CA3135466A1 - User loan willingness prediction method and device and computer system - Google Patents
User loan willingness prediction method and device and computer system Download PDFInfo
- Publication number
- CA3135466A1 CA3135466A1 CA3135466A CA3135466A CA3135466A1 CA 3135466 A1 CA3135466 A1 CA 3135466A1 CA 3135466 A CA3135466 A CA 3135466A CA 3135466 A CA3135466 A CA 3135466A CA 3135466 A1 CA3135466 A1 CA 3135466A1
- Authority
- CA
- Canada
- Prior art keywords
- user
- predicting
- model
- user data
- loan
- 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 42
- 230000006399 behavior Effects 0.000 claims abstract description 34
- 230000008569 process Effects 0.000 claims abstract description 16
- 238000012549 training Methods 0.000 claims description 51
- 238000012216 screening Methods 0.000 abstract description 5
- 238000002474 experimental method Methods 0.000 abstract description 2
- 238000007477 logistic regression Methods 0.000 description 17
- 239000013598 vector Substances 0.000 description 8
- 238000004891 communication Methods 0.000 description 6
- 230000004927 fusion Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000007418 data mining Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 241000699670 Mus sp. Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- 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
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Economics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Development Economics (AREA)
- Evolutionary Computation (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Business, Economics & Management (AREA)
- Technology Law (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The present invention discloses to a predicting of user's loan willingness method, apparatus, and computer system. The method comprises: obtaining user data of to be predicted user, the user data comprises user properties and historical consuming behaviors; using a plurality of pre-set classifiers of first pre-set model, generating predicting result of the to be predicted user's category according to the user data respectively, the pre-set classifiers are used for classifying users; integrating predicting result of each pre-set classifier to generate predicting features; and using second pre-set model to predict whether the to be predicted user needs loan according to the predicting features, predicting user loan willingness according to obtained user data which has avoided manual screening process on user data, avoided inaccurate predicting led by inaccurate manual screening, shortened time period required for features experiment, and improved accuracy and efficiency of model predicting.
Description
USER LOAN WILLINGNESS PREDICTION METHOD AND DEVICE AND
COMPUTER SYSTEM
Field [0001] The present disclosure relates to machine learning field, particularly a predicting of user's loan willingness method, apparatus, and system.
Background
COMPUTER SYSTEM
Field [0001] The present disclosure relates to machine learning field, particularly a predicting of user's loan willingness method, apparatus, and system.
Background
[0002] In recent years, the use of data mining technology in financial industry has become more and more extensive and general. The data mining technology can help financial industry analyze and evaluate customers, help financial industry better dig out potential information, rules and use them to provide more suitable products to customers.
[0003] For financial industry, how to make full use of existing data indicator resource through effective algorithm model to dig out customers with loan request and willingness is the urgent problem in the present field. In the prior art, score card model and some other linear models are often used to evaluate and predict whether customers have loan request and willingness.
However, the learning ability of these models is limited, and manual pre-screening is required for features, a large number of features process's pre-analysis is also required, predicting according to determined effective features and combination of features takes a long time and it can not guarantee the predicting accuracy.
Invention Content
However, the learning ability of these models is limited, and manual pre-screening is required for features, a large number of features process's pre-analysis is also required, predicting according to determined effective features and combination of features takes a long time and it can not guarantee the predicting accuracy.
Invention Content
[0004] To solve the shortcomings of the prior art, the main purpose of the present invention is to provide a predicting of user's loan willingness method, apparatus, and computer system.
[0005] To achieve the above-mentioned purpose, the first aspect of the present invention provides a method for predicting user's loan willingness, the method comprises:
Date recue / Date received 202 1-1 1-29
Date recue / Date received 202 1-1 1-29
[0006] Obtaining user data of to be predicted user, the user data comprises user properties and historical consuming behaviors;
[0007] Using a plurality of pre-set classifiers of first pre-set model, generating predicting result of the to be predicted user's category according to the user data respectively, the pre-set classifiers are used for classifying users;
[0008] Integrating predicting result of each pre-set classifier to generate predicting features;
[0009] Using second pre-set model to predict whether the to be predicted user needs loan according to the predicting features.
[0010] In some implementations, the first pre-set model is obtained through iterative training, the iterative training process of the first pre-set model comprises:
[0011] Using obtained first training dataset to train first pre-set model, generating pre-set classifier, the first training dataset includes historical user data and whether correspondingly user has loan behavior within a pre-set time period;
[0012] According to residuals of pre-set classifier generated in previous round of iterative training, the first pre-set model is iterative trained to generate correspondingly pre-set classifier, until the first pre-set model meets the correspondingly pre-set conditions, the first pre-set model consists of a plurality of pre-set classifiers.
