CN107424007A - A kind of method and apparatus for building electronic ticket susceptibility identification model - Google Patents
A kind of method and apparatus for building electronic ticket susceptibility identification model Download PDFInfo
- Publication number
- CN107424007A CN107424007A CN201710565218.5A CN201710565218A CN107424007A CN 107424007 A CN107424007 A CN 107424007A CN 201710565218 A CN201710565218 A CN 201710565218A CN 107424007 A CN107424007 A CN 107424007A
- Authority
- CN
- China
- Prior art keywords
- electronic ticket
- user
- sample data
- identification model
- susceptibility
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0207—Discounts or incentives, e.g. coupons or rebates
- G06Q30/0211—Determining the effectiveness of discounts or incentives
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/285—Clustering or classification
-
- 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0207—Discounts or incentives, e.g. coupons or rebates
-
- 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0207—Discounts or incentives, e.g. coupons or rebates
- G06Q30/0224—Discounts or incentives, e.g. coupons or rebates based on user history
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses the method and apparatus of structure electronic ticket susceptibility identification model, it is related to field of computer technology.One embodiment of this method includes：User's sample data is obtained according to the application scenarios of electronic ticket；Electronic ticket is sorted out to obtain electronic ticket species according at least one of the denomination of electronic ticket, limit or discount rate；Behavioural characteristic is determined according to the application scenarios of electronic ticket；Electronic ticket susceptibility identification model is trained according to user's sample data, electronic ticket species and behavioural characteristic.The embodiment can establish electronic ticket susceptibility identification model exactly, more to accurately judge that susceptibility of the user to electronic ticket, so as to reduce businessman's cost and improve the using effect of electronic ticket.
Description
Technical field
The present invention relates to field of computer technology, more particularly to a kind of method for building electronic ticket susceptibility identification model and
Device.
Background technology
In the marketing, it is often necessary to the consumer psychology of user is stimulated by various promotion methods, for example buys one and gives
First, completely subtract, discount, land vertically, purchase by group, providing the various promotion methods such as electronic ticket.In electric business field, electronic ticket is a kind of common
Marketing methods, by the granting of electronic ticket, the liveness of user can be lifted, maintaining healthy, firm user group can be with
Promote the sale of specific category commodity.Although economic benefit can be brought by providing electronic ticket, certain cost is also undertaken simultaneously,
In order to cost-effective, it is necessary to accomplish the accurate dispensing of electronic ticket, can so create cost-effective to greatest extent while
Earned value.
In process of the present invention is realized, inventor has found that not a kind of effectively determination electronic ticket launches bar in the prior art
The method of part, the dispensing of electronic ticket is mostly just carried out using the spending limit of user or consuming frequency as condition, it is difficult to meet
The demand accurately launched, is unfavorable for the cost control of businessman, and is difficult to produce a desired effect.
The content of the invention
In view of this, the embodiment of the present invention provides a kind of method and apparatus for building electronic ticket susceptibility identification model, energy
It is enough to establish electronic ticket susceptibility identification model exactly, more to accurately judge that susceptibility of the user to electronic ticket, so as to drop
Low businessman's cost simultaneously improves the using effect of electronic ticket.
To achieve the above object, one side according to embodiments of the present invention, there is provided one kind structure electronic ticket susceptibility
The method of identification model, including：
User's sample data is obtained according to the application scenarios of electronic ticket；
Electronic ticket is sorted out according to the electronic ticket information of electronic ticket to obtain electronic ticket species；
Behavioural characteristic is determined according to the application scenarios of electronic ticket；
Electronic ticket susceptibility identification model is trained according to user's sample data, electronic ticket species and behavioural characteristic.
Optionally, the identification of electronic ticket susceptibility is trained according to user's sample data, electronic ticket species and behavioural characteristic
The step of model, includes：
Training set is selected from user's sample data；
According to the training set, electronic ticket species and behavioural characteristic, using there is electronics described in the classification algorithm training of supervision
Certificate susceptibility identification model.
Optionally, methods described also includes：
It is electric described in the classification algorithm training of supervision using having according to the training set, electronic ticket species and behavioural characteristic
After the step of sub- certificate susceptibility identification model,
Checking collection is selected from user's sample data in addition to the training set, uses the electronic ticket susceptibility
Identification model is predicted to the checking collection, judges whether prediction result reaches expected requirement；
If reaching, the electronic ticket susceptibility identification model is preserved；If not up to, selecting new behavioural characteristic, with
And the electronic ticket susceptibility identification model according to the new behavioural characteristic re -training.
Optionally, the sorting algorithm for having a supervision include K- nearest neighbor algorithms, decision Tree algorithms, NB Algorithm,
At least one of Logistic regression algorithms, SVM algorithm or AdaBoost Meta algorithms.
Optionally, the sorting algorithm for having supervision is Gradient Iteration decision Tree algorithms.
Optionally, methods described also includes：
After the step of obtaining user's sample data according to the application scenarios of electronic ticket,
Service condition according to user in user's sample data to electronic ticket, is added in user's sample data
The electronic ticket of user uses mark.
Optionally, the electronic ticket information includes at least one of denomination, limit or discount rate.
To achieve the above object, other side according to embodiments of the present invention, there is provided one kind structure electronic ticket is sensitive
The device of identification model is spent, including：
Sample chooses module, for obtaining user's sample data according to the application scenarios of electronic ticket；
Electronic ticket classifying module, electronic ticket is sorted out to obtain electronic ticket for the electronic ticket information according to electronic ticket
Species；
Characteristic determination module, behavioural characteristic is determined for the application scenarios according to electronic ticket；
Training module, electronic ticket susceptibility is trained to know according to user's sample data, electronic ticket species and behavioural characteristic
Other model.
Optionally, the training module is additionally operable to：
Training set is selected from user's sample data；
According to the training set, electronic ticket species and behavioural characteristic, using there is electronics described in the classification algorithm training of supervision
Certificate susceptibility identification model.
Optionally, described device also includes：
Recruitment evaluation module, for selecting checking collection from user's sample data in addition to the training set, make
The checking collection is predicted with the electronic ticket susceptibility identification model, judges whether prediction result reaches expected requirement；
If reaching, the electronic ticket susceptibility identification model is preserved；If new behavioural characteristic not up to, is selected, and according to institute
State electronic ticket susceptibility identification model described in new behavioural characteristic re -training.
