the client of ranking 1-100 is divided into the first kind, the client of ranking 101-200 is divided into second
Class carries out the client of ranking 201-300 to be divided into third class, first to first kind pushes customer loan product information, when first
When the marketing result of class client is more than the marketing limit of such loan product, stop to the second class pushes customer loan product, it is no
Then, continue to the second class pushes customer loan product, and so on.
Further, obtained ranking results can be sent to the corresponding terminal of marketing personnel by server.Wherein, server can
It is obtained with the ranking results in the carrying marketing personnel's user identifier for receiving the transmission of marketing personnel's terminal after asking according to the row
Sequence result obtains request and sends ranking results, can also select the marketing personnel user bound in advance after ranking results are obtained
It identifies corresponding terminal and sends ranking results in real time.After marketing personnel's terminal receives ranking results, corresponding marketing personnel can
It is selectively marketed according to the ranking results, so as to improve marketing success rate.
In above-mentioned loan data processing method, current first each first person for borrowing client in borrowing client set is being got
It, can be according to personally identifiable information and first application information of borrowing using preset first prediction model when part information and first loan application information
The corresponding first continuous loan prediction probability of each first loan client is obtained, each head borrows client and exists in obtaining current head and borrowing client set
When monthly mortgage in preset time period is than data, it can obtain than data each first borrowing visitor using preset second prediction model according to monthly mortgage
Family corresponding second, which continues, borrows prediction probability, then according to the first continuous prediction probability and the second continuous loan prediction probability borrowed using preset
Third prediction model obtains each first corresponding synthesis of client of borrowing and continues loan prediction probability, finally corresponding according to each first loan client
Comprehensive continuous prediction probability of borrowing is ranked up to obtain ranking results to each first loan client, and loan product letter is pushed according to ranking results
Breath, due to being predicted twice from different dimensions, and obtains integrated forecasting probability, this is comprehensive according to the result that this is predicted twice
The possibility that client continues loan can be reacted, and carry out the push for having target according to the size of the possibility by closing probability, can be saved
A large amount of computer resource.
In one embodiment, as shown in figure 3, the generation step of the first prediction model includes in step S202：
Step S302 obtains the personally identifiable information of each loan customer in history loan customer set, first loan letter of application
Breath and continuous loan probability.
Step S304 obtains the first training sample according to personally identifiable information, first loan application information and continuous probability of borrowing.
Step S306 carries out model training according to the first training sample and obtains the first prediction model.
Specifically, history loan customer set refers to having occurred and that the client of first loan behavior before current slot
The set formed.Since the continuous loan of each loan customer in history loan customer set is the result is that known, can will
People's identity information, first loan application information and continuous probability of borrowing obtain the first training sample, wherein, there is the loan customer of continuous loan behavior
Continuous probability of borrowing is 100%, and the continuous loan probability of the loan customer of not continuous loan behavior is 0.
First training sample obtains the first prediction model for being trained to model, with training.Carrying out model training
When, the model training mode of supervision, such as decision-tree model, Logic Regression Models, SVM (Support may be used
Vector Machine, support vector machines) and Bayesian model etc..Training sample can include positive sample and negative sample
This, when carrying out having the model training of supervision, there is the corresponding personally identifiable information of loan customer of continuous loan behavior, first loan letter of application
Breath and loan probability form positive sample, and the corresponding personally identifiable information of loan customer, the head of not continuous loan behavior borrow application information
And loan probability forms positive sample.
In the present embodiment, by by history loan set in each loan customer historical data be used as training sample come
It carries out model training and obtains the first prediction model, the accuracy rate of prediction can be improved.
In one embodiment, the generation step of the second prediction model includes in step S204：Obtain history loan customer
Monthly mortgage of each loan customer in preset time period compares data in set；According to monthly mortgage second is obtained than data and continuous probability of borrowing
Training sample；Model training is carried out according to the second training sample and obtains the second prediction model.
What deserves to be explained is in the present embodiment, if data whole in usage history loan customer set
It practises, it is easy to lead to over-fitting.So in one embodiment, consider cross validation when model training.Intersect
Verification is a kind of practical approach that data sample is statistically cut into relatively small subset, can first be divided in a subset
Analysis, and other subsets are then used for doing the subsequently confirmation and verification to this analysis.Subset at the beginning is referred to as training set, and its
Its subset is then referred to as verification collection or test set.The target of cross validation is the model for defining a data set to test,
Training stage is to reduce as the problem of over-fitting.