[0013] In some implementations, the first pre-set model comprises trained gradient boosting random tree model, the second pre-set model comprises trained logistic regression model.
[0014] In some implementations, training process of the second pre-set model comprises:
[0015] According to predicting result of each pre-set classifier of the first pre-set model, Date recue / Date received 202 1-1 1-29 generating predicting features corresponding to each historical user data;
[0016] According to predicting features corresponding to the historical user data and whether correspondingly user has loan behavior within a pre-set time period, generating second training dataset;
[0017] Using the second training dataset to train the second pre-set model.
[0018] In some implementations, integrating predicting result of each pre-set classifier to generate predicting features, comprising:
[0019] Integrating the predicting result according to generation order of the pre-set classifiers.
[0020] In some implementations, the method comprises:
[0021] When the to be predicted user is predicted to need loan, determining the to be predicted user as target user, so as to provide loan product to the target user.
[0022] The second aspect is that the present invention provides a predicting apparatus of user's loan willingness, wherein, the apparatus comprises:
[0023] An obtaining module for obtaining user data of to be predicted user, the user data comprises user properties and historical consuming behaviors;
[0024] A process module for using a plurality of pre-set classifiers of first pre-set model, generating predicting result of the to be predicted user's category according to the user data respectively, the pre-set classifiers are used for classifying users;
integrating predicting result of each pre-set classifier to generate predicting features; and using second pre-set model to predict whether the to be predicted user needs loan according to the predicting features.
Date recue / Date received 202 1-1 1-29
integrating predicting result of each pre-set classifier to generate predicting features; and using second pre-set model to predict whether the to be predicted user needs loan according to the predicting features.
Date recue / Date received 202 1-1 1-29
[0025] In some implementations, the apparatus comprises a training module configured to use first training dataset to train first pre-set model and generate pre-set classifier, the first training dataset includes historical user data and whether correspondingly user has loan behavior within a pre-set time period; according to residuals of pre-set classifier generated in previous round of iterative training, the first pre-set model is iterative trained to generate correspondingly pre-set classifier, until the first pre-set model meets the correspondingly pre-set conditions, the first pre-set model consists of a plurality of pre-set classifiers.
[0026] In some implementations, the training module is also used for according to predicting result of each pre-set classifier of the first pre-set model, generating predicting features corresponding to each historical user data; according to predicting features corresponding to the historical user data and whether correspondingly user has loan behavior within a pre-set time period, generating second training dataset; and using the second training dataset to train the second pre-set model.
[0027] The third aspect of the present invention provides a computer system, the system comprises:
[0028] One or a plurality of processors;
[0029] A memory associated with one or a plurality of processors, the memory is configured to store program commands, if the program commands are executed by one or a plurality of processors, executing following operations:
[0030] Obtaining user data of to be predicted user, the user data comprises user properties and historical consuming behaviors;
[0031] Using a plurality of pre-set classifiers of first pre-set model, generating predicting result of the to be predicted user's category according to the user data respectively, the pre-set classifiers are used for classifying users;
Date recue / Date received 202 1-1 1-29
Date recue / Date received 202 1-1 1-29
[0032] Integrating predicting result of each pre-set classifier to generate predicting features;
[0033] Using second pre-set model to predict whether the to be predicted user needs loan according to the predicting features.
[0034] The beneficial effects achieved by the present invention are:
[0035] The present invention provides a predicting of user's loan willingness method, comprising: obtaining user data of to be predicted user, the user data comprises user properties and historical consuming behaviors; using a plurality of pre-set classifiers of first pre-set model, generating predicting result of the to be predicted user's category according to the user data respectively, the pre-set classifiers are used for classifying users;
integrating predicting result of each pre-set classifier to generate predicting features; and using second pre-set model to predict whether the to be predicted user needs loan according to the predicting features, predicting user loan willingness according to obtained user data has avoided manual screening process on user data, avoided inaccurate predicting led by inaccurate manual screening, shortened time period required for features experiment, and improved accuracy and efficiency of model predicting.
integrating predicting result of each pre-set classifier to generate predicting features; and using second pre-set model to predict whether the to be predicted user needs loan according to the predicting features, predicting user loan willingness according to obtained user data has avoided manual screening process on user data, avoided inaccurate predicting led by inaccurate manual screening, shortened time period required for features experiment, and improved accuracy and efficiency of model predicting.
[0036] All products of the present invention do not need to have all the above-mentioned effects.