Optionally, the sorting algorithm for having a supervision include K- nearest neighbor algorithms, decision Tree algorithms, NB Algorithm,
At least one of Logistic regression algorithms, SVM algorithm or AdaBoost Meta algorithms.
Optionally, the sorting algorithm for having supervision is Gradient Iteration decision Tree algorithms.
Optionally, described device also includes：
Certificate uses mark module, for the service condition according to user in user's sample data to electronic ticket, in institute
The electronic ticket for stating addition user in user's sample data uses mark.
Optionally, the electronic ticket information includes at least one of denomination, limit or discount rate.
To achieve the above object, another aspect according to embodiments of the present invention, there is provided one kind structure electronic ticket is sensitive
The electronic equipment of identification model is spent, including：
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are by one or more of computing devices so that one or more of processing
Device is at least realized：
User's sample data is obtained according to the application scenarios of electronic ticket；
Electronic ticket is sorted out according to the electronic ticket information of electronic ticket to obtain electronic ticket species；
Behavioural characteristic is determined according to the application scenarios of electronic ticket；
Electronic ticket susceptibility identification model is trained according to user's sample data, electronic ticket species and behavioural characteristic.
To achieve the above object, another aspect according to embodiments of the present invention, there is provided a kind of computer-readable medium, its
On be stored with computer program, at least realized when described program is executed by processor：
User's sample data is obtained according to the application scenarios of electronic ticket；
Electronic ticket is sorted out according to the electronic ticket information of electronic ticket to obtain electronic ticket species；
Behavioural characteristic is determined according to the application scenarios of electronic ticket；
Electronic ticket susceptibility identification model is trained according to user's sample data, electronic ticket species and behavioural characteristic.
One embodiment in foregoing invention has the following advantages that or beneficial effect：Because using the application according to electronic ticket
Scene selects user's sample data and determines behavioural characteristic, and electronic ticket is classified, and then carries out the skill of model training
Art means, the model of user in application-specific scene for electronic ticket susceptibility is established, so as to more accurately judge that user
To the susceptibility of electronic ticket, and then reduce businessman's cost and improve the technique effect of electronic ticket using effect.
Further effect adds hereinafter in conjunction with embodiment possessed by above-mentioned non-usual optional mode
With explanation.
Brief description of the drawings
Accompanying drawing is used to more fully understand the present invention, does not form inappropriate limitation of the present invention.Wherein：
Fig. 1 is the signal of the key step of the method for structure electronic ticket susceptibility identification model according to embodiments of the present invention
Figure；
Fig. 2 is the signal of the main modular of the device of structure electronic ticket susceptibility identification model according to embodiments of the present invention
Figure；
Fig. 3 is that the embodiment of the present invention can apply to exemplary system architecture figure therein；
Fig. 4 is adapted for the structural representation for realizing the terminal device of the embodiment of the present invention or the computer system of server
Figure.
Embodiment
The one exemplary embodiment of the present invention is explained below in conjunction with accompanying drawing, including the various of the embodiment of the present invention
Details should think them only exemplary to help understanding.Therefore, those of ordinary skill in the art should recognize
Arrive, various changes and modifications can be made to the embodiments described herein, without departing from scope and spirit of the present invention.Together
Sample, for clarity and conciseness, the description to known function and structure is eliminated in following description.
Inventor has found during the present invention is realized：User on network when doing shopping, according to the difference of purchasing habits
Various Shopping Behaviors features can be produced, for example, some users just pay special attention to electronic ticket, often go to get electronic ticket, or
When system of users provides electronic ticket, user can just utilize electronic ticket to produce buying behavior, show such user to electronics
Certificate is more sensitive；And some users are not bought because whether providing electronic ticket, show such user to electronic ticket not
It is sensitive.User is divided into different susceptibility colonies based on these behavior expression cans, this operation for website and
The marketing of user all plays vital effect.The embodiment provides for different electronic ticket application scenarios,
The method and device of behavioural characteristic structure electronic ticket susceptibility identification model based on user, the electronic ticket susceptibility identification model
Using the behavioural characteristic of user as input, using user for different type electronic ticket susceptibility in other words use tendency as defeated
Go out, electronic ticket is predicted using probability.
Fig. 1 is the signal of the key step of the method for structure electronic ticket susceptibility identification model according to embodiments of the present invention
Figure.
A kind of as shown in figure 1, side of the structure electronic ticket susceptibility identification model provided according to one embodiment of the invention
Method, including：
S10, user's sample data is obtained according to the application scenarios of electronic ticket.
S11, electronic ticket is sorted out according to the electronic ticket information of electronic ticket to obtain electronic ticket species.
S12, behavioural characteristic is determined according to the application scenarios of electronic ticket.
S13, electronic ticket susceptibility identification mould is trained according to user's sample data, electronic ticket species and behavioural characteristic
Type.
In step S10, the different application scene that is directed to when being launched according to electronic ticket, the present embodiment chooses different user's samples
Notebook data carries out model training.Such as drawing new scene, it is therefore an objective to attract new user to use electronic ticket purchase commodity, it is necessary to
The new user in history section time range is chosen, mainly according to the user's user's sample of single time as model training under first
Data；The scene and for example purchased again for guiding, it is therefore an objective to attract the user for having produced consumption to reuse electronic ticket purchase commodity,
Just need to choose the user for being produced in history section time range and purchasing behavior again, model training is used as using the behavior of these users
User's sample data；For another example it is sunk into sleep user's scene for wake-up, it is therefore an objective to attract the user not consumed for a long time
Buy commodity, therefore the definition for the user that need to determine to be sunk into sleep, for example place an order first away from modern more than 6 months, and nearly 3 months without effective
Order but there is valid order etc. in nearly 6 months, thus need to choose the use of being sunk into sleep that the history section time also corresponds to above-mentioned standard
Family, user's sample data using the behavior of these users as model training.