In one embodiment, as shown in figure 4, the generation step of third prediction model includes in step S206：
Step S402, according to the personally identifiable information of each loan customer in history loan customer set and first loan letter of application
Breath obtains the corresponding first continuous loan prediction probability of each loan customer using preset first prediction model.
Specifically, server gets the personally identifiable information of each loan customer and first loan in history loan customer set
After information, these information are inputted into the trained first continuous loan prediction model, obtain each loan customer corresponding first
It is continuous to borrow prediction probability.
Step S404 compares data according to monthly mortgage of each loan customer in preset time period in history loan customer set
The corresponding second continuous loan prediction probability of each loan customer is obtained using preset second prediction model.
Specifically, server each loan customer in history loan customer set is got is each in preset time period
After the monthly mortgage ratio of the moon, these monthly mortgages are input to the second prediction model than data, it is pre- to obtain the corresponding second continuous loan of each client
Survey probability.It is appreciated that in the present embodiment, predicted time section phase when preset time period is trained with the second prediction model
Together, e.g., for the second prediction model when carrying out model training, the data that training set includes are the monthly mortgage ratio of first trimester, then are obtaining
Into history loan set during the second continuous loan prediction probability of each loan customer, the monthly mortgage of acquisition is similarly first three than data
The monthly mortgage ratio of the moon.
Step S406 continuous borrows prediction probability, the according to loan customer each in history loan customer set corresponding first
Two continuous loan prediction probabilities and continuous probability of borrowing obtain third training sample, and carrying out model training according to third training sample obtains third
Prediction model.
Further, due to the continuous loan situation of each loan customer in history loan customer set be it is known, can root
Continue loan prediction probability, the second continuous prediction probability and continuous probability of borrowing borrowed as training set progress model according to the first of each loan customer
Training obtains third prediction model, wherein, the loan probability for having the loan customer of continuous loan behavior is 100%, not continuous loan behavior
The continuous loan of loan customer be summarised as 0, it is in the present embodiment, similary using the model training for having supervision when carrying out model training
Mode, such as Logic Regression Models, SVM (Support Vector Machine, support vector machines) and Bayesian model etc..
In one embodiment, the third prediction model Y=α of an initialization can be set0+α1X1+α2X2, wherein, Y generations
Table integrated forecasting probability, X1Represent the first prediction probability, X2Represent the second prediction probability.In the prediction model for establishing initialization
Afterwards, need to build the training set that study is trained to the third prediction model model of the initialization, by existing customer data
The first prediction model, the second prediction model are substituted into, obtains the first prediction probability, the second prediction probability, by the first prediction probability,
Two prediction probabilities carry out mould to the prediction model of initialization with corresponding continuous result composition training set of borrowing using machine learning algorithm
To get corresponding coefficient value, when carrying out model training, the model training mode of supervision may be used in type training,
Such as Logic Regression Models, SVM (Support Vector Machine, support vector machines) and Bayesian model etc..
By taking SVM as an example, stochastic gradient algorithm may be used and carry out model training, it is required that generation in gradient descent procedures
Valency function J (θ) is minimum, and in one embodiment, the following formula expression may be used in cost function：It represents, wherein, m represents the number of sample characteristics, x(i)
For input, y(i)Represent the value of the integrated forecasting probability in training sample, positive sample can be set as 1, and negative sample can be set as -1.
hθ(x(i)) represent every time trained output valve, wherein,Wherein,That is θTX be equal to the product of feature and parameter and.
In the present embodiment, by by history loan set in each loan customer historical data be used as training sample come
It carries out model training and obtains the second prediction model, the accuracy rate of prediction can be improved.
In one embodiment, as shown in figure 5, step S208 includes：
Step S502 obtains continuous loan client set according to history loan customer set.
Wherein, continue loan client set and refer to the set that the client by continuous loan behavior is formed.
Step S504 chooses head in current first loan client set according to ranking results and borrows client as current goal visitor
Family.
Specifically, since ranking results are ranked up according to the comprehensive continuous size for borrowing prediction probability, it is sorting
Afterwards, the forward client of comprehensive continuous loan prediction probability can be selected as current goal visitor by server successively according to the ranking results
Family, for example, when ranking results for descending arrange when, since the client to make number one, successively will preceding 100 clients as ought
Preceding target customer.
Step S506, by current goal client and it is continuous borrow each continuous loan client mappings in client set arrive with head borrow information with
Identity information is the vectorial coordinate in the multi-C vector space of coordinate, respectively obtaining current goal client and each continuous loan client couple
The vectorial coordinate answered.