Drawing Description
Drawing Description
[0037] In order to describe the technical solutions clearer in the implementations of the present application or the prior art, the following are drawings that need to be used are briefly introduced. Obviously, the drawings in the following description are only some implementations of the application, for those of ordinary skill in the art, without creative work, they can also obtain other drawings based on these drawings.
Date recue / Date received 202 1-1 1-29
Date recue / Date received 202 1-1 1-29
[0038] Figure 1 is a schematic diagram of the user sample's extraction provided by the implementation of the present invention;
[0039] Figure 2 is an ROC graph based on a GBDT and logistic regression fusion model and a single logistic regression model provided by the implementation of the present invention;
[0040] Figure 3 is a flow chart provided by the implementation of the present invention;
[0041] Figure 4 is a structural diagram of apparatus provided by the implementation of the present invention;
[0042] Figure 5 is a structural diagram of computer system provided by the implementation of the present invention;
[0043] Figure 6 is a model structural diagram based on a GBDT and logistic regression fusion model provided by the present invention.
Specific Implementation Methods
Specific Implementation Methods
[0044] To make clearer invention purpose, technical solution and benefits, the following will describe the technical solutions of the implementations in the present application with accompanying drawings, obviously the described implementations are only a part of the implementations in the present application. Based on the implementations in the present invention, all other implementations obtained by those of ordinary skilled in the art will fall in the protection scope of the present invention.
[0045] As the above-mentioned in the technical background, in order to solve the technical problems, the present application provides a predicting of user's loan willingness method, classifying users with the classifiers of pre-trained first pre-set model according to the obtained Date recue / Date received 202 1-1 1-29 user data. Then integrating the classified result by classifiers, generating predicting features, inputting predicting features into the second pre-set model, the second pre-set model predicts the user's loan willingness according to the inputted predicting features.
[0046] The first predicting model can be gradient boosting random tree model (GBDT), the second pre-set model can be logistic regression model (LR), the training of the above two models comprises:
[0047] Si, obtaining data sample;
[0048] As shown in Figure 1, a pre-set time can be selected, obtaining the users who do not conduct transactions and other activities with pre-set financial platform in three months after the pre-set time. Determining the users who have loan behaviors with pre-set platform within one week after three months as the positive samples, determining the users who have no loan behaviors with pre-set financial platform within one week after three months as the negative samples. For the negative users and the positive users, obtaining correspondingly historical user data including user properties, user financial behaviors and one or a plurality of user's consuming behaviors. Wherein the user properties can include features such as gender, age, and mobile phone number's attribution; user financial behaviors include user's financial membership level on the pre-set platform, credit card and debit card bound to the pre-set platform and other data; user consuming behaviors include the user's purchase frequency, consuming amount, and other data on the pre-set platform within the pre-set time period.
[0049] Pre-processing the obtained historical user data with data cleaning, data conversion, and data standardization, generating the correspondingly basic features variables.
[0050] Preferably, the pre-set platform can be a platform that provides user with financial service such as loan.
[0051] Preferably, 70% of the obtained positive samples and negative samples can be used as Date recue / Date received 202 1-1 1-29 the training dataset, and 30% can be as the test dataset.
[0052] S2, using the training dataset to train the model;
[0053] Preferably, using the basic features variables contained in the positive samples and the negative samples to train GBDT model. The GBDT model generates a weak classifier also called as tree during each iteration process, each weak classifier is trained according to the previous round of iteration of classifier's residuals, after a pre-set number of iterations' training, a pre-set number of weak trained classifiers are generated. Each weak classifier can classify users according to user data.
[0054]
According to the classification result of all weak classifiers, generating correspondingly predicting features. For example, GBDT model generates two weak classifiers Ti, T2, the number of leaf nodes of two weak classifiers are respectively three and two, Li, j represents that the i-th classifier classifies the sample to the j-th leaf node, taking sample X1 as example:
According to the classification result of all weak classifiers, generating correspondingly predicting features. For example, GBDT model generates two weak classifiers Ti, T2, the number of leaf nodes of two weak classifiers are respectively three and two, Li, j represents that the i-th classifier classifies the sample to the j-th leaf node, taking sample X1 as example:
[0055] The sample X1 is classified to the second leaf node on the first classifier, in other words, L1,2, then this sample's vector expression is (0, 1, 0) on Ti;
[0056] The sample X1 is classified to the second leaf node on the second classifier, in other words, L2,1, then this sample's vector expression is (1, 0) on Ti;
[0057] Then the obtained vector expression of sample X1 according to the entire GDBT
model is (0, 1, 0, 1, 0).
model is (0, 1, 0, 1, 0).
[0058] Similarly, the classification result of each sample will be expressed as a 0/1 features vector with length 5.