In step S11, electronic ticket has been carried out in detail according to the electronic ticket such as the denomination of electronic ticket, limit and discount rate information
Classification.Electronic ticket has a limit and denomination mostly, such as the electronic ticket of " full 100 yuan subtract 5 yuan ", and 100 be limit, and 5 be denomination, by
It is various in electronic ticket species, in order to ensure the accuracy rate and reusability of model, electronic ticket need to be sorted out, i.e., by limit and
Denomination is close or identical electronic ticket is classified as one kind, for example the such two kinds of electronic ticket cans of 99-5 and 100-5 are classified as one kind.For
Be easy to sort out, can also according to limit and denomination arithmetical discount rate, such as before " full 100 yuan subtract 5 yuan " electronic ticket folding
Button rate is 5%, is then sorted out according to electronic ticket limit and discount rate, the method and threshold value of classification visually specifically apply and
It is fixed.Some electronic tickets and limit is not provided with addition, also having, can carried out unconditional preferential such as " subtract 5 yuan " or " preferential
5% " etc., when needing to establish sensitivity model to such electronic ticket, then limit and denomination can not be added classification
Scope, and individually investigate discount rate.
In step S12, for the user group of different application scene, its behavioural characteristic is also what is differed, therefore
Need according to application scenarios, select it is representative, clearly distinguished as principal character by user with certificate and without certificate
Behavioural characteristic carry out model training, so that obtained susceptibility identification model can be come in advance according to these behavioural characteristics of user
Survey the probability that user uses electronic ticket.For example, selectable behavioural characteristic is as shown in table 1：
The behavioural characteristic table of table 1
From the above it can be seen that the method for the structure electronic ticket susceptibility identification model that the present embodiment provides, because
User's sample data and determination behavioural characteristic are selected using according to the application scenarios of electronic ticket, and electronic ticket is classified,
Then the technological means of model training is carried out, establishes the model of user in application-specific scene for electronic ticket susceptibility, from
And susceptibility of the user to electronic ticket is more accurately judged that, and then reduction businessman cost and improve electronic ticket and use effect
The technique effect of fruit.
In some optional embodiments, S13, instructed according to user's sample data, electronic ticket species and behavioural characteristic
The step of practicing electronic ticket susceptibility identification model includes：
Training set is selected from user's sample data.In order to guarantee fairness, the selection of training set can use
Randomly selected mode, such as 70% user's sample data is randomly chosen as training set, remaining user's sample data
It can be used for the checking of subsequent step.
According to the training set, electronic ticket species and behavioural characteristic, using there is electronics described in the classification algorithm training of supervision
Certificate susceptibility identification model.There is the sorting algorithm of supervision, refer to there is training sample, and the sorting algorithm with attribute tags, this
Training sample can be divided into different classifications by class algorithm according to attribute tags.By taking the embodiment of the present invention as an example, with user's sample
Notebook data, using behavioural characteristic as attribute tags, model training is carried out for different types of electronic ticket as training sample.
In some optional embodiments, the sorting algorithm for having supervision includes K- nearest neighbor algorithms, decision Tree algorithms, Piao
At least one of plain bayesian algorithm, Logistic regression algorithms, SVM algorithm or AdaBoost Meta algorithms.It is preferable at some
Embodiment in, the sorting algorithm for having supervision is Gradient Iteration decision Tree algorithms.
Gradient Iteration decision Tree algorithms GBDT (Gradient Boosting Decision Tree) is a kind of determining for iteration
Plan tree algorithm, the algorithm may be used with nearly all regression problems, and applicable surface is very wide, simultaneously can be used for two classification problems,
Given threshold, it is positive example more than threshold value, otherwise is negative example.The algorithm can be very good the differentiation for solving electronic ticket susceptibility, can
Its susceptibility score value to electronic ticket is exported for each user, and then supports accurate, fine integral method, and then guides user's purchase
Thing, promote merchandise sales.
GBDT is made up of more decision trees, and the conclusion of all trees, which adds up, does final result.GBDT disassembled is GB, DT
And step-length, GB are Gradient Iterations, its thought is that to establish model each time be under the gradient for the model loss function established before
Direction drops.The foundation of each new model is in order that the residual error of model is reduced to gradient direction before obtaining.DT is regression tree, mesh
Mark is exactly to minimize loss function to minimize mean square deviation, and such functional vector means that into previous regression tree in sample space
On predicted value, then ask functional vector gradient to be equal to desired value and subtract predicted value, i.e. residual vector, therefore next recurrence
Tree is exactly to be fitted this residual vector.The main thought of step-length is the effect for walking the gradual Approaching Results of a small step every time than every
The mode of Approaching Results is easier to avoid over-fitting secondary major step advanced in years quickly.
In some optional embodiments, methods described also includes：
In S10, after the step of obtaining user's sample data according to the application scenarios of electronic ticket,
Service condition according to user in user's sample data to electronic ticket, is added in user's sample data
The electronic ticket of user uses mark.
There is the sorting algorithm of supervision in use, such as during GBDT algorithms progress model training, markd need to learn
Sample, therefore the present embodiment judges each user in user's sample data true according to application scenarios according to user's sample data
Whether the electronic ticket got in fixed time range used, and used mark according to using result addition electronic ticket；For example,
Mark can be used to be arranged to 1 electronic ticket if user has used certain electronic ticket, if not being arranged to 0 if.
In some optional embodiments, it is ready to after the behavioural characteristic of user, it is necessary first to turn behavioural characteristic collection
Change sparse matrix format, such as libsvm forms into；Libsvm data format is as follows：Lable 1:value 2:
Value ..., wherein lable are that the mark i.e. electronic ticket of classification uses mark, and value is the behavioural characteristic that participate in training, is counted
Separated between with space, if certain characteristic value is 0, this feature can be omitted, and so do the use that can reduce internal memory, carry
High arithmetic speed.For example, table 2 is the user behavior examples of features after chosen.
The behavioural characteristic example of table 2
It is as follows that the behavioural characteristic of table 2 is converted into libsvm forms：
In some optional embodiments, when training electronic ticket susceptibility identification model, using Distributed Calculation engine.
Such as Spark engines, the Mahout engines based on Hadoop or distributed XGBoost engines etc. can be selected.