Specifically, first information of borrowing is various credit informations and refund information of the client when provide a loan for the first time.At one
In embodiment, head borrows information and includes the first loan amount of the loan, refund issue.Identity information includes age, gender, monthly income etc..
In the present embodiment, information and identity information are borrowed as in the multi-C vector space of coordinate, each dimension represents a difference using head
Information, server is by current goal client and continuous borrow each client mappings in client set and arrives with first loan information and identity information
For in the multi-C vector space of coordinate, each client can uniquely correspond to the space vector of one, the space vector using origin as
Starting point, therefore, vectorial coordinate can be represented with the coordinate of terminal.It is understood that the dimension of vector space and each
The corresponding information of a dimension can be previously set as needed, and dimension is more, the obtained customer information representated by space vector
It is more accurate.
In one embodiment, when vector space is to borrow the amount of the loan, refund issue, age as x-axis, y using head
The three-dimensional vector space of axis, z-axis, the then corresponding vectorial coordinate of each client obtained are (x, y, z), as the head of target customer is borrowed
The amount of the loan is 100000, and refund issue is 5, and the age 30, then the corresponding vectorial coordinate of the client is (100000,5,30).
Step S508 calculates current goal client and the continuous vector distance borrowed in client set between each continuous loan client.
Specifically, vector distance can calculate by the following method：If the continuous vectorial coordinate for borrowing client is (x1, y1,
z1... ...), the vectorial coordinate of target customer is (x2, y2, z2... ...), then vector distance is：
Step S510, corresponding continuous loan client is as the reference pair of current goal client when vector distance is less than preset value
As obtaining references object set, loan product information being pushed according to references object set.
Specifically, the vector distance between target customer may have multiple less than the continuous loan client of preset value, by these
The continuous references object for borrowing client as target customer, obtains references object set, according to the references object set to target customer
Push loan product.
In one embodiment, loan product is pushed to target customer according to the references object set to include：Inquiry reference
The continuous loan product of each references object in object set obtains continuous loan product set；Obtain each continuous loan in continuous loan product set
The corresponding references object quantity of product；Target push product is obtained according to each continuous corresponding references object quantity of product of borrowing.
In the present embodiment, server inquires each references object in references object set from the historical data of preservation
It is continuous to borrow product, obtain it is continuous borrow product set, and count continuous and borrow the corresponding continuous loan number of each continuous loan product in product set, choose
The most continuous loan product of continuous loan number pushes product as target.It, can when the most continuous loan product quantity of continuous loan number is more than 1
These continuous loan products are all pushed into product as target, a continuous product of borrowing can also be randomly choosed and push production as target
Product.
Further, server obtains the corresponding user identifier of target customer, and target push product information is pushed to the use
Family identifies corresponding terminal.
In above-described embodiment, by determining references object, then the continuous loan product of references object is counted, is continued
The most continuous loan product of loan number pushes product as target, can so as to targetedly carry out the push of loan product
To improve marketing hit rate when server carries out loan product push.
It should be understood that although each step in the flow chart of Fig. 2-5 is shown successively according to the instruction of arrow,
These steps are not that the inevitable sequence indicated according to arrow performs successively.Unless it expressly states otherwise herein, these steps
Execution there is no stringent sequences to limit, these steps can perform in other order.Moreover, at least one in Fig. 2-5
Part steps can include multiple sub-steps, and either these sub-steps of multiple stages or stage are not necessarily in synchronization
Completion is performed, but can be performed at different times, the execution sequence in these sub-steps or stage is also not necessarily successively
It carries out, but can either the sub-step of other steps or at least part in stage be in turn or alternately with other steps
It performs.
In one embodiment, as shown in fig. 6, providing a kind of loan data processing unit 600, including：First continuous loan
Prediction probability acquisition module 602, second is continuous to borrow prediction probability acquisition module 604, comprehensive continuous loan prediction probability module 606 and sequence
Module 608, wherein：First continuous loan prediction probability acquisition module 602, for obtaining each first loan visitor in current first loan client set
The personally identifiable information at family and first loan application information, it is pre- using preset first according to personally identifiable information and first loan application information
It surveys model and obtains the corresponding first continuous loan prediction probability of each first loan client；Second continuous loan prediction probability acquisition module 604, is used for
Each first monthly mortgage for borrowing client in preset time period in current first loan client set is obtained, than data, to be adopted than data according to monthly mortgage
The corresponding second continuous loan prediction probability of each first loan client is obtained with preset second prediction model；Comprehensive continuous loan prediction probability mould
Block 606, for being obtained respectively using preset third prediction model according to the first continuous loan prediction probability and the second continuous prediction probability of borrowing
A first corresponding synthesis of client of borrowing continues loan prediction probability；Sorting module 608, for being continued according to each first corresponding synthesis of client of borrowing
It borrows prediction probability each first loan client is ranked up to obtain ranking results, loan product information is pushed according to ranking results.