[0059] Training the logistic regression model according to the features vectors generated by Date recue / Date received 202 1-1 1-29 the positive sample and the negative sample respectively, obtaining the trained logistic regression model.
[0060] Specifically, the input variables of the logistic regression model can be expressed as Xgbdt = tx1, x2, ... , xii}, wherein, Xgbdt refers to the 0/1 features vector corresponding to each sample obtained through the training of GBDT model; the logistic regression model can be expressed by formula: f0 (x) = ¨ 1-F1e-z = 1-Fe-19,x wherein, z = 00 +
01.1C + ... + 071x71 = OT X, 0 is obtained pre-set parameter through the training.
01.1C + ... + 071x71 = OT X, 0 is obtained pre-set parameter through the training.
[0061] As shown in Figure 6, training to obtain the GBDT model and the logistic regression model which can form a fusion model, testing the model by using the test dataset, determining the model's predicting result.
[0062] Specifically, the above-mentioned model can be used to predict the user's loan willingness according to user's user data, the predicting process comprises:
[0063] Step 1, obtaining user data of the target user;
[0064] Preferably, the target user can be the user who does not conduct transactions or other behaviors on the pre-set platform within a pre-set time period.
[0065] The user data includes the target user's user properties, user financial behaviors and one or a plurality of user's consuming behaviors.
[0066] Step2, inputting the user data into the trained GDBT model, each weak classifier of the trained GDBT model predicts the classification of the target user according to the user data.
[0067] 5tep3, integrating the classification result of each weak classifier, generating the correspondingly predicting vector;
Date recue / Date received 202 1-1 1-29
Date recue / Date received 202 1-1 1-29
[0068] Step4, using the trained logistic regression model, predicting whether the target user needs loan according to the predicting vector.
[0069] Figure 2 is an ROC graph (receiver operating's features curve) result of recognizing the pre-set user's loan willingness based on a GBDT and logistic regression fusion model and a single logistic regression model, the test dataset based on the fusion model of the GDBT and the logistic regression has better predicting result than the single logistic regression model.
[0070] Implementation 2
[0071] Corresponding to the above-mentioned implementations, as shown in Figure 3, the present application provides a predicting of user's loan willingness method, the method comprises:
[0072] 310, obtaining user data of to be predicted user, the user data comprises user properties and historical consuming behaviors;
[0073] 320, using a plurality of pre-set classifiers of first pre-set model, generating predicting result of the to be predicted user's category according to the user data respectively, the pre-set classifiers are used for classifying users;
[0074] 330, integrating predicting result of each pre-set classifier to generate predicting features;
[0075] 340, using second pre-set model to predict whether the to be predicted user needs loan according to the predicting features.
[0076] Preferably, the method comprises:
Date recue / Date received 202 1-1 1-29
Date recue / Date received 202 1-1 1-29
[0077] 350, when the to be predicted user is predicted to need loan, determining the to be predicted user as target user, so as to provide loan product to the target user.
[0078] 360, using obtained first training dataset to train first pre-set model, generating pre-set classifier, the first training dataset includes historical user data and whether correspondingly user has loan behavior within a pre-set time period;
[0079] 361, according to residuals of pre-set classifier generated in previous round of iterative training, the first pre-set model is iterative trained to generate correspondingly pre-set classifier, until the first pre-set model meets the correspondingly pre-set conditions, the first pre-set model consists of a plurality of pre-set classifiers.
[0080]
Preferably, integrating predicting result of each pre-set classifier to generate predicting features, comprising:
Preferably, integrating predicting result of each pre-set classifier to generate predicting features, comprising:
[0081] 362, integrating the predicting result according to generation order of the pre-set classifiers.
[0082] Preferably, training process of the second pre-set model comprises:
[0083] 363, according to predicting result of each pre-set classifier of the first pre-set model, generating predicting features corresponding to each historical user data;
[0084] 364, according to predicting features corresponding to the historical user data and whether correspondingly user has loan behavior within a pre-set time period, generating second training dataset;
[0085] 365, using the second training dataset to train the second pre-set model.
[0086] Preferably, the second pre-set model comprises the trained logistic regression model.
Date recue / Date received 202 1-1 1-29
Date recue / Date received 202 1-1 1-29
[0087] Preferably, the first pre-set model comprises the trained gradient boosting random tree model.
[0088] Preferably, the user data also comprises the historical financial behaviors of the to be predicted user.