In some optional embodiments, methods described also includes：
It is electric described in the classification algorithm training of supervision using having according to the training set, electronic ticket species and behavioural characteristic
After the step of sub- certificate susceptibility identification model,
Checking collection is selected from user's sample data in addition to the training set, uses the electronic ticket susceptibility
Identification model is predicted to the checking collection, judges whether prediction result reaches expected requirement.If reaching, the electricity is preserved
Sub- certificate susceptibility identification model；If not up to, new behavioural characteristic is selected, and instruct again according to the new behavioural characteristic
Practice the electronic ticket susceptibility identification model.
In order to verify whether the model that training obtains meets the requirements, user's sample data can be randomly divided into two
Point, choose a portion and collect as training set, another part as checking；Such as can be by user characteristic data with 7:3
Ratio random division is that training set and checking collect.After model training is carried out using training set, further by verifying that set pair is instructed
It is experienced to model verified, judged whether to need the condition (mainly behavioural characteristic) to model training according to the result
Optimize improvement.
In some optional embodiments, tri- accurate rate (Precision), recall rate (Recall) and AUC fingers are used
Mark comes whether assessment models reach expected requirement.Specific calculating logic is described below：4 basic definitions clear and definite first, TP：Refer to
Model prediction is just (1), and is actually also the quantity of the just object of observation of (1)；TN：It is negative (0) to refer to model prediction,
And the quantity of the object of observation for really bearing (0) actually also；FP：Refer to model prediction as just (1), but be actually negative
(0) quantity of the object of observation；FN：It is negative (0) to refer to model prediction, but is actually the quantity of the just object of observation of (1).
So, accurate rate, which refers to model and is correctly identified as the object of just (1), accounts for ratio of the Model Identification for the object of observation sum of just (1),
Formula is as follows：Accurate rate=TP/ (TP+FP).Recall rate reflects the proportion that the positive example being appropriately determined accounts for total positive example, formula
It is as follows：Recall rate=TP/ (TP+FN).ROC curve (receiver operating characteristic, subject's work
Feature) and AUC (Area Under roc Curve, roc TG-AUC) be also usually utilized to evaluate a two-value grader
Quality, AUC is the area under ROC curve, and the numerical value of this area is not more than 1, and ROC curve be normally at y=x this
The top of bar straight line, so between AUC spans 0.5 to 1, in general, AUC is bigger, and classifying quality is better.Actually make
, can be by setting whether the mode such as threshold value judgment models reach expected requirement in.
In order that the method for the embodiment of the present invention is clearer, the present invention is implemented below by a specific embodiment
The method of example illustrates.
In this specific embodiment, except training obtain susceptibility identification model in addition to, also further using the model to
Family behavior is predicted that the present embodiment mainly includes the following steps that：
Crowd chooses
Corresponding crowd's (user's sample data i.e. in previous embodiment) is selected according to the application scenarios of reality, for example is drawn
New scene and wake-up are sunk into sleep user's scene, for drawing new scene, just choose the new user i.e. user of nearly 1 year of certain category at this
Under category first under single time in nearly 1 year；It is sunk into sleep user's scene for waking up, choosing nearly half a year under certain category does not all have
Single time is spent under last time of the user to place an order the i.e. user under the category before half a year.
The classification of electronic ticket
The electronic ticket of history granting may have many kinds, as shown in table 3, electricity be calculated according to the limit of electronic ticket and denomination
The discount rate of sub- certificate, sorts out further according to limit and discount rate to electronic ticket, first the present invention by limit within 100 yuan
Electronic ticket is divided into four gears：25th, 50,75,100, i.e., electronic ticket of the limit between 37.5 and 62.5 is classified as 50, with such
Push away.Discount rate is one section according to 5 percentage points by the present embodiment, i.e., 5%, 10%, 15%...... thus enters electronic ticket
Classification is gone, classifying method can customize.
The electronic ticket classification chart of the history granting of table 3
| Electronic ticket type after classification | Limit | Denomination | Discount rate |
| C:25-25% | 20 | 5 | 25.0% |
| C:25-50% | 20 | 10 | 50.0% |
| C:25-45% | 22 | 10 | 45.5% |
| C:25-45% | 23 | 10 | 43.5% |
| C:25-40% | 25 | 10 | 40.0% |
| C:25-15% | 29 | 5 | 17.2% |
| C:50-10% | 45 | 5 | 11.1% |
| C:50-10% | 49 | 5 | 10.2% |
| C:50-30% | 49 | 15 | 30.6% |
| C:50-40% | 49 | 20 | 40.8% |
| C:50-10% | 50 | 5 | 10.0% |
| C:50-20% | 50 | 10 | 20.0% |
| C:50-50% | 50 | 25 | 50.0% |
| C:75-15% | 69 | 10 | 14.5% |
| C:75-45% | 69 | 30 | 43.5% |
| C:75-10% | 79 | 8 | 10.1% |
| C:75-20% | 79 | 15 | 19.0% |
| C:100-10% | 99 | 10 | 10.1% |
| C:100-10% | 99 | 9 | 9.1% |
Electronic ticket uses mark
Use feelings of this certain customers of first step selection on the electronic ticket of nearly 1 year are seen according to different scenes first
Condition, for example new scene is drawn, it is necessary to see electronic ticket service condition of the new user of category of nearly 1 year of selection before placing an order first, if
Placed an order when newly user is single under first with certificate and be then labeled as 1, be otherwise 0；For waking up user's scene of being sunk into sleep, it is possible to see this
Certain customers before active period is half a year to the service condition of electronic ticket, if electronic ticket that user gets before half a year and making
With being then labeled as 1, get and be not used then labeled as 0.
Behavioural characteristic is processed
Needed to choose different behavioural characteristics according to different application scenarios, for example draw new scene, due to being that category is newly used
Family is, it is necessary to observe the behavioural characteristic before the new user's first purchase in this part, therefore be the purchase feature under no category, therefore
Need to be differentiated according to behavioural characteristic before the purchase such as the browsing of user plus purchase, concern, search (specific features refer to table 1,
It will not be repeated here)；For waking up user's scene of being sunk into sleep, it is possible to special using class according to the purchase category feature of user, electronic ticket
Sign is differentiated.