In one embodiment, above device further includes：First prediction model establishes module, for obtaining history loan visitor
The personally identifiable information of each loan customer, first loan application information and continuous loan probability in the set of family, according to personally identifiable information, head
It borrows application information and continuous probability of borrowing obtains the first training sample, carrying out model training according to the first training sample obtains the first prediction
Model.
In one embodiment, above device further includes：Second prediction model establishes module, for obtaining history loan visitor
Monthly mortgage of each loan customer in preset time period be than data in the set of family, and the is obtained than data and continuous probability of borrowing according to monthly mortgage
Two training samples carry out model training according to the second training sample and obtain the second prediction model.
In one embodiment, above device further includes：Third prediction model establishes module, objective for being provided a loan according to history
The personally identifiable information of each loan customer and first application information of borrowing are obtained respectively using preset first prediction model in the set of family
A loan customer corresponding first, which continues, borrows prediction probability, according to each loan customer in history loan customer set in preset time
Monthly mortgage in section obtains the corresponding second continuous loan prediction probability of each loan customer than data using preset second prediction model,
According to the corresponding first continuous loan prediction probability of loan customer each in history loan customer set, the second continuous loan prediction probability and continue
It borrows probability and obtains third training sample, carrying out model training according to third training sample obtains third prediction model.
In one embodiment, as shown in fig. 7, above device further includes：It is continuous to borrow client set acquisition module 702, it is used for
Continuous loan client set is obtained according to history loan customer set；Target customer chooses module 704, for being worked as according to ranking results
Preceding first borrow chooses first loan client as current goal client in client set；Vectorial coordinate acquisition module 706, for by current mesh
It marks each continuous client mappings of borrowing in client and continuous loan client set and borrows the multi-C vector of information and identity information as coordinate to using first
In space, the vectorial coordinate of current goal client and the corresponding vectorial coordinate of each continuous loan client are respectively obtained；Vector distance meter
Module 708 is calculated, for calculating current goal client and the continuous vector distance borrowed in client set between each continuous loan client；With reference to
Object set acquisition module 710, for using vector distance be less than preset value when it is corresponding it is continuous loan client as current goal client
References object, obtain references object set, according to references object set push loan product information.
In one embodiment, above device further includes：Target pushes product acquisition module, for inquiring references object collection
The continuous loan product of each references object in conjunction obtains continuous loan product set, obtains each continuous loan product pair in continuous loan product set
The references object quantity answered obtains target push product according to each continuous corresponding references object quantity of product of borrowing.
In one embodiment, above device further includes：Product pushing module, for target push product to be pushed
To the corresponding terminal of current goal client.
Specific restriction about loan data processing unit may refer to the limit above for loan data processing method
Fixed, details are not described herein.Modules in above-mentioned loan data processing unit can fully or partially through software, hardware and its
It combines to realize.Above-mentioned each module can be embedded in or in the form of hardware independently of in the processor in computer equipment, can also
It is stored in a software form in the memory in computer equipment, in order to which processor calls execution more than modules 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 8.The computer equipment include the processor connected by system bus, memory, network interface and
Database.Wherein, the processor of the computer equipment is for offer 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 operating system in non-volatile memory medium and the operation of computer program.The calculating
The database of machine equipment handles data for storing loan data.The network interface of the computer equipment is used for and external terminal
It is communicated by network connection.To realize a kind of loan data processing method when the computer program is executed by processor.
It will be understood by those skilled in the art that the structure shown in Fig. 8, only part knot relevant with application scheme
The block diagram of structure does not form the restriction for the computer equipment being applied thereon to application scheme, specific computer equipment
It can include either combining certain components than components more or fewer shown in figure or be arranged with different components.