[0089] Implementation 3
[0090] Corresponding to the above-mentioned implementations, the present application provides one predicting apparatus of user's loan willingness, the apparatus comprises:
[0091] An obtaining module 410 for obtaining user data of to be predicted user, the user data comprises user properties and historical consuming behaviors;
[0092] A process module 420 for using a plurality of pre-set classifiers of first pre-set model, generating predicting result of the to be predicted user's category according to the user data respectively, the pre-set classifiers are used for classifying users;
integrating predicting result of each pre-set classifier to generate predicting features; and using second pre-set model to predict whether the to be predicted user needs loan according to the predicting features.
integrating predicting result of each pre-set classifier to generate predicting features; and using second pre-set model to predict whether the to be predicted user needs loan according to the predicting features.
[0093] Preferably, the apparatus comprises a training module 430 configured to use first training dataset to train first pre-set model and generate pre-set classifier, the first training dataset includes historical user data and whether correspondingly user has loan behavior within a pre-set time period; according to residuals of pre-set classifier generated in previous round of iterative training, the first pre-set model is iterative trained to generate correspondingly pre-set classifier, until the first pre-set model meets the correspondingly pre-set conditions, the first pre-set model consists of a plurality of pre-set classifiers.
[0094] Preferably, the training module is also used for according to predicting result of each Date recue / Date received 202 1-1 1-29 pre-set classifier of the first pre-set model, generating predicting features corresponding to each historical user data; according to predicting features corresponding to the historical user data and whether correspondingly user has loan behavior within a pre-set time period, generating second training dataset; and using the second training dataset to train the second pre-set model.
[0095] Preferably, the second pre-set model comprises the trained logistic regression model.
[0096] Preferably, the first pre-set model comprises the trained gradient boosting random tree model.
[0097] Preferably, the user data also comprises the historical financial behaviors of the to be predicted user.
[0098] Preferably, the process module 420 also can be used for integrating the predicting result according to generation order of the pre-set classifiers.
[0099] Preferably, the process module 420 also can be used for when the to be predicted user is predicted to need loan, determining the to be predicted user as target user, so as to provide loan product to the target user.
[0100] Implementation 4
[0101]
Corresponding the above-mentioned method, apparatus and system, the implementation 4 of the present application provides aa computer system, comprising: one or a plurality of processors; and a memory associated with one or a plurality of processors, the memory is configured to store program commands, if the program commands are executed by one or a plurality of processors, executing following operations:
Corresponding the above-mentioned method, apparatus and system, the implementation 4 of the present application provides aa computer system, comprising: one or a plurality of processors; and a memory associated with one or a plurality of processors, the memory is configured to store program commands, if the program commands are executed by one or a plurality of processors, executing following operations:
[0102] Obtaining user data of to be predicted user, the user data comprises user properties and historical consuming behaviors;
Date recue / Date received 202 1-1 1-29
Date recue / Date received 202 1-1 1-29
[0103] Using a plurality of pre-set classifiers of first pre-set model, generating predicting result of the to be predicted user's category according to the user data respectively, the pre-set classifiers are used for classifying users;
[0104] Integrating predicting result of each pre-set classifier to generate predicting features;
[0105] Using second pre-set model to predict whether the to be predicted user needs loan according to the predicting features.
[0106] Wherein, Figure 5 exemplarily shows the architecture of the computer system , which can specifically include a processor 1510, video display adapter 1511, disk driver 1512, input/output interface 1513, network interface 1514, and memory 1520. The above-mentioned processor 1510, video display adapter 1511, disk driver 1512, input/output interface 1513, network interface 1514 and memory 1520 can be connected through a communication bus 1530.
[0107]
Wherein, the processor 1510 can be achieved by using a general CPU
(Central Processing Unit), Microprocessor, Application Specific Integrated Circuit (ASIC) , or one or more integrated circuits, which are used to execute relative program to achieve the technical solutions provided in this application.
Wherein, the processor 1510 can be achieved by using a general CPU
(Central Processing Unit), Microprocessor, Application Specific Integrated Circuit (ASIC) , or one or more integrated circuits, which are used to execute relative program to achieve the technical solutions provided in this application.
[0108] The memory 1520 can adopt ROM ( Read Only Memory), RAM (Random Access Memory), static storage devices and dynamic storage devices to achieve. The memory 1520 can store operate system 1521 used to control the running of the computer system 1500, used to control the low-level operation of the computer system 1500's Basic Input Output System (BIOS) 1522. In addition, storing a web browser 1523, data storage management 1524, and icon font processing system 1525 and so on. The above-mentioned icon font processing system 1525 can be the specific application that implements the above-mentioned steps. To sum up, when achieving the technical solutions provided by this Date recue / Date received 202 1-1 1-29 application through software or firmware, related program codes are stored in the memory 1520 and executed by a processor 1510. Input / output interface 1513 is used for connecting input / output modules to achieve the information input and output.