Model training and recruitment evaluation
The feature processed is processed into libsvm form first, by training set with 7:3 ratio splits into instruction at random
Practice collection and checking collection, then carry out model training with the GBDT algorithms based on Spark, such as iterations is set 30 times, tree
Depth is 8, calls train functions to carry out model training, and then calls predict functions to be predicted on checking collection, will be pre-
The lable values of the result measured and reality are contrasted, and then are calculated AUC and carried out modelling effect assessment.
Predict that user chooses
Chosen for different scenes with the user of training set same type to form forecast set, such as drawing new scene choosing
Take never had had the user of buying behavior under the category since history；It is chosen at for waking up user's scene of being sunk into sleep in the category
User of single time before half a year under last time.
Model prediction
It is first according to preset path and reads the model trained, then forecast set is equally converted into libsvm lattice
Formula, the predict functions for recalling gbdt are predicted.
Electronic ticket susceptibility exports
Each user will be exported after model prediction and uses certificate probability under different electronic ticket types under different categories, such as
Shown in table 4, these data storages are easy to use into hdfs files.
The electronic ticket of table 4 uses probability signal table
From the above it can be seen that the method for the structure electronic ticket susceptibility identification model that the present embodiment provides, because
User's sample data and determination behavioural characteristic are selected using according to the application scenarios of electronic ticket, and electronic ticket is classified,
Then the technological means of model training is carried out, establishes the model of user in application-specific scene for electronic ticket susceptibility, from
And susceptibility of the user to electronic ticket is more accurately judged that, and then reduction businessman cost and improve electronic ticket and use effect
The technique effect of fruit.
Fig. 2 is the signal of the main modular of the device of structure electronic ticket susceptibility identification model according to embodiments of the present invention
Figure.
A kind of as shown in Fig. 2 dress of the structure electronic ticket susceptibility identification model provided according to one embodiment of the invention
Put 200, including：
Sample chooses module 201, for obtaining user's sample data according to the application scenarios of electronic ticket；
Electronic ticket classifying module 202, at least one of the denomination according to electronic ticket, limit or discount rate to electronics
Certificate is sorted out to obtain electronic ticket species；
Characteristic determination module 203, behavioural characteristic is determined for the application scenarios according to electronic ticket；
Training module 204, electronic ticket susceptibility is trained according to user's sample data, electronic ticket species and behavioural characteristic
Identification model.
From the above it can be seen that the device for the structure electronic ticket susceptibility identification model that the present embodiment provides, because
User's sample data and determination behavioural characteristic are selected using according to the application scenarios of electronic ticket, and electronic ticket is classified,
Then the technological means of model training is carried out, establishes the model of user in application-specific scene for electronic ticket susceptibility, from
And susceptibility of the user to electronic ticket is more accurately judged that, and then reduction businessman cost and improve electronic ticket and use effect
The technique effect of fruit.
In some optional embodiments, the training module 204 is additionally operable to：
Training set is selected from user's sample data；
According to the training set, electronic ticket species and behavioural characteristic, using there is electronics described in the classification algorithm training of supervision
Certificate susceptibility identification model.
In some optional embodiments, described device 200 also includes：
Recruitment evaluation module 205, for selecting checking collection from user's sample data in addition to the training set,
The checking collection is predicted using the electronic ticket susceptibility identification model, judges whether prediction result reaches expected and want
Ask；If reaching, the electronic ticket susceptibility identification model is preserved；If new behavioural characteristic not up to, is selected, and according to
Electronic ticket susceptibility identification model described in the new behavioural characteristic re -training.
In some optional embodiments, the sorting algorithm for having supervision includes K- nearest neighbor algorithms, decision Tree algorithms, Piao
At least one of plain bayesian algorithm, Logistic regression algorithms, SVM algorithm or AdaBoost Meta algorithms.
In some optional embodiments, the sorting algorithm for having supervision is Gradient Iteration decision Tree algorithms.
In some optional embodiments, described device 200 also includes：
Certificate uses mark module 206, for the service condition according to user in user's sample data to electronic ticket,
The electronic ticket that user is added in user's sample data uses mark.
Fig. 3 shows the method or structure electricity for the structure electronic ticket susceptibility identification model that can apply the embodiment of the present invention
The exemplary system architecture 300 of the device of sub- certificate susceptibility identification model.
As shown in figure 3, system architecture 300 can include terminal device 301,302,303, network 304 and server 305.
Network 304 between terminal device 301,302,303 and server 305 provide communication link medium.Network 304 can be with
Including various connection types, such as wired, wireless communication link or fiber optic cables etc..
User can be interacted with using terminal equipment 301,302,303 by network 304 with server 305, to receive or send out
Send message etc..Various telecommunication customer end applications, such as the application of shopping class, net can be installed on terminal device 301,302,303
(merely illustrative) such as the application of page browsing device, searching class application, JICQ, mailbox client, social platform softwares.
Terminal device 301,302,303 can have a display screen and a various electronic equipments that supported web page browses, bag
Include but be not limited to smart mobile phone, tablet personal computer, pocket computer on knee and desktop computer etc..
Server 305 can be to provide the server of various services, such as utilize terminal device 301,302,303 to user
User profile be collected, store, and carry out model training, the calculation server of compliance test result and training.Computational service
Device can analyze etc. processing to receiving user's sample data, and (such as the susceptibility for training to obtain is known by result
Other model and the prediction result being calculated according to the model) feed back to terminal device.
It should be noted that the embodiment of the present invention provided structure electronic ticket susceptibility identification model method typically by
Server 305 is performed, and correspondingly, the device of structure electronic ticket susceptibility identification model is generally positioned in server 305.
It should be understood that the number of the terminal device, network and server in Fig. 3 is only schematical.According to realizing need
Will, can have any number of terminal device, network and server.
According to an embodiment of the invention, present invention also offers a kind of electronic equipment and a kind of readable storage medium storing program for executing.
Fig. 4 is adapted for the structural representation for realizing the terminal device of the embodiment of the present invention or the computer system of server
Figure.
Below with reference to Fig. 4, it illustrates suitable for for realizing the computer system 400 of the terminal device of the embodiment of the present invention
Structural representation.Terminal device shown in Fig. 4 is only an example, to the function of the embodiment of the present invention and should not use model
Shroud carrys out any restrictions.