In one embodiment, a kind of computer equipment is provided, including memory, processor and storage on a memory
And the computer program that can be run on a processor, processor realize following steps when performing computer program：It obtains current first
It borrows each head in client set and borrows the personally identifiable information of client and first loan application information, according to personally identifiable information and first loan Shen
Please information obtain each first borrowing client corresponding first and continuous borrowing prediction probability using preset first prediction model；It obtains current first
Each first monthly mortgage of the client in preset time period of borrowing is borrowed in client set than data, according to monthly mortgage than data using preset the
Two prediction models obtain the corresponding second continuous loan prediction probability of each first loan client；Continued according to the first continuous prediction probability and second of borrowing
Loan prediction probability obtains each first corresponding synthesis of client of borrowing using preset third prediction model and continues loan prediction probability；According to each
It is a it is first borrow the corresponding synthesis of client continue borrow prediction probability to it is each it is first borrow client and be ranked up to obtain ranking results, tied according to sequence
Fruit pushes loan product information.
In one embodiment, following steps are also realized when processor performs computer program：Obtain history loan customer
The personally identifiable information of each loan customer, first loan application information and continuous loan probability in set；According to personally identifiable information, first loan
Application information and continuous probability of borrowing obtain the first training sample；Model training is carried out according to the first training sample and obtains the first prediction mould
Type.
In one embodiment, following steps are also realized when processor performs computer program：Obtain history loan customer
Monthly mortgage of each loan customer in preset time period compares data in set；According to monthly mortgage second is obtained than data and continuous probability of borrowing
Training sample；Model training is carried out according to the second training sample and obtains the second prediction model.
In one embodiment, following steps are also realized when processor performs computer program：According to history loan customer
The personally identifiable information of each loan customer and first loan application information are obtained each using preset first prediction model in set
The corresponding first continuous loan prediction probability of loan customer；According to each loan customer in history loan customer set in preset time period
Interior monthly mortgage obtains that each loan customer is corresponding by second continuous to borrow prediction probability than data using preset second prediction model；Root
Continue to borrow prediction probability and continue according to the corresponding first continuous loan prediction probability of loan customer each in history loan customer set, second and borrow
Probability obtains third training sample, and carrying out model training according to third training sample obtains third prediction model.
In one embodiment, following steps are also realized when processor performs computer program：According to history loan customer
Set obtains continuous loan client set；Head is chosen in current first loan client set borrow client as current goal according to ranking results
Client；By current goal client and it is continuous borrow each continuous loan client mappings in client set and arrive information and identity information are borrowed as seat using head
In target multi-C vector space, the vectorial coordinate and the corresponding vector of each continuous loan client that respectively obtain current goal client are sat
Mark；Calculate current goal client and the continuous vector distance borrowed in client set between each continuous loan client；Vector distance is less than
The corresponding continuous references object for borrowing client as current goal client, obtains references object set, according to reference pair during preset value
As set pushes loan product information.
In one embodiment, in one embodiment, following steps are also realized when processor performs computer program：It looks into
The continuous loan product of each references object in references object set is ask, obtains continuous loan product set；Obtain continuous borrow in product set respectively
The corresponding references object quantity of a continuous loan product；Target push production is obtained according to each continuous corresponding references object quantity of product of borrowing
Product.
In one embodiment, following steps are also realized when processor performs computer program：The target is pushed and is produced
Product push to the corresponding terminal of current goal client.
In one embodiment, a kind of computer readable storage medium is provided, is stored thereon with computer program, is calculated
Machine program realizes following steps when being executed by processor：Obtain the current first personal identification borrowed each head in client set and borrow client
Information and first loan application information, are obtained respectively according to personally identifiable information and first application information of borrowing using preset first prediction model
A first loan client corresponding first, which continues, borrows prediction probability；It obtains each head in current first loan client set and borrows client in preset time
It is corresponding by that monthly mortgage in section than data, according to monthly mortgage than data obtains each first loan client using preset second prediction model
Two continuous loan prediction probabilities；It is obtained according to the first continuous loan prediction probability and the second continuous prediction probability of borrowing using preset third prediction model
Continue loan prediction probability to each first corresponding synthesis of client of borrowing；Loan prediction probability pair is continued according to each first corresponding synthesis of client of borrowing
Each first loan client is ranked up to obtain ranking results, and loan product information is pushed according to ranking results.
In one embodiment, following steps are also realized when computer program is executed by processor：Obtain history loan visitor
The personally identifiable information of each loan customer, first loan application information and continuous loan probability in the set of family；According to personally identifiable information, head
It borrows application information and continuous probability of borrowing obtains the first training sample；Model training is carried out according to the first training sample and obtains the first prediction
Model.