Input! output module can be configured in the device as a component ( not shown in the figure ), or it can be connected to the device to provide corresponding functions. Wherein, Input devices can include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices can include monitors, speakers, vibrators, lights and so on.
Input! output module can be configured in the device as a component ( not shown in the figure ), or it can be connected to the device to provide corresponding functions. Wherein, Input devices can include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices can include monitors, speakers, vibrators, lights and so on.
[0109] The network interface 1514 is used to connect a communication module (not shown in the figure) to achieve the communication interaction between this device and other devices. Wherein, the communication module can achieve communication through wired means (such as USB , network cable, etc. ) , or through wireless methods ( such as mobile network, WIFI, Bluetooth, etc. ) to achieve communication.
[0110] The bus 1530 includes a path and transmits information among various components of the device (such as the processor 1510 , the video display adapter 1511, the disk driver 1512, the input/output interface 1513, the network interface 1514, and the memory 1520).
[0111] In addition, the computer system 1500 also can obtain information with specific receiving conditions from the virtual resource object's receiving condition information database 1541 for condition judgement and son on.
[0112] What should be noted is that although the above device only shows the processor 1510, the video display adapter 1511, the disk driver 1512, input/output interface 1513, network interface 1514, memory 1520, bus 1530, etc., but in the process of the specific implementation, the device may also include other essential components for normal operation. In addition, those skilled in the art can understand that the above apparatus can comprise only the essential components of the present application to achieve the implementation, but there is no need to contain all the components as shown in figure.
Date recue / Date received 202 1-1 1-29
Date recue / Date received 202 1-1 1-29
[0113] According to the description of the above implementations that those skilled in the art can clearly understand that the application can be achieved with the help of software and essential general hardware platform. Based on this understanding, the essence of the technical solution of this application, or in other words, the part that contributes to the existing technology can be implemented in the form of a software product, the computer software product can be stored in storage media, such as ROM/RAM, magnetic disks, optical disks, etc., including several commands to make a computer device (can be a personal computer, a cloud server, or a network device, etc.) execute the methods described in each implementation or some of the implementations of the present application.
[0114] The various implementations in this description are described in a progressive manner, the same and similar parts among the various implementations can be referred to each other separately, and each implementation focuses on the differences compared with the other implementations. Especially for the concern of the system or the system implementations, since it is basically similar to the implementation method, the description is relatively simple. For related details, please refer to the implementation method. The system and system implementations described in the above are only schematic, and the units described by separate parts may or may not be physically separated, and the parts displayed as units may or may not be physical units, which means, it can be in one place, or it may be distributed to a plurality of network units. Some or all of the modules are selected according to actual needs to achieve the purpose of implementation's solution. The ordinary skill in the art can understand and implement without creative work.
[0115] The above-mentioned descriptions are only preferred implementations of the present invention and are not intended to limit the present invention, anything within the spirit and principle of the present invention, any modifications, equivalent replacements, improvements, etc., which shall be included in the protection scope of the present invention.
Date recue / Date received 202 1-1 1-29
Date recue / Date received 202 1-1 1-29
Claims (10)
1. A predicting method of user's loan willingness comprises:
obtaining user data of to be predicted user, the user data comprises user properties and historical consuming behaviors;
using a plurality of pre-set classifiers of first pre-set model, generating predicting result of the to be predicted user's category according to the user data respectively, the pre-set classifiers are used for classifying users;
integrating predicting result of each pre-set classifier to generate predicting features;
and using second pre-set model to predict whether the to be predicted user needs loan according to the predicting features.
obtaining user data of to be predicted user, the user data comprises user properties and historical consuming behaviors;
using a plurality of pre-set classifiers of first pre-set model, generating predicting result of the to be predicted user's category according to the user data respectively, the pre-set classifiers are used for classifying users;
integrating predicting result of each pre-set classifier to generate predicting features;
and using second pre-set model to predict whether the to be predicted user needs loan according to the predicting features.
2. The method according to claim 1, wherein, the first pre-set model is obtained through iterative training, the iterative training process of the first pre-set model comprises:
using obtained first training dataset to train first pre-set model, generating pre-set classifier, the first training dataset includes historical user data and whether correspondingly user has loan behavior within a pre-set time period; and according to residuals of pre-set classifier generated in previous round of iterative training, the first pre-set model is iterative trained to generate correspondingly pre-set classifier, until the first pre-set model meets the correspondingly pre-set conditions, the first pre-set model consists of a plurality of pre-set classifiers.
using obtained first training dataset to train first pre-set model, generating pre-set classifier, the first training dataset includes historical user data and whether correspondingly user has loan behavior within a pre-set time period; and according to residuals of pre-set classifier generated in previous round of iterative training, the first pre-set model is iterative trained to generate correspondingly pre-set classifier, until the first pre-set model meets the correspondingly pre-set conditions, the first pre-set model consists of a plurality of pre-set classifiers.