As shown in figure 4, computer system 400 includes CPU (CPU) 401, it can be read-only according to being stored in
Program in memory (ROM) 402 or be loaded into program in random access storage device (RAM) 403 from storage part 408 and
Perform various appropriate actions and processing.In RAM 403, also it is stored with system 400 and operates required various programs and data.
CPU 401, ROM 402 and RAM 403 are connected with each other by bus 404.Input/output (I/O) interface 405 is also connected to always
Line 404.
I/O interfaces 405 are connected to lower component：Importation 406 including keyboard, mouse etc.；Penetrated including such as negative electrode
The output par, c 407 of spool (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.；Storage part 408 including hard disk etc.；
And the communications portion 409 of the NIC including LAN card, modem etc..Communications portion 409 via such as because
The network of spy's net performs communication process.Driver 410 is also according to needing to be connected to I/O interfaces 405.Detachable media 411, such as
Disk, CD, magneto-optic disk, semiconductor memory etc., it is arranged on as needed on driver 410, in order to read from it
Computer program be mounted into as needed storage part 408.
Especially, according to an embodiment of the invention, the process of the schematic diagram description of key step above may be implemented as
Computer software programs.For example, embodiments of the invention include a kind of computer program product, it includes being carried on computer can
The computer program on medium is read, the computer program includes the program for being used for performing the method shown in the schematic diagram of key step
Code.In such embodiments, the computer program can be downloaded and installed by communications portion 409 from network, and/
Or it is mounted from detachable media 411.When the computer program is performed by CPU (CPU) 401, the present invention is performed
System in the above-mentioned function that limits.
It should be noted that the computer-readable medium shown in the present invention can be computer-readable signal media or meter
Calculation machine readable storage medium storing program for executing either the two any combination.Computer-readable recording medium for example can be --- but not
Be limited to --- electricity, magnetic, optical, electromagnetic, system, device or the device of infrared ray or semiconductor, or it is any more than combination.Meter
The more specifically example of calculation machine readable storage medium storing program for executing can include but is not limited to：Electrical connection with one or more wires, just
Take formula computer disk, hard disk, random access storage device (RAM), read-only storage (ROM), erasable type and may be programmed read-only storage
Device (EPROM or flash memory), optical fiber, portable compact disc read-only storage (CD-ROM), light storage device, magnetic memory device,
Or above-mentioned any appropriate combination.In the present invention, computer-readable recording medium can any include or store journey
The tangible medium of sequence, the program can be commanded the either device use or in connection of execution system, device.And at this
In invention, computer-readable signal media can include in a base band or as carrier wave a part propagation data-signal,
Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including but unlimited
In electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be that computer can
Any computer-readable medium beyond storage medium is read, the computer-readable medium, which can send, propagates or transmit, to be used for
By instruction execution system, device either device use or program in connection.Included on computer-readable medium
Program code can be transmitted with any appropriate medium, be included but is not limited to：Wirelessly, electric wire, optical cable, RF etc., or it is above-mentioned
Any appropriate combination.
Flow chart and block diagram in accompanying drawing, it is illustrated that according to the system of various embodiments of the invention, method and computer journey
Architectural framework in the cards, function and the operation of sequence product.At this point, each square frame in flow chart or block diagram can generation
The part of one module of table, program segment or code, a part for above-mentioned module, program segment or code include one or more
For realizing the executable instruction of defined logic function.It should also be noted that some as replace realization in, institute in square frame
The function of mark can also be with different from the order marked in accompanying drawing generation.For example, two square frames succeedingly represented are actual
On can perform substantially in parallel, they can also be performed in the opposite order sometimes, and this is depending on involved function.Also
It is noted that the combination of each square frame and block diagram in block diagram or flow chart or the square frame in flow chart, can use and perform rule
Fixed function or the special hardware based system of operation are realized, or can use the group of specialized hardware and computer instruction
Close to realize.
Being described in module involved in the embodiment of the present invention can be realized by way of software, can also be by hard
The mode of part is realized.Described module can also be set within a processor, for example, can be described as：A kind of processor bag
Include sample selection module, electronic ticket classifying module, characteristic determination module, training module, recruitment evaluation module and certificate and use mark
Module.Wherein, the title of these modules does not form the restriction to the module in itself under certain conditions, for example, sample is chosen
Module is also described as " being used to obtain user's sample data according to the application scenarios of electronic ticket ".
As on the other hand, present invention also offers a kind of computer-readable medium, the computer-readable medium can be
Included in equipment described in above-described embodiment；Can also be individualism, and without be incorporated the equipment in.Above-mentioned calculating
Machine computer-readable recording medium carries one or more program, when said one or multiple programs are performed by the equipment, makes
Obtaining the equipment includes：
User's sample data is obtained according to the application scenarios of electronic ticket；
Electronic ticket is sorted out according to the electronic ticket information of electronic ticket to obtain electronic ticket species；
Behavioural characteristic is determined according to the application scenarios of electronic ticket；
Electronic ticket susceptibility identification model is trained according to user's sample data, electronic ticket species and behavioural characteristic.
Technical scheme according to embodiments of the present invention, because selecting user's sample number using according to the application scenarios of electronic ticket
According to determine behavioural characteristic, and electronic ticket is classified, then carries out the technological means of model training, establish it is specific should
The user in scene of model with to(for) electronic ticket susceptibility, so as to more accurately judge that susceptibility of the user to electronic ticket, enter
And reduction businessman's cost is reached and has improved the technique effect of electronic ticket using effect.
Above-mentioned embodiment, does not form limiting the scope of the invention.Those skilled in the art should be bright
It is white, depending on design requirement and other factors, various modifications, combination, sub-portfolio and replacement can occur.It is any
Modifications, equivalent substitutions and improvements made within the spirit and principles in the present invention etc., should be included in the scope of the present invention
Within.
Claims (16)
- A kind of 1. method for building electronic ticket susceptibility identification model, it is characterised in that including：User's sample data is obtained according to the application scenarios of electronic ticket；Electronic ticket is sorted out according to the electronic ticket information of electronic ticket to obtain electronic ticket species；Behavioural characteristic is determined according to the application scenarios of electronic ticket；Electronic ticket susceptibility identification model is trained according to user's sample data, electronic ticket species and behavioural characteristic.