In one embodiment, following steps are also realized when computer program is executed by processor：Obtain history loan visitor
Monthly mortgage of each loan customer in preset time period compares data in the set of family；According to monthly mortgage the is obtained than data and continuous probability of borrowing
Two training samples；Model training is carried out according to the second training sample and obtains the second prediction model.
In one embodiment, following steps are also realized when computer program is executed by processor：It is provided a loan according to history objective
The personally identifiable information of each loan customer and first application information of borrowing are obtained respectively using preset first prediction model in the set of family
A loan customer corresponding first, which continues, borrows prediction probability；According to each loan customer in history loan customer set in preset time
Monthly mortgage in section obtains the corresponding second continuous loan prediction probability of each loan customer than data using preset second prediction model；
According to the corresponding first continuous loan prediction probability of loan customer each in history loan customer set, the second continuous loan prediction probability and continue
It borrows probability and obtains third training sample, carrying out model training according to third training sample obtains third prediction model.
In one embodiment, following steps are also realized when computer program is executed by processor：It is provided a loan according to history objective
Gather to obtain continuous loan client set in family；Head is chosen in current first loan client set borrow client as current mesh according to ranking results
Mark client；By current goal client and it is continuous borrow each continuous loan client mappings in client set arrive using first loan information and identity information as
In the multi-C vector space of coordinate, the vectorial coordinate and the corresponding vector of each continuous loan client that respectively obtain current goal client are sat
Mark；Calculate current goal client and the continuous vector distance borrowed in client set between each continuous loan client；Vector distance is less than
The corresponding continuous references object for borrowing client as current goal client, obtains references object set, according to reference pair during preset value
As set pushes loan product information.
In one embodiment, following steps are also realized when computer program is executed by processor：Inquire references object collection
The continuous loan product of each references object in conjunction obtains continuous loan product set；Obtain each continuous loan product pair in continuous loan product set
The references object quantity answered；Target push product is obtained according to each continuous corresponding references object quantity of product of borrowing.
In one embodiment, following steps are also realized when computer program is executed by processor：The target is pushed
Product pushes to the corresponding terminal of current goal client.
One of ordinary skill in the art will appreciate that realizing all or part of flow 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 flow of the embodiment of above-mentioned each method.Wherein,
Any reference to 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 example can be combined arbitrarily, to make description succinct, 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 is all considered to be the range of this specification record.
Embodiment described above only expresses the several embodiments of the application, and description 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 those of ordinary skill in the art are come
It says, under the premise of the application design is not departed from, various modifications and improvements can be made, these belong to the protection of the application
Range.Therefore, the protection domain of the application patent should be determined by the appended claims.
Claims (10)
1. a kind of loan data processing method, the method includes：
It obtains current head and borrows each first personally identifiable information for borrowing client and first loan application information in client set, according to described
People's identity information and head borrow application information and obtain the corresponding first continuous loan of each first loan client using preset first prediction model
Prediction probability；
It obtains each head in current first loan client set and borrows monthly mortgage of the client in preset time period than data, according to the monthly mortgage
It obtains using preset second prediction model each first borrowing that client is corresponding by second continuous to borrow prediction probability than data；
It is obtained according to the described first continuous loan prediction probability and the second continuous prediction probability of borrowing using preset third prediction model
Each first corresponding synthesis of client of borrowing continues loan prediction probability；
Continue loan prediction probability according to the corresponding synthesis of each head loans client each first loan client is ranked up to obtain ranking results,
Loan product information is pushed according to the ranking results.
2. according to the method described in claim 1, it is characterized in that, the generation step of first prediction model includes：
Obtain the personally identifiable information of each loan customer in history loan customer set, first loan application information and continuous loan probability；
First training sample is obtained according to the personally identifiable information, first loan application information and continuous probability of borrowing；
Model training is carried out according to first training sample and obtains the first prediction model.
3. according to the method described in claim 2, it is characterized in that, the generation step of second prediction model includes：
It obtains monthly mortgage of each loan customer in preset time period in history loan customer set and compares data；
Second training sample is obtained than data and continuous probability of borrowing according to the monthly mortgage；
Model training is carried out according to second training sample and obtains the second prediction model.
4. according to the method described in claim 3, it is characterized in that, the generation step of the third prediction model includes：
According to the personally identifiable information of each loan customer in the history loan customer set and first application information of borrowing using pre-
If the first prediction model obtain each loan customer corresponding first and continuous borrow prediction probability；
According to monthly mortgage of each loan customer in preset time period in the history loan customer set than data using default
The second prediction model obtain each loan customer corresponding second and continuous borrow prediction probability；
According to the corresponding first continuous loan prediction probability of loan customer each in history loan customer set, the second continuous loan prediction probability
And continuous probability of borrowing obtains third training sample, carrying out model training according to the third training sample obtains third prediction model.