3. The method according to claim 1 or 2, wherein, the first pre-set model comprises trained gradient boosting random tree model, the second pre-set model comprises trained logistic Date recue / Date received 2021-11-29 regression model.
4. The method according to claim 2, wherein, training process of the second pre-set model comprises:
according to predicting result of each pre-set classifier of the first pre-set model, generating predicting features corresponding to each historical user data;
according to predicting features corresponding to the historical user data and whether correspondingly user has loan behavior within a pre-set time period, generating second training dataset; and using the second training dataset to train the second pre-set model.
according to predicting result of each pre-set classifier of the first pre-set model, generating predicting features corresponding to each historical user data;
according to predicting features corresponding to the historical user data and whether correspondingly user has loan behavior within a pre-set time period, generating second training dataset; and using the second training dataset to train the second pre-set model.
5. The method according to claim 2, wherein, integrating predicting result of each pre-set classifier to generate predicting features, comprising:
integrating the predicting result according to generation order of the pre-set classifiers.
integrating the predicting result according to generation order of the pre-set classifiers.
6. The method according to claim 2 comprises:
when the to be predicted user is predicted to need loan, determining the to be predicted user as target user, so as to provide loan product to the target user.
when the to be predicted user is predicted to need loan, determining the to be predicted user as target user, so as to provide loan product to the target user.
7. A predicting apparatus of user's loan willingness, wherein, the apparatus comprises:
an obtaining module for obtaining user data of to be predicted user, the user data comprises user properties and historical consuming behaviors; and a process module for using a plurality of pre-set classifiers of first pre-set model, generating predicting result of the to be predicted user's category according to the user Date recue / Date received 2021-11-29 data respectively, the pre-set classifiers are used for classifying users;
integrating predicting result of each pre-set classifier to generate predicting features;
and using second pre-set model to predict whether the to be predicted user needs loan according to the predicting features.
an obtaining module for obtaining user data of to be predicted user, the user data comprises user properties and historical consuming behaviors; and a process module for using a plurality of pre-set classifiers of first pre-set model, generating predicting result of the to be predicted user's category according to the user Date recue / Date received 2021-11-29 data respectively, the pre-set classifiers are used for classifying users;
integrating predicting result of each pre-set classifier to generate predicting features;
and using second pre-set model to predict whether the to be predicted user needs loan according to the predicting features.
8. The predicting apparatus according to claim 7, wherein, the apparatus comprises a training module configured to use first training dataset to train first pre-set model and generate pre-set classifier, the first training dataset includes historical user data and whether correspondingly user has loan behavior within a pre-set time period; according to residuals of pre-set classifier generated in previous round of iterative training, the first pre-set model is iterative trained to generate correspondingly pre-set classifier, until the first pre-set model meets the correspondingly pre-set conditions, the first pre-set model consists of a plurality of pre-set classifiers.
9. The predicting apparatus according to claim 8, wherein, the training module is also used for according to predicting result of each pre-set classifier of the first pre-set model, generating predicting features corresponding to each historical user data; according to predicting features corresponding to the historical user data and whether correspondingly user has loan behavior within a pre-set time period, generating second training dataset; and using the second training dataset to train the second pre-set model.
10. A computer system comprises:
one or a plurality of processors; and a memory associated with one or a plurality of processors, the memory is configured to store program commands, if the program commands are executed by one or a plurality of processors, executing following operations:
Date recue / Date received 2021-11-29 obtaining user data of to be predicted user, the user data comprises user properties and historical consuming behaviors;
using a plurality of pre-set classifiers of first pre-set model, generating predicting result of the to be predicted user's category according to the user data respectively, the pre-set classifiers are used for classifying users;
integrating predicting result of each pre-set classifier to generate predicting features; and using second pre-set model to predict whether the to be predicted user needs loan according to the predicting features.
Date recue / Date received 2021-11-29
one or a plurality of processors; and a memory associated with one or a plurality of processors, the memory is configured to store program commands, if the program commands are executed by one or a plurality of processors, executing following operations:
Date recue / Date received 2021-11-29 obtaining user data of to be predicted user, the user data comprises user properties and historical consuming behaviors;
using a plurality of pre-set classifiers of first pre-set model, generating predicting result of the to be predicted user's category according to the user data respectively, the pre-set classifiers are used for classifying users;
integrating predicting result of each pre-set classifier to generate predicting features; and using second pre-set model to predict whether the to be predicted user needs loan according to the predicting features.
Date recue / Date received 2021-11-29
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011043449.8 | 2020-09-28 | ||
| CN202011043449.8A CN112328869A (en) | 2020-09-28 | 2020-09-28 | User loan willingness prediction method and device and computer system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA3135466A1 true CA3135466A1 (en) | 2022-03-28 |
Family
ID=74303290
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA3135466A Pending CA3135466A1 (en) | 2020-09-28 | 2021-09-28 | User loan willingness prediction method and device and computer system |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN112328869A (en) |
| CA (1) | CA3135466A1 (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116071103A (en) * | 2023-03-07 | 2023-05-05 | 天津金城银行股份有限公司 | Method and device for prompting client to borrow and electronic equipment |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112785095A (en) * | 2021-03-12 | 2021-05-11 | 北京顶象技术有限公司 | Loan prediction method, loan prediction device, electronic device, and computer-readable storage medium |
| CN113065946A (en) * | 2021-03-17 | 2021-07-02 | 上海浦东发展银行股份有限公司 | Classification updating promoting method and device for overdue credit card certificate clients and storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107895277A (en) * | 2017-09-30 | 2018-04-10 | 平安科技（深圳）有限公司 | Method, electronic installation and the medium of push loan advertisement in the application |
| CN110971460A (en) * | 2019-12-03 | 2020-04-07 | 北京红山信息科技研究院有限公司 | Off-network prediction method, device, server and storage medium |
-
2020
- 2020-09-28 CN CN202011043449.8A patent/CN112328869A/en active Pending
-
2021
- 2021-09-28 CA CA3135466A patent/CA3135466A1/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116071103A (en) * | 2023-03-07 | 2023-05-05 | 天津金城银行股份有限公司 | Method and device for prompting client to borrow and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112328869A (en) | 2021-02-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10360631B1 (en) | Utilizing artificial intelligence to make a prediction about an entity based on user sentiment and transaction history | |
| CN110995459B (en) | Abnormal object identification method, device, medium and electronic equipment | |
| CA3135466A1 (en) | User loan willingness prediction method and device and computer system | |
| CN111523677B (en) | Method and device for realizing interpretation of prediction result of machine learning model | |
| CN111800289B (en) | Communication network fault analysis method and device | |
| CN111127178A (en) | Data processing method and device, storage medium and electronic equipment | |
| CN111783039B (en) | Risk determination method, risk determination device, computer system and storage medium | |
| CN111738331A (en) | User classification method and device, computer-readable storage medium and electronic device | |
| CN106537423A (en) | Adaptive Characterization as a Service | |
| CN111340240A (en) | Method and device for realizing automatic machine learning | |
| CN111179051A (en) | Financial target customer determination method and device and electronic equipment | |
| CN111210332A (en) | Method and device for generating post-loan management strategy and electronic equipment | |
| US11099107B2 (en) | Component testing plan considering distinguishable and undistinguishable components | |
| US20230029218A1 (en) | Feature engineering using interactive learning between structured and unstructured data | |
| CN113610625A (en) | Overdue risk warning method and device and electronic equipment | |
| CN113139381A (en) | Unbalanced sample classification method and device, electronic equipment and storage medium | |
| CN111026973B (en) | Commodity interest degree prediction method and device and electronic equipment | |
| CN112419025A (en) | User data processing method and device, storage medium and electronic equipment | |
| US12106617B2 (en) | Method and system for auto generating automotive data quality marker | |
| CN115511187A (en) | Asset recovery prediction method, device, equipment, medium and computer program product | |
| CN113516398A (en) | Risk equipment identification method and device based on hierarchical sampling and electronic equipment | |
| CN114020916A (en) | Text classification method and device, storage medium and electronic equipment | |
| CN113297479A (en) | User portrait generation method and device and electronic equipment | |
| US20230351211A1 (en) | Scoring correlated independent variables for elimination from a dataset | |
| US20240144144A1 (en) | Machine-generated process transformation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| EEER | Examination request |
Effective date: 20220726 |
|
| EEER | Examination request |
Effective date: 20220726 |
|
| EEER | Examination request |
Effective date: 20220726 |
|
| EEER | Examination request |
Effective date: 20220726 |
|
| EEER | Examination request |
Effective date: 20220726 |
|
| EEER | Examination request |
Effective date: 20220726 |
|
| EEER | Examination request |
Effective date: 20220726 |
|
| EEER | Examination request |
Effective date: 20220726 |
|
| EEER | Examination request |
Effective date: 20220726 |
|
| EEER | Examination request |
Effective date: 20220726 |
|
| EEER | Examination request |
Effective date: 20220726 |