- 2. according to the method for claim 1, it is characterised in that according to user's sample data, electronic ticket species and row The step of being characterized training electronic ticket susceptibility identification model includes：Training set is selected from user's sample data；According to the training set, electronic ticket species and behavioural characteristic, using having, electronic ticket described in the classification algorithm training of supervision is quick Sensitivity identification model.
- 3. according to the method for claim 2, it is characterised in that methods described also includes：According to the training set, electronic ticket species and behavioural characteristic, using there is electronic ticket described in the classification algorithm training of supervision After the step of susceptibility identification model,Checking collection is selected from user's sample data in addition to the training set, is identified using the electronic ticket susceptibility Model is predicted to the checking collection, judges whether prediction result reaches expected requirement；If reaching, the electronic ticket susceptibility identification model is preserved；If not up to, select new behavioural characteristic, Yi Jigen According to electronic ticket susceptibility identification model described in the new behavioural characteristic re -training.
- 4. according to the method for claim 2, it is characterised in that the sorting algorithm for having a supervision include K- nearest neighbor algorithms, At least one in decision Tree algorithms, NB Algorithm, Logistic regression algorithms, SVM algorithm or AdaBoost Meta algorithms Kind.
- 5. according to the method for claim 4, it is characterised in that the sorting algorithm for having supervision is Gradient Iteration decision tree Algorithm.
- 6. according to the method for claim 1, it is characterised in that the electronic ticket information includes denomination, limit or discount rate At least one of.
- 7. according to the method for claim 1, it is characterised in that methods described also includes：After the step of obtaining user's sample data according to the application scenarios of electronic ticket,Service condition according to user in user's sample data to electronic ticket, user is added in user's sample data Electronic ticket using mark.
- A kind of 8. device for building electronic ticket susceptibility identification model, it is characterised in that including：Sample chooses module, for obtaining user's sample data according to the application scenarios of electronic ticket；Electronic ticket classifying module, for being sorted out according to the electronic ticket information of electronic ticket to electronic ticket to obtain electronic ticket kind Class；Characteristic determination module, behavioural characteristic is determined for the application scenarios according to electronic ticket；Training module, electronic ticket susceptibility identification mould is trained according to user's sample data, electronic ticket species and behavioural characteristic Type.
- 9. device according to claim 8, it is characterised in that the training module is additionally operable to：Training set is selected from user's sample data；According to the training set, electronic ticket species and behavioural characteristic, using having, electronic ticket described in the classification algorithm training of supervision is quick Sensitivity identification model.
- 10. device according to claim 9, it is characterised in that described device also includes：Recruitment evaluation module, for selecting checking collection from user's sample data in addition to the training set, use institute State electronic ticket susceptibility identification model to be predicted the checking collection, judge whether prediction result reaches expected requirement；If reach Arrive, then preserve the electronic ticket susceptibility identification model；If new behavioural characteristic not up to, is selected, and according to described new Behavioural characteristic re -training described in electronic ticket susceptibility identification model.
- 11. device according to claim 9, it is characterised in that the sorting algorithm for having a supervision include K- nearest neighbor algorithms, At least one in decision Tree algorithms, NB Algorithm, Logistic regression algorithms, SVM algorithm or AdaBoost Meta algorithms Kind.
- 12. device according to claim 11, it is characterised in that the sorting algorithm for having supervision is Gradient Iteration decision-making Tree algorithm.
- 13. device according to claim 8, it is characterised in that described device also includes：Certificate uses mark module, for the service condition according to user in user's sample data to electronic ticket, in the use The electronic ticket that user is added in the sample data of family uses mark.
- 14. device according to claim 8, it is characterised in that the electronic ticket information includes denomination, limit or discount rate At least one of.
- A kind of 15. electronic equipment for building electronic ticket susceptibility identification model, it is characterised in that including：One or more processors；Storage device, for storing one or more programs,When one or more of programs are by one or more of computing devices so that one or more of processors are real The now method as described in any in claim 1-7.
- 16. a kind of computer-readable medium, is stored thereon with computer program, it is characterised in that described program is held by processor The method as described in any in claim 1-7 is realized during row.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710565218.5A CN107424007A (en) | 2017-07-12 | 2017-07-12 | A kind of method and apparatus for building electronic ticket susceptibility identification model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710565218.5A CN107424007A (en) | 2017-07-12 | 2017-07-12 | A kind of method and apparatus for building electronic ticket susceptibility identification model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107424007A true CN107424007A (en) | 2017-12-01 |
Family
ID=60427861
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710565218.5A Pending CN107424007A (en) | 2017-07-12 | 2017-07-12 | A kind of method and apparatus for building electronic ticket susceptibility identification model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107424007A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108776909A (en) * | 2018-06-07 | 2018-11-09 | 上海在赢端网络科技有限公司 | A kind of electronic ticket derives the management system and method for value-added service |
| CN109299975A (en) * | 2018-09-03 | 2019-02-01 | 拉扎斯网络科技（上海）有限公司 | Object characteristic parameter determination method and device, electronic equipment and readable storage medium |
| CN109509033A (en) * | 2018-12-14 | 2019-03-22 | 重庆邮电大学 | A kind of user buying behavior big data prediction technique under consumer finance scene |
| CN109635010A (en) * | 2018-12-26 | 2019-04-16 | 深圳市梦网百科信息技术有限公司 | A kind of user characteristics and characterization factor extract, querying method and system |
| CN110348897A (en) * | 2019-06-29 | 2019-10-18 | 上海淇馥信息技术有限公司 | Financial service product marketing method, apparatus and electronic equipment |
| CN110472137A (en) * | 2019-07-05 | 2019-11-19 | 中国平安人寿保险股份有限公司 | The negative sample construction method of identification model, device and system |
| CN110766431A (en) * | 2018-07-27 | 2020-02-07 | 北京京东尚科信息技术有限公司 | Method and device for judging whether user is sensitive to coupon |
| CN111242661A (en) * | 2018-11-29 | 2020-06-05 | 北京京东尚科信息技术有限公司 | Coupon issuing method and device, computer system and medium |
| CN112734470A (en) * | 2021-01-05 | 2021-04-30 | 中国工商银行股份有限公司 | Electronic coupon pushing method and device based on customer preference |
| CN113689247A (en) * | 2021-10-27 | 2021-11-23 | 冰联(广州)网络科技有限公司 | Block chain electronic ticket marking method and system based on information flow parallel connection |
| CN114169906A (en) * | 2020-09-11 | 2022-03-11 | 腾讯科技（深圳）有限公司 | Electronic ticket pushing method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102129645A (en) * | 2011-03-08 | 2011-07-20 | 孙亦嘉 | Oriented advertisement publishing platform and method |
| CN103402177A (en) * | 2013-08-02 | 2013-11-20 | 南京市海聚信息科技有限公司 | Information pushing system of WiFi (wireless fidelity) terminal and implementation method thereof |
| CN106355454A (en) * | 2016-10-31 | 2017-01-25 | 腾讯科技（深圳）有限公司 | Information recommendation method, server and client side |
-
2017
- 2017-07-12 CN CN201710565218.5A patent/CN107424007A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102129645A (en) * | 2011-03-08 | 2011-07-20 | 孙亦嘉 | Oriented advertisement publishing platform and method |
| CN103402177A (en) * | 2013-08-02 | 2013-11-20 | 南京市海聚信息科技有限公司 | Information pushing system of WiFi (wireless fidelity) terminal and implementation method thereof |
| CN106355454A (en) * | 2016-10-31 | 2017-01-25 | 腾讯科技（深圳）有限公司 | Information recommendation method, server and client side |
Non-Patent Citations (1)
| Title |
|---|
| 立刻有: "《https://blog.csdn.net/shine19930820/article/details/53995369》", 3 January 2017 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108776909A (en) * | 2018-06-07 | 2018-11-09 | 上海在赢端网络科技有限公司 | A kind of electronic ticket derives the management system and method for value-added service |
| CN110766431A (en) * | 2018-07-27 | 2020-02-07 | 北京京东尚科信息技术有限公司 | Method and device for judging whether user is sensitive to coupon |
| CN109299975B (en) * | 2018-09-03 | 2021-06-08 | 拉扎斯网络科技（上海）有限公司 | Object characteristic parameter determination method and device, electronic equipment and readable storage medium |
| CN109299975A (en) * | 2018-09-03 | 2019-02-01 | 拉扎斯网络科技（上海）有限公司 | Object characteristic parameter determination method and device, electronic equipment and readable storage medium |
| CN111242661A (en) * | 2018-11-29 | 2020-06-05 | 北京京东尚科信息技术有限公司 | Coupon issuing method and device, computer system and medium |
| CN109509033A (en) * | 2018-12-14 | 2019-03-22 | 重庆邮电大学 | A kind of user buying behavior big data prediction technique under consumer finance scene |
| CN109635010A (en) * | 2018-12-26 | 2019-04-16 | 深圳市梦网百科信息技术有限公司 | A kind of user characteristics and characterization factor extract, querying method and system |
| CN109635010B (en) * | 2018-12-26 | 2021-10-08 | 深圳市梦网视讯有限公司 | User characteristic and characteristic factor extraction and query method and system |
| CN110348897A (en) * | 2019-06-29 | 2019-10-18 | 上海淇馥信息技术有限公司 | Financial service product marketing method, apparatus and electronic equipment |
| CN110472137A (en) * | 2019-07-05 | 2019-11-19 | 中国平安人寿保险股份有限公司 | The negative sample construction method of identification model, device and system |
| CN110472137B (en) * | 2019-07-05 | 2023-07-25 | 中国平安人寿保险股份有限公司 | Negative sample construction method, device and system of recognition model |
| CN114169906A (en) * | 2020-09-11 | 2022-03-11 | 腾讯科技（深圳）有限公司 | Electronic ticket pushing method and device |
| CN114169906B (en) * | 2020-09-11 | 2024-03-22 | 腾讯科技（深圳）有限公司 | Electronic coupon pushing method and device |
| CN112734470A (en) * | 2021-01-05 | 2021-04-30 | 中国工商银行股份有限公司 | Electronic coupon pushing method and device based on customer preference |
| CN113689247A (en) * | 2021-10-27 | 2021-11-23 | 冰联(广州)网络科技有限公司 | Block chain electronic ticket marking method and system based on information flow parallel connection |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107424007A (en) | A kind of method and apparatus for building electronic ticket susceptibility identification model | |
| CN109615020A (en) | Characteristic analysis method, device, equipment and medium based on machine learning model | |
| CN107273436A (en) | The training method and trainer of a kind of recommended models | |
| CN107784390A (en) | Recognition methods, device, electronic equipment and the storage medium of subscriber lifecycle | |
| CN107391680A (en) | Content recommendation method, device and equipment | |
| US20140108190A1 (en) | Recommending product information | |
| CN104572775B (en) | Advertisement classification method, device and server | |
| CN110163647A (en) | A kind of data processing method and device | |
| CN109685537B (en) | User behavior analysis method, device, medium and electronic equipment | |
| CN108960719A (en) | Selection method and apparatus and computer readable storage medium | |
| CN107688966A (en) | Data processing method and its system and non-volatile memory medium | |
| CN107346344A (en) | The method and apparatus of text matches | |
| US20190392295A1 (en) | Information processing device, method, and program that use deep learning | |
| CN108388563A (en) | Information output method and device | |
| CN109784352A (en) | A kind of method and apparatus for assessing disaggregated model | |
| CN110263255A (en) | Acquisition methods, system, server and the storage medium of customer attribute information | |
| CN108595448A (en) | Information-pushing method and device | |
| CN111966886A (en) | Object recommendation method, object recommendation device, electronic equipment and storage medium | |
| CN109214501A (en) | The method and apparatus of information for identification | |
| Agarwal et al. | Sentiment analysis in stock price prediction: a comparative study of algorithms | |
| US20240232232A1 (en) | Automated data set enrichment, analysis, and visualization | |
| CN116308505A (en) | Resource allocation method, device, storage medium and computer equipment | |
| CN108629355A (en) | Method and apparatus for generating workload information | |
| Kalaivani et al. | Predicting the price range of mobile phones using machine learning techniques | |
| WO2021129368A1 (en) | Method and apparatus for determining client type |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171201 |