5. according to the method described in claim 2-4 any one, which is characterized in that described corresponding according to each first loan client
Comprehensive continuous prediction probability of borrowing is ranked up to obtain ranking results to each first loan client, and loan production is pushed according to the ranking results
The step of product information, includes：
Continuous loan client set is obtained according to the history loan customer set；
Head is chosen in current first loan client set borrow client as current goal client according to the ranking results；
By current goal client and it is continuous borrow each continuous loan client mappings in client set and arrive information and identity information are borrowed as seat using head
In target multi-C vector space, the vectorial coordinate and the corresponding vector of each continuous loan client that respectively obtain current goal client are sat
Mark；
Calculate current goal client and the continuous vector distance borrowed in client set between each continuous loan client；
Corresponding continuous loan client obtains reference pair as the references object of current goal client when vector distance is less than preset value
As set；
Loan product information is pushed according to the references object set.
6. according to the method described in claim 5, it is characterized in that, described push loan product according to the references object set
The step of information, includes：
The continuous loan product of each references object in references object set is inquired, obtains continuous loan product set；
It obtains and continuous borrows the corresponding references object quantity of each continuous loan product in product set；
Current goal client correspondence is obtained according to the continuous corresponding references object quantity of product of borrowing each in the continuous loan product set
Target push product.
7. according to the method described in claim 6, it is characterized in that, the method further includes：
Target push product is pushed into the corresponding terminal of current goal client.
8. a kind of loan data processing unit, which is characterized in that described device includes：
First continuous loan prediction probability acquisition module, for obtaining the current first personal identification for borrowing each first loan client in client set
Information and first loan application information, are obtained according to the personally identifiable information and first application information of borrowing using preset first prediction model
Continue to each first loan client corresponding first and borrow prediction probability；
Second it is continuous borrow prediction probability acquisition module, each in client set first borrow client in preset time for obtaining current first borrow
Monthly mortgage in section obtains each first loan client using preset second prediction model than data according to the monthly mortgage and corresponds to than data
Second continuous borrow prediction probability；
Comprehensive continuous loan prediction probability module, for being adopted according to the described first continuous loan prediction probability and the second continuous prediction probability of borrowing
Each first corresponding synthesis of client of borrowing, which is obtained, with preset third prediction model continues loan prediction probability；
Sorting module is ranked up each first loan client for continuing loan prediction probability according to the corresponding synthesis of each head loans client
Ranking results are obtained, loan product information is pushed according to the ranking results.
9. a kind of computer equipment including memory, processor and stores the meter that can be run on a memory and on a processor
Calculation machine program, which is characterized in that the processor realizes any one of claim 1 to 7 institute when performing the computer program
The step of stating method.
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 claim 1 to 7 is realized when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810091577.6A CN108182633B (en) | 2018-01-30 | 2018-01-30 | Loan data processing method, loan data processing device, loan data processing program, and computer device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810091577.6A CN108182633B (en) | 2018-01-30 | 2018-01-30 | Loan data processing method, loan data processing device, loan data processing program, and computer device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108182633A true CN108182633A (en) | 2018-06-19 |
| CN108182633B CN108182633B (en) | 2020-08-04 |
Family
ID=62551877
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810091577.6A Active CN108182633B (en) | 2018-01-30 | 2018-01-30 | Loan data processing method, loan data processing device, loan data processing program, and computer device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108182633B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109165983A (en) * | 2018-09-04 | 2019-01-08 | 中国平安人寿保险股份有限公司 | Insurance products recommended method, device, computer equipment and storage medium |
| CN110321393A (en) * | 2019-06-28 | 2019-10-11 | 深圳前海微众银行股份有限公司 | Method for computing data, device, equipment and readable storage medium storing program for executing |
| CN111582898A (en) * | 2019-02-18 | 2020-08-25 | 北京奇虎科技有限公司 | Data processing method, device, equipment and storage medium |
| CN112116405A (en) * | 2020-09-29 | 2020-12-22 | 中国银行股份有限公司 | Data processing method, device, electronic equipment and medium |
| CN112541817A (en) * | 2020-12-22 | 2021-03-23 | 建信金融科技有限责任公司 | Marketing response processing method and system for potential customers of personal consumption loan |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090248567A1 (en) * | 2008-03-28 | 2009-10-01 | American Express Travel Related Services Company, Inc. | Consumer behaviors at lender level |
| CN104978689A (en) * | 2015-06-19 | 2015-10-14 | 徐晓晖 | Stock market combination characteristic based network loan risk quantitative analysis method |
| CN106600369A (en) * | 2016-12-09 | 2017-04-26 | 广东奡风科技股份有限公司 | Real-time recommendation system and method of financial products of banks based on Naive Bayesian classification |
-
2018
- 2018-01-30 CN CN201810091577.6A patent/CN108182633B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090248567A1 (en) * | 2008-03-28 | 2009-10-01 | American Express Travel Related Services Company, Inc. | Consumer behaviors at lender level |
| CN104978689A (en) * | 2015-06-19 | 2015-10-14 | 徐晓晖 | Stock market combination characteristic based network loan risk quantitative analysis method |
| CN106600369A (en) * | 2016-12-09 | 2017-04-26 | 广东奡风科技股份有限公司 | Real-time recommendation system and method of financial products of banks based on Naive Bayesian classification |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109165983A (en) * | 2018-09-04 | 2019-01-08 | 中国平安人寿保险股份有限公司 | Insurance products recommended method, device, computer equipment and storage medium |
| CN111582898A (en) * | 2019-02-18 | 2020-08-25 | 北京奇虎科技有限公司 | Data processing method, device, equipment and storage medium |
| CN110321393A (en) * | 2019-06-28 | 2019-10-11 | 深圳前海微众银行股份有限公司 | Method for computing data, device, equipment and readable storage medium storing program for executing |
| CN112116405A (en) * | 2020-09-29 | 2020-12-22 | 中国银行股份有限公司 | Data processing method, device, electronic equipment and medium |
| CN112116405B (en) * | 2020-09-29 | 2024-02-02 | 中国银行股份有限公司 | Data processing method, device, electronic equipment and medium |
| CN112541817A (en) * | 2020-12-22 | 2021-03-23 | 建信金融科技有限责任公司 | Marketing response processing method and system for potential customers of personal consumption loan |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108182633B (en) | 2020-08-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI788529B (en) | Credit risk prediction method and device based on LSTM model | |
| CN108182633A (en) | Loan data processing method, device, computer equipment and storage medium | |
| CN110060144B (en) | Method for training credit model, method, device, equipment and medium for evaluating credit | |
| CN109783730A (en) | Products Show method, apparatus, computer equipment and storage medium | |
| CN109165983A (en) | Insurance products recommended method, device, computer equipment and storage medium | |
| CN110363645A (en) | Asset data processing method, device, computer equipment and storage medium | |
| CN109272408A (en) | Vehicle loan financial product intelligent recommendation method, apparatus, equipment and medium | |
| CN109767319A (en) | The accrediting amount determines method, apparatus, computer equipment and storage medium | |
| CN109615280A (en) | Employee's data processing method, device, computer equipment and storage medium | |
| CN110415036B (en) | User grade determining method, device, computer equipment and storage medium | |
| CN109784654A (en) | Task creating method, device, computer equipment and storage medium | |
| CN113918884A (en) | Traffic prediction model construction method and traffic prediction method | |
| CN109583682A (en) | Recognition methods, device and the computer equipment of business finance fraud risk | |
| CN109801101A (en) | Label determines method, apparatus, computer equipment and storage medium | |
| CN108334625A (en) | Processing method, device, computer equipment and the storage medium of user information | |
| CN110728301A (en) | Credit scoring method, device, terminal and storage medium for individual user | |
| CN109325868A (en) | Questionnaire data processing method, device, computer equipment and storage medium | |
| CN109360111A (en) | Questionnaire data modification method, device, computer equipment and storage medium | |
| CN110866637B (en) | Scoring prediction method, scoring prediction device, computer equipment and storage medium | |
| CN111310931A (en) | Parameter generation method and device, computer equipment and storage medium | |
| CN115796548A (en) | Resource allocation method, device, computer equipment, storage medium and product | |
| CN114925919A (en) | Service resource processing method and device, computer equipment and storage medium | |
| CN114925275A (en) | Product recommendation method and device, computer equipment and storage medium | |
| CN113779116A (en) | Object sorting method, related equipment and medium | |
| WO2017004348A1 (en) | Normalizing data sets for predicting an attribute of the data sets |